[Senate Hearing 115-782]
[From the U.S. Government Publishing Office]


                                                        S. Hrg. 115-782

                           THE OPIOID CRISIS:
                         THE ROLE OF TECHNOLOGY
                         AND DATA IN PREVENTING
                         AND TREATING ADDICTION

=======================================================================

                                HEARING

                                 OF THE

                    COMMITTEE ON HEALTH, EDUCATION,
                          LABOR, AND PENSIONS

                          UNITED STATES SENATE

                     ONE HUNDRED FIFTEENTH CONGRESS

                             SECOND SESSION

                                   ON

  EXAMINING THE OPIOID CRISIS, FOCUSING ON THE ROLE OF TECHNOLOGY AND 
               DATA IN PREVENTING AND TREATING ADDICTION

                               __________

                           FEBRUARY 27, 2018

                               __________

 Printed for the use of the Committee on Health, Education, Labor, and 
                                Pensions
                                
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]                                


        Available via the World Wide Web: http://www.govinfo.gov
        
        
        
          COMMITTEE ON HEALTH, EDUCATION, LABOR, AND PENSIONS

                  LAMAR ALEXANDER, Tennessee, Chairman
                  
MICHAEL B. ENZI, Wyoming		PATTY MURRAY, Washington
RICHARD BURR, North Carolina		BERNARD SANDERS (I), Vermont
JOHNNY ISAKSON, Georgia			ROBERT P. CASEY, JR., Pennsylvania
RAND PAUL, Kentucky			MICHAEL F. BENNET, Colorado
SUSAN M. COLLINS, Maine			TAMMY BALDWIN, Wisconsin
BILL CASSIDY, M.D., Louisiana		CHRISTOPHER S. MURPHY, Connecticut
TODD YOUNG, Indiana			ELIZABETH WARREN, Massachusetts
ORRIN G. HATCH, Utah			TIM KAINE, Virginia
PAT ROBERTS, Kansas			MAGGIE HASSAN, New Hampshire
LISA MURKOWSKI, Alaska			TINA SMITH, Minnesota
TIM SCOTT, South Carolina		DOUG JONES, Alabama   
                                     
               David P. Cleary, Republican Staff Director
         Lindsey Ward Seidman, Republican Deputy Staff Director
                 Evan Schatz, Democratic Staff Director
             John Righter, Democratic Deputy Staff Director
                            
                            C O N T E N T S

                              ----------                              

                               STATEMENTS

                       TUESDAY, FEBRUARY 27, 2018

                                                                   Page

                           Committee Members

Alexander, Hon. Lamar, Chairman, Committee on Health, Education, 
  Labor, and Pensions, Opening statement.........................     1
Murray, Hon. Patty, Ranking Member, a U.S. Senator from the State 
  of Washington, Opening statement...............................     3

                               Witnesses

Mahon, Snezana, Pharm.D., Vice President, Clinical Product 
  Development, Express Scripts, St. Louis, MO....................     5
    Prepared statement...........................................     7
    Summary statement............................................    11
Green, Sherry L., J.D., Chief Executive Officer, Sherry L. Green 
  & Associations, LLC, Co-founder, National Alliance for Model 
  State Drug Laws, Sante Fe, NM..................................    12
    Prepared statement...........................................    13
    Summary statement............................................    14
Clark, H. Westley, M.D., J.D., M.P.H., Dean's Executive 
  Professor, Public Health Program, Santa Clara University, Santa 
  Clara, CA......................................................    15
    Prepared statement...........................................    17
    Summary statement............................................    23
Shah, Sanket J., Clinical Assistant Professor, Health 
  Informatics, University of Illinois at Chicago, Hinsdale, IL...    24
    Prepared statement...........................................    26
    Summary statement............................................    27

 
                           THE OPIOID CRISIS:
                         THE ROLE OF TECHNOLOGY
                         AND DATA IN PREVENTING
                         AND TREATING ADDICTION

                              ----------                              


                       Tuesday, February 27, 2018

                                       U.S. Senate,
       Committee on Health, Education, Labor, and Pensions,
                                                    Washington, DC.
    The Committee met, pursuant to notice, at 10:07 a.m. in 
room SD-430, Dirksen Senate Office Building, Hon. Lamar 
Alexander, presiding.
    Present: Senators Alexander [presiding], Isakson, Cassidy, 
Young, Roberts, Murray, Casey, Bennet, Murphy, Warren, Hassan, 
Kaine, and Smith.

                 OPENING STATEMENT OF SENATOR ALEXANDER

    The Chairman. Good morning. The Senate Committee on Health, 
Education, Labor, and Pensions will please come to order.
    This is the fifth in a series of hearings this Congress has 
had on the opioid crisis to date.
    Today, we are looking at the role data and technology can 
play in combating the opioid crisis. The Committee plans to 
hold a mark-up on legislation to address opioids as soon as the 
end of March.
    We are getting a number of good ideas from all directions. 
We had a meeting yesterday with Governors, and more than 30 
Senators attended to hear their thoughts. We will have a 
hearing with Governors in another week or 10 days to formally 
receive many of their thoughts.
    Like many doctors, Johns Hopkins surgeon, Martin Makary, 
was taught to prescribe opioids for patients during medical 
school and, quote, ``Gave out opioids like candy. My colleagues 
and I were unaware that about 1 in 16 patients became chronic 
users.''
    After seeing his own father recover from gallbladder 
surgery with a single Ibuprofen tablet, Dr. Makary realized the 
extent to which the medical community had been over prescribing 
opioid painkillers.
    Data and technology can help identify these intentional or 
unintentional actions so that behaviors can be improved.
    For example, Dr. Makary said, quote, ``My colleagues at 
Johns Hopkins and I have used data to identify the average 
number of opioids a doctor prescribes after a routine C-
section. The range of doctors' prescribing patterns is 
stunning. Some doctors average 3 to 10 opioid tablets after a 
C-section, while other doctors still average 30 or 60.''
    As we have examined this crisis over the last 5 months, we 
have learned that strong local communities are key to finding 
solutions, and Washington's role is usually to support those 
efforts and create an environment for success.
    When we look at what the Federal Government can do, sharing 
more data and utilizing new technologies may be the most 
helpful thing we can do. Data can paint a more complete picture 
of the opioid crisis, for example, revealing which communities 
are seeing a spike in prescriptions such as a West Virginia 
town of 3,191 people that saw 20.8 million painkillers shipped 
to its two pharmacies for over a decade.
    Helping doctors avoid prescribing opioids to someone 
recovering from addiction; we have suggested that in Jessie's 
Law, which the Senate passed last year. We can make it easier 
for doctors to be alerted to a patient's opioid abuse history, 
and recording the last time someone who overdosed had filled a 
prescription by checking health records and the prescription 
drug monitoring system.
    Tennessee has found that less than 50 percent of people who 
died from an overdose in 2017 had filled a prescription in the 
prior 60 days, suggesting that more people dying from an 
overdose are buying heroin or fentanyl illegally.
    Quality data gives everyone the ability to make informed 
decisions about how best to address the opioid crisis.
    For state and local governments, it means having 
Prescription Drug Monitoring Programs, or PDMP's, that are easy 
to use. PDMP's are data bases that nearly every state uses to 
track controlled substance prescriptions so state officials can 
see what is happening at the community level, and doctors and 
pharmacists can check a patient's history with controlled 
substances before writing or filling a prescription.
    For Tennessee, this has proved to be an invaluable tool. 
Between 2015 and 2017, the number of prescriptions written for 
opioids has decreased by 14 percent. The Tennessee Department 
of Health attributes these decreases to doctors and pharmacists 
using a PDMP more. However, in our state of 6.6 million people, 
there were still 7.6 million opioid prescriptions written in 
2016.
    For individual doctors, nurses, and patients, data can mean 
helping prevent more people from sliding down the slope of 
addiction.
    As we consider new legislation, I want to hear specific 
suggestions about how the Federal Government can help states 
and local communities take full advantage of the amazing 
potential that technology has to offer in solving the opioid 
crisis.
    Then there are prescription drug monitoring programs. 
Questions we have about those include the Federal Government 
preventing states from using the PDMP's in ways they think 
would best help their communities, as well as other questions.
    E-prescribing, while we do not want to mandate burdensome 
rules for doctors to follow, how can we encourage doctors to 
prescribe medications, especially controlled substances, 
electronically wherever possible?
    Then, predicting problems and avoiding them. In the private 
sector, healthcare companies like clearinghouses, hospitals, 
and insurance companies have a lot of data, and these 
businesses use that data to their advantage to make 
improvements in their businesses.
    I would like to hear about how the Federal Government can 
start using all the data it collects to identify and prevent 
over prescribing before it leads to addiction and overdoes.
    Finally, privacy. New technology and more data can be 
useful tools, but we have to consider everyone's privacy and 
there is a lot at stake.
    Addiction can impact every aspect of a person's life from 
their ability to find a job, and housing, and keeping custody 
of their child. So we need to ensure that whatever action we 
take, privacy is protected.
    Senator Murray.

                  OPENING STATEMENT OF SENATOR MURRAY

    Senator Murray. Well, thank you very much, Mr. Chairman.
    Last week, I was home in Washington State, and I heard 
firsthand from families and communities who are facing this 
epidemic. I heard from a young man who ended up in foster care 
because his parents struggled with drug use and mental health 
issues.
    I heard from a mother who felt helpless as her son 
experienced addiction. He sought help. He struggled with 
relapse and tragically died of an opioid overdose.
    I heard from another young parent whose children were taken 
away because she was unable to overcome her addiction.
    I heard from an elementary school principal, whose students 
should be learning to read and write, should be running around 
full of energy and excitement, but are, instead, struggling to 
focus at school because of the trauma of their parents' 
addiction at home.
    On a previous trip to a hospital in Longview, Washington, I 
heard how half--half--of the babies there were born to mothers 
addicted to opioids.
    Just as I have been listening to voices back in Washington 
State, this Committee has been listening here in Washington, DC 
and learning more about what our communities need to fight this 
national epidemic.
    I am very grateful to all of our witnesses today for adding 
your voices to this discussion.
    Technology and data offer important opportunities to 
address the opioid crisis, to prevent addition, and avoid the 
tragedy so many families are facing.
    While we can see the promise of this approach through 
efforts, such as Prescription Drug Monitoring Programs, which 
almost every state has established, there is a lot more we can 
do.
    These electronic data bases, which keep track of 
prescriptions for controlled substances, can be a great 
resource to fight opioid misuse. However, Big Data, by itself, 
does not guarantee a big impact. We need to use that data 
creatively, protecting the privacy of patients while finding 
innovative ways to protect their safety and health. Some states 
are pioneering ways to make that happen.
    For example, some states require doctors to check the data 
base so they are aware of a patient's existing prescriptions 
before they write new ones. States that require this saw a drop 
in opioid prescriptions, drug-related hospitalizations, and 
overdose deaths.
    Others have built-in systems that alert doctors of alarming 
patterns or signs that a patient may be at-risk of drug misuse. 
Others have made PDMP's easier to access and use by integrating 
them with Electronic Health Records.
    Still, while many states are looking at new ways to put 
their data to good use, there is also important data that many 
still cannot use, like the data collected by their neighboring 
states. Too many of the PDMP systems are not sufficiently 
interoperable, and that means they are not sharing data and 
working effectively with each other.
    As a result, the prescription a patient receives in one 
state may not show up in the system of another state. Two 
doctors in different states may not see they are writing 
prescriptions for the same patient. They may not have the data 
to see a pattern of substance misuse or to prevent one from 
developing.
    We need more states to move toward an interoperable PDMP 
system so that we can put the pieces together and give doctors 
a full picture of a patient's prescription history.
    It is not just states that have to understand the role that 
technology and data can play in addressing this epidemic. 
Pharmacies, prescribers, patients, all stakeholders have to 
take a look at how advances can help prevent and address opioid 
misuse, whether through more responsible prescribing based on 
risk factors, safe disposal of controlled substances, or safety 
checks by pharmacists when they dispense.
    We must also be mindful of the importance of patient 
privacy; data has to be easy to use, but hard to misuse. People 
dealing with addiction already face stigma and may fear 
speaking out or seeking help. We have to ensure that our data 
practices and protocols do not create new risks or fears that a 
patient's most private battles might be made public against 
their wishes.
    Striking that balance is not an easy task, but it is an 
important one. We have made great progress so far in striking 
difficult balances in responding to this crisis with bipartisan 
solutions.
    I am hopeful that through conversations like the one we are 
having today, we can continue that bipartisan progress as we 
look at how technology can empower partners fighting this 
crisis at every level, from Federal, state, and local 
governments, to healthcare providers, to educators, to public 
safety officials, and to families like the ones I meet in 
Washington State.
    I look forward to hearing from all of our witnesses today, 
and learning more from your expertise and experience.
    Mr. Chairman, I very much look forward to our continued 
work together, and with many Senators to find and move forward 
key legislative solutions to the challenges we face fighting 
this crisis.
    The Chairman. Thank you, Senator Murray.
    Thanks for your cooperation, and that of your healthcare 
staff and others, for working on the bipartisan legislation we 
hope to mark-up soon.
    I am pleased to welcome our four witnesses. I thank each of 
you for being here.
    The first witness we will hear from is Snezana Mahon. Ms. 
Mahon serves as the Vice President of Clinical Product 
Development at Express Scripts.
    In this role, she oversees the company's clinical 
initiatives and utilization management programs that aim to 
make the use of prescription medicine safer, more affordable, 
and more accessible for both patients and payers.
    Prior to this role, she served as the Senior Director of 
Medicare Strategy at Express Scripts, providing guidance to 
Medicare Advantage and Part D plans.
    Next, is Ms. Sherry Green. She has 24 years of experience 
in developing opioid and other drug abuse-related policies. She 
draws on this expertise to consult state legislators and 
healthcare professionals on strategies to prevent opioid abuse.
    She is the co-founder of the National Alliance for Model 
State Drug Laws where she served as the Chief Executive Officer 
for 20 years.
    Next, is Dr. Westley Clark. He serves as the Dean's 
Executive Professor of Public Health at Santa Clara University 
in California.
    Previously, Dr. Clark served as Director of the Center for 
Substance Abuse Treatment at the Substance Abuse and Mental 
Health Services Administration, which we call SAMHSA. He led 
the agency's effort to provide individuals with addictive 
disorders effective and accessible treatments.
    He has also been Chief of the Associated Substance Abuse 
programs at the U.S. Department of Veterans Affairs Medical 
Center in San Francisco and Associate Clinical Professional in 
the Department of Psychiatry at the University of California at 
San Francisco.
    Last, we will hear from Mr. Sanket Shah. Mr. Shah is a 
Clinical Assistant Professor of Health Informatics in the 
Department of Biomedical and Health Information Sciences at the 
University of Illinois at Chicago.
    Mr. Shah has developed course curricula focused on 
healthcare business intelligence, healthcare data, knowledge 
management, and consumer informatics. He is also Director at 
Blue Health Intelligence, an independent licensee of the Blue 
Cross Blue Shield Association and home to one of the Nation's 
largest commercial healthcare data bases of medical and 
pharmacy claims.
    Welcome, again, to our witnesses.
    Ms. Mahon, we will hear from you to begin.

STATEMENT OF SNEZANA MAHON, PHARM.D., VICE PRESIDENT, CLINICAL 
   PRODUCT DEVELOPMENT, EXPRESS SCRIPTS, ST. LOUIS, MISSOURI

    Dr. Mahon. Good morning, Chairman Alexander, Ranking Member 
Murray, and Members of the Committee.
    I am Snezana Mahon. I am the Vice President of Clinical 
Product Development at Express Scripts.
    It is an honor to come before the Committee today to 
discuss the solutions that can address the opioid epidemic that 
is not only devastating our healthcare system, but also 
splintering American families. I applaud the Committee's 
attention to this crisis.
    I am a registered pharmacist and received my Doctorate of 
Pharmacy from the St. Louis College of Pharmacy. Before joining 
Express Scripts, I practiced in the retail pharmacy setting for 
7 years. Now, I lead a clinical product organization that is 
responsible for focusing on systematic programs and changes 
that can address this crisis on a broader scale.
    Express Scripts is the Nation's largest standalone Pharmacy 
Benefit Manager. We manage the pharmacy benefit for more than 
80 million Americans. I want to leave the Committee today three 
figures and three ideas. I will tackle the figures first.
    Sixty percent; that is the amount our program has reduced 
the average day supply of first-time opioid users.
    Eighty-seven percent of new prescriptions initially written 
for a long-acting opioid were subsequently filled for a short-
acting opioid.
    Ninety-six percent of initially written prescriptions for a 
longer day supply were rewritten and filled for a 7-day supply.
    How did we achieve these results?
    First, pharmacies and pharmacists are frequently not aware 
of other medications patients may be taking. At the pharmacy 
counter, we leveraged several utilization management 
strategies.
    We have instituted a 7-day limit for first-time opioid 
users. We also implemented morphine equivalent dose 
interventions on new and current users in line with the CDC 
guidelines, as well as we preferred a short-acting opioid 
before a long-acting one.
    Second, our program also engages patients by educating them 
about the risks before it occurs, and sending them educational 
letters in their home after their first fill.
    If the patient continues to refill their prescription, we 
have a specialized pharmacist reach out to the patient and talk 
to them about the risks of overuse and abuse, and we also give 
them instructions on proper safe disposal.
    We directly provide patients with a drug deactivation 
disposal bag that enables them to safely dispose of their 
unused medications.
    Third, for prescribers, we focused on making information 
more readily available to them. Similar to pharmacies, the 
average prescriber is not always aware of prescriptions their 
patients are taking from other prescribers.
    We alert the provider via the Electronic Health Record, or 
fax, or letter on the potential of misuse and abuse, as well as 
the member's morphine equivalent dose, so the prescriber has a 
more complete picture of the patient's entire history.
    So far, our data suggests that the program is working. 
However, we continue to develop new strategies and implement 
new best practices. As a result of our program, we are now 
launching additional point of sale interventions regarding 
fentanyl.
    We also are recommending the addition of Naloxone for 
patients via Electronic Health Record, as well as a fax or a 
letter to the provider for certain individuals. And finally, we 
are educating providers on their prescribing pattern by 
comparing them to their peers, as well as communicating with 
them to encourage more conservative opioid prescribing.
    These private sector efforts could be accelerated by 
policies that support safe opioid utilization. The first is 
electronic prescribing. Currently, increasing numbers of states 
now require its use for these medications.
    Encouraging e-prescribing controlled substances would 
restrict pharmacy shopping, enable better prescription 
tracking, as well as reducing fraud, waste, and abuse.
    Express Scripts supports H.R. 3528, the Every Prescription 
Conveyed Securely Act, which would move Medicare to a system of 
e-prescribing for opioids.
    We also support requiring a 7-day fill for initial opioid 
prescriptions with exceptions for hospice, cancer, and 
palliative care.
    The S. 892 Opioid Addiction Prevention Act introduced by 
Senators Gillibrand and McCain would be a positive step forward 
in preventing addiction before it begins.
    The final policy recommendation is one we would continue to 
recommend the Committee avoid is mandating coverage of so-
called abuse deterrent opioids.
    Manufacturers have been developing and selling these novel 
approaches with a goal of making their products less 
susceptible to abuse. Unfortunately, abuse deterrent is not the 
same as abuse proof.
    Mandating the coverage of abuse deterrent opioids is a 
flawed approach and the FDA acknowledges that these products 
are not abuse proof. We do not want prescribers or patients to 
believe that these products are less addictive and cause over-
utilization patterns to continue.
    Again, thank you for the incredible opportunity to present 
Express Scripts' data-driven solutions today as we continue to 
lead the industry in developing strategies to prevent 
addiction.
    [The prepared of Dr. Mahon follows:]
                  prepared statement of snezana mahon
    Good morning Chairman Alexander, Ranking Member Murray, and Members 
of the Committee. My name is Snezana Mahon and I am the Vice President 
of Clinical Product Development for Express Scripts. It is an honor to 
come before the Committee today to discuss solutions that can address 
an epidemic that is not only devastating our health care system but 
also splintering American families: opioid addiction. I applaud the 
Committee's attention to this crisis, having already held two hearings 
this year following two other hearings last fall, and I am honored that 
you asked me before the Committee today to share what Express Scripts 
is doing to address opioid addiction--namely, developing and offering 
new tools aimed at preventing addiction from starting even before a 
patient picks up their first opioid prescription at the pharmacy 
counter.
                         About Express Scripts
    Headquartered in St. Louis, Express Scripts is the Nation's largest 
stand-alone pharmacy benefit manager (PBM). We manage the pharmacy 
benefits for more than 80 million Americans, including those in health 
plans, union-sponsored plans, state employee health plans, and public 
purchasers, including TRICARE, Medicare Part D, and Medicaid. Our 
services include providing network-pharmacy claims processing, home 
delivery pharmacy care, specialty pharmacy care, benefit-design 
consultation, drug utilization review, formulary management, and 
medical and drug data analysis services.
    Because Express Scripts interacts with patients, pharmacies, 
prescribers, and payers, our company is uniquely situated to collect 
data when patients receive and fill a prescription for an opioid under 
their pharmacy benefit. We can leverage that data across the care 
continuum in order to design interventions aimed at preventing opioid 
addiction from beginning in the first place. With 2 million Americans 
addicted to prescription narcotics, and more than 1,000 people treated 
daily in emergency departments for misusing prescription opioids, this 
is a $53 billion public health crisis.
                 Our Advanced Opioid Management Program
    To test out how we could help minimize early opioid exposure and 
prevent progression to overuse and abuse, we started with a pilot in 
2016. In a study of just more than 100,000 Express Scripts members new 
to opioid therapy, we observed a 38 percent reduction in 
hospitalizations and 40 percent reduction in emergency room (ER) visits 
in the intervention group versus control group during 6 months of 
follow-up. Half of the members received an educational letter from the 
Express Scripts Neuroscience Therapeutic Resource Center (TRC) and half 
no intervention at all. A subset of those receiving the TRC educational 
letter who had high-risk patterns of opioid use also received a 
counseling call from a Neuroscience TRC specialized pharmacist. Among 
this subset, we observed a 19 percent decrease in the days' supply of 
opioid dispensing during 6 months of follow-up. Most importantly, 
patients got the medicine they needed while we helped prevent 
unnecessary refills that could put patients at risk of harm.
    With such success, we expanded the program as an offering to our 
clients more broadly. This past September, Express Scripts launched our 
comprehensive Advanced Opioid Management (AOM) solution focused on 
opioid abuse education and prevention. This product was developed by 
leveraging our substantial healthcare data analytics capabilities and 
works across the full prescription drug continuum: from providing new 
tools for physicians at the point of care, patient education and 
outreach--including safe disposal of unused opioids--to safety checks 
for dispensing pharmacies.

    More specifically:

    Engaging Prescribers----

    A prescription from a physician or other prescriber remains the 
only lawful means for a patient to receive an opioid from a pharmacy. 
These clinicians are not always aware of prescriptions from other 
prescribers that their patients are taking. Nor are they necessarily 
aware of CDC recommendations to start short acting opioid therapy 
before advancing to longer acting forms.

     The AOM solution delivers automated messages at the 
provider point of care via Electronic Health Record (EHR) on potential 
misuse and abuse, along with morphine equivalent dose (MED) 
communications to ensure prescribers have a more complete picture of 
their patient's history;

     Enhanced Prior Authorization is applied to long-acting 
opioid prescriptions for patients without such drugs existing in their 
claim history to help encourage use of such a medication only where 
clinically appropriate; and

     When data suggests potential ``doctor shopping'' behavior, 
limiting patients to a single provider for obtaining these medications.

    Patient Education and Outreach----

    Using our data analytics capabilities as a PBM, we have found that 
one of the keys to address prescription drug abuse is patient outreach 
and education, and believe this approach could be applied across both 
the public and private payer-based healthcare insurance marketplace. 
The AOM solution engages patients by communication, specifically:

     Proactive Member Education: An important step in 
preventing opioid overuse is educating members about such risks before 
they occur. Through our AOM solution, ESI provides proactive education 
to members new to opioid therapy through an educational letter;

     Proactive Member Education through Specialized Pharmacist 
Outreach: If the member continues opioid therapy, specific utilization 
trends will trigger an Express Scripts specialized pharmacist from our 
Neuroscience Therapeutic Resource Center (TRC) to contact that member 
and provide a live clinical consultation educating the member on 
potential risks, and instructions on safe use--including proper storage 
and disposal of unused pills; and

     Providing Drug Disposal Bags: The AOM solution also 
directly provides patients with drug deactivation disposal bags that 
chemically neutralize opioids that enables them to safely dispose 
unused medications and thereby prevent future opioid diversion or 
misuse.

    Engaging Pharmacies----

    Similar to prescribers, pharmacies and pharmacists are frequently 
not aware of other medications a patient is taking. AOM endeavors to 
close these gaps in care by aggregating a patient's entire opioid 
utilization profile and limit initial opioid prescriptions.

     The AOM solution involves an intervention at the pharmacy 
point of sale (POS) for members accumulating greater than 200mg 
Morphine Equivalent Dose (MED)--a widely accepted clinical threshold at 
which greater quantities of Morphine Milligram Equivalents (MME) may be 
considered dangerous and potentially an indicator for misuse/abuse. 
Pharmacists are alerted at doses of 90 mg MME. A prior authorization is 
required for members accumulating quantities of opioid medication 
exceeding 200mg MME per day;

     Concurrent drug utilization review programs are run to 
help pharmacists identify the most pertinent clinical patient safety 
and utilization concerns; and

     First-time users prescribed short-acting opioids are 
restricted under the solution to an initial 7-day supply.

    Most importantly, we know the AOM solution works, based on data 
collected from both our initial pilot test on 100,000 members conducted 
in 2016, and the first 2 months of full operation for 4.6 million 
patients currently benefiting from this program. Key results include:

     Since becoming fully operational for nearly five million 
patients beginning on September 1, 2017 we have seen:

     59.5 percent reduction in the average days' supply per 
claim for first time opioid users

     95.9 percent of the prescriptions that were reprocessed 
because of our utilization management edits were filled for a 7-days' 
supply or less;

     Only 4.1 percent of opioid prescriptions providing more 
than a 7-day supply were approved for patients after a prior 
authorization (PA) requirement was triggered; and

     87 percent of new opioid prescriptions initially written 
for a long-acting opioid were subsequently filled with a short-acting 
opioid first due to implementation of the new enhanced prior 
authorization program.

    Continuing to Develop and Implement Best Practices

    In less than 6 months, our program has grown to nearly 7 million 
Americans enrolled. As a data driven firm, we're constantly evaluating 
marketplace behavior and trends and recommending changes to our program 
as a result. We recently announced some changes to our opioid program:

     New point-of-sale alerts: Fentanyl is being targeted 
specifically, as it is an incredibly potent drug, and fentanyl-related 
deaths are on the rise. New requirements are being added to the 
coverage approval criteria to tighten the criteria for fentanyl 
products.

     Additionally, a new drug quantity management (DQM) program 
for fentanyl patches has been created for a complete and comprehensive 
DQM solution for opioids.

     New physician care alert: We're recommending the addition 
of naloxone for potentially high-risk members who are receiving a large 
number of opioid prescriptions where treatment does not appear to be 
coordinated.

     Physician education/peer comparison: Prescriber 
educational messaging that leverages behavioral science and social 
norming based on area of practice and peer comparison to encourage more 
conservative opioid prescription.

                   Policies Lawmakers Should Consider
    Given the success of our program, Express Scripts also advocates 
for meaningful policy change that we think could expand on the some of 
the lessons we've learned. Acknowledging that some of the following 
policy options extend beyond the scope of this Committee's 
jurisdiction, should the Senate take up another legislative package on 
opioid abuse, I wanted to highlight them today given this Committee's 
comprehensive look at the problem.

    Electronic Prescribing----

    Electronic prescribing (or ``e-prescribing'') has been shown to 
dramatically reduce medication errors and fraud; yet, until 2010 the 
Drug Enforcement Agency (DEA) barred its use for ordering controlled 
substances. Currently, increasing numbers of states now require its use 
for these medications. Mandating e-prescribing controlled substances 
would restrict pharmacy shopping, enable better prescription tracking, 
and reduce fraud and waste as well. ESI supports H.R. 3528, the Every 
Prescription Conveyed Securely (EPCS) Act, as it would move Medicare to 
a system of mandatory e-prescribing for opioids as this would go a long 
way toward saving lives and stopping addiction by eliminating the 
possibility of fraudulent paper claims. Express Scripts urges the 
Senate to examine policies that increase the use of electronic 
prescribing for controlled substances, whether it is through the 
Medicare program, the DEA, or through the commercial insurance market 
through policies in this Committee's jurisdiction.

    Mandating 7-Day Fill Limit on Initial Opioid Prescriptions----

    Another effective tool for reducing opioid abuse in the program 
would involve implementing a 7-day supply limit for first fills of 
short acting opioids, with exceptions allowed for hospice and 
palliative care patients. S. 892, the Opioid Addiction Prevention Act, 
introduced by Senators Gillibrand and McCain, would also be a positive 
step forward to preventing addiction before it begins. Though this 
legislation falls outside of the HELP Committee's jurisdiction, the 
bill would benefit commercially insured patients across the country.
    Currently, there is a patchwork of state laws around the country on 
fill limits. To illustrate this, below is a table that shows how these 
laws currently vary depending on geography:

               Table 1: Opioid Related Quantity Limit Laws
------------------------------------------------------------------------
      State              Qty           1st Fill Qty      Qty for Minors
------------------------------------------------------------------------
Alaska            ----              7 days             7 days
------------------------------------------------------------------------
Arizona           ----              5 days             ----
------------------------------------------------------------------------
Connecticut       ----              7 days             7 days
------------------------------------------------------------------------
Delaware          31 days           7 days             7 days
------------------------------------------------------------------------
Hawaii            30 days           ----               ----
------------------------------------------------------------------------
Illinois          30 days           ----               ----
------------------------------------------------------------------------
Louisiana         ----              7 days             7 days
------------------------------------------------------------------------
Maine             7 acute, 30       ----               ----
                   chronic
------------------------------------------------------------------------
Massachusetts     30 days           7 days             ----
------------------------------------------------------------------------
New Hampshire     34 days           ----               ----
------------------------------------------------------------------------
New Jersey        30 days           5 days             ----
------------------------------------------------------------------------
New York          30 days           7 days             ----
------------------------------------------------------------------------
Ohio              90 days           14 days            ----
------------------------------------------------------------------------
Pennsylvania      ----              ----               7 days
------------------------------------------------------------------------
Rhode Island      30 days           20 doses & 30 MME/ ----
                                     day
------------------------------------------------------------------------
South Carolina    30 days           ----               ----
------------------------------------------------------------------------
Tennessee         30 days           ----               ----
------------------------------------------------------------------------
Utah              30 days           7 days             ----
------------------------------------------------------------------------
Vermont           ----              7 days & 50 MME/   3 days & 24 MME/
                                     day                day
------------------------------------------------------------------------
Virginia          7 acute, 14       ----               ----
                   surgical
------------------------------------------------------------------------
West Virginia     30 days           ----               ----
------------------------------------------------------------------------

    Resisting the False Appeal of Incentivizing Use of Abuse Deterrent 
Opioid Formulations----

    Opioid manufacturers have been developing and selling novel (and 
expensive) approaches with a stated goal of making their products less 
susceptible to abuse, which typically means the product is engineered 
in some way to make it more difficult (but not impossible) to crush it 
up and make it injectable. Unfortunately--as tacitly admitted by use of 
the term ``abuse deterrent'' vs. ``abuse proof''--these efforts are 
consistently defeated and, in any event, remain equally susceptible to 
misuse as any other oral medication. Nevertheless, over the last 2 
years approximately 50 pieces of legislation requiring coverage of 
Abuse-Deterrent Formulations (ADF) of opioid products have been 
introduced in more than 30 different states. While the goal of these 
bills--to reduce opioid abuse--is laudable, mandating coverage of ADF 
opioids fails to take into account several substantial flaws with this 
approach, namely:

      The FDA fully acknowledges that these products are not 
abuse proof;

     Concerns expressed by clinical experts that ADF opioids 
will mislead prescribers and patients into thinking the products are 
less addictive, and thus overprescribing patterns will continue or, 
potentially, increase; and

     While ADF opioids make tampering more difficult, these 
products are considerably more expensive than non-ADF opioids, thereby 
shrinking available coverage dollars for other drugs offered by a 
health plan payer.

    Instead of mandating first-line coverage for ADF opioids, we 
reiterate that the best approach to reducing opioid misuse is through 
comprehensive, well-coordinated efforts among providers, public and 
private healthcare payers, and law enforcement that emphasizes patient 
education on drug safety--including counseling and addiction treatment.
    Again, thank you for the incredible opportunity to present Express 
Scripts' data-driven solutions as we continue to lead our industry in 
developing strategies to prevent addiction. I am happy to answer any 
questions you may have, and offer to continue to be a resource to this 
Committee as you consider further legislation to address this epidemic 
and save lives.
                                 ______
                                 
                  [summary statement of snezana mahon]
    Express Scripts has developed a novel and comprehensive opioid 
management program for plan sponsors of prescription drug benefits. The 
program touches the entire patient care continuum and has promising 
early outcomes.

    Express Scripts Advanced Opioid Management Program

          The AOM solution delivers automated messages at the 
        prescriber point of care via Electronic Health Record (EHR) on 
        potential misuse and abuse, along with morphine equivalent dose 
        (MED) communications to ensure prescribers have a complete 
        picture of their patient's history.

          If the patient continues opioid therapy, specific 
        utilization trends will trigger an Express Scripts specialized 
        pharmacist from our Neuroscience Therapeutic Resource Center 
        (TRC) to contact that member and provide a live clinical 
        consultation educating the member on potential risks, and 
        instructions on safe use--including proper storage and disposal 
        of unused pills.

          Intervening at the pharmacy point of sale (POS) for 
        members accumulating greater than 200mg Morphine Equivalent 
        Dose (MED).

    Outcomes Data

          59.5 percent reduction in the average days' supply 
        per claim for first time opioid users.

          95.9 percent of the prescriptions that were 
        reprocessed because of our utilization management edits were 
        filled for a 7-days' supply or less.

          Only 4.1 percent of opioid prescriptions providing 
        more than a 7-day supply were approved for patients after a 
        prior authorization (PA) requirement was triggered.

          87 percent of new opioid prescriptions initially 
        written for a long-acting opioid were subsequently filled with 
        a short-acting opioid first due to implementation of the new 
        enhanced prior authorization program.

    Policies Lawmakers Should Consider

          Electronic Prescribing--Mandating e-prescribing 
        controlled substances would restrict pharmacy shopping, enable 
        better prescription tracking, and reduce fraud and waste as 
        well.

          7-Day Fill Limit on Initial Opioid Prescriptions--S. 
        892, the Opioid Addiction Prevention Act, introduced by 
        Senators Gillibrand and McCain, would also be a positive step 
        forward to preventing addiction before it begins.

          Resisting the False Appeal of Incentivizing Use of 
        Abuse Deterrent Opioid Formulations--We remain concerned that 
        ADF opioids will mislead prescribers and patients into thinking 
        the products are less addictive, and thus overprescribing 
        patterns will continue or, potentially, increase.
                                 ______
                                 
    The Chairman. Thank you, Dr. Mahon.
    Ms. Green, welcome.

 STATEMENT OF SHERRY L. GREEN, J.D., CHIEF EXECUTIVE OFFICER, 
 SHERRY L. GREEN AND ASSOCIATES; CO-FOUNDER, NATIONAL ALLIANCE 
        FOR MODEL STATE DRUG LAWS, SANTA FE, NEW MEXICO

    Ms. Green. Thank you, Chairman Alexander, Ranking Member 
Murray, and Members of the Committee.
    Thank you for this opportunity to make recommendations 
about the proper role of policy and technology in data 
projects. I am going to recommend three quality control 
principles that should be involved in all data and technology 
projects.
    The first is that the best practices for healthcare that we 
want the technology projects to actually achieve need to remain 
very clear and very consistent. When those standards appear to 
be uncertain or sometimes in conflict, the result is that the 
technology projects lose their focus. And what we see is that 
the measures of success for some vendors simply becomes staying 
ahead of their competitors.
    This is less about what technology can do and more about 
what technology should do to achieve the best practices that we 
believe are necessary to help address the opioid epidemic.
    Second, the legislative and policy changes that we actually 
need to be in place to optimize technology projects need to 
keep pace with the adoption of those technology projects.
    When those legislative changes, as often happens, lag 
behind they actually hinder, not support, the technological 
enhancements. And that causes the enhancements to be completely 
unable to fully operationalize the best practices we want them 
to achieve.
    The third principle is that the federally funded projects 
from now on should actually breakdown data silos, not 
incentivize the creation of new silos. We need to continue to 
heed the lessons that we learned from the Federal efforts to 
computerize medical records.
    While we obviously gained many benefits from that effort, 
we also now understand that the significant Federal funding 
that was put out there actually incentivized data vendors to 
make data sharing difficult and more costly. So much so, the 
Congress had to step back in and actually prohibit information 
blocking.
    We should proactively avoid this kind of situation again 
with any new funding for data projects by proactively including 
in technology grants and contracts those safeguards that will 
actually facilitate data sharing, as well as prevent the 
exclusionary data access that sometimes happens when people are 
simply trying to increase their market dominance.
    I want to thank you again for this opportunity to make 
these recommendations, and at the end of the panel, I am happy 
to answer any questions you might have.
    [The prepared statement of Ms. Green follows:]
                 prepared statement of sherry l. green
    Chairman Alexander, Ranking Member Murray, and Members of the 
Committee, I thank you for this opportunity to share recommendations 
regarding the proper role of policy in data and technology projects 
intended to help address the opioid crisis. My recommendations 
represent quality control principles necessary to ensure that the 
activities of such projects remain aligned with the goals of 
congressional opioid abuse prevention strategies, including the 
Protecting Our Infants Act, the Comprehensive Addiction and Recovery 
Act, and the 21st Century Cures Act. Based on 24 years of drafting and 
helping implement drug and alcohol policies, including those for opioid 
abuse, I respectfully offer the following points of consideration.
    Health care standards of quality and best practices which 
technology and data projects are to facilitate must remain clear and 
consistent. Technology and data solutions can significantly advance 
improved responses by health care professionals and public health 
officials to the opioid epidemic. The unrelenting misuse, abuse, 
addiction to and diversion of opioids and other potentially addictive 
substances place new demands on prescribers and dispensers. Training 
and beliefs of years past must be set aside. Professionals and 
officials must learn and use new approaches to manage pain, 
particularly chronic noncancer pain, and treat drug and alcohol 
addiction. More than ever before treatment decisions for each patient 
must represent a careful weighing of multiple factors to balance 
appropriate patient care with prevention of misuse, abuse, addiction to 
and diversion of medication. This transition in practice must be 
expeditious rather than gradual. Technology and data solutions can 
effect a more timely transition through (1) efficient delivery of new 
education and training, and (2) improved coordination and analysis of 
data relied upon for clinical treatment and public health decisions.
    As the search for tools to address the opioid epidemic ramps up, so 
too do the competing claims that various technology and data solutions 
can do more, and do more faster. But the true value of a solution can 
only be realized in its use to achieve or improve upon new standards 
and best practices for clinical care and public health. Where the 
standards are uncertain or seemingly in conflict, the focus for a 
technology and data vendor can become doing more, and doing more faster 
than its competitors.
    The use of technology and data solutions to enhance prescription 
drug monitoring programs (PDMPs) is informative. Over the past 18 
months, numerous well-intended technology and data vendors promoted 
their solutions to PDMP Administrators (Administrators). The vendors 
described in detail how their solutions can improve the Administrators' 
ability to ``catch'' doctor shoppers and detect fraud. Detecting and 
preventing fraud is certainly one of the goals of PDMPs. However, 
states are transforming their programs into better health care 
information delivery tools. The vendors were silent regarding how their 
solutions can help accomplish PDMPs' health care goals. Policymakers, 
professionals, and officials must articulate consistently and 
repeatedly the standards which technology and data solutions are to 
facilitate.
    Only by doing so will technology and data solutions remain 
effective as means to a new health care and public health practice and 
approach to addressing the opioid epidemic.
    Legislative and regulatory changes necessary to optimize technology 
and data solutions must keep pace with the adoption of the solutions. 
Processes for refining and updating technology and data often proceed 
at a faster pace than amendments to statutes or even regulations. 
Technology and data solutions do not operate in a vacuum; they must 
comply with applicable policies that govern access, use, and disclosure 
of various types of data. When those policies fail to support the 
standards of quality and best practices for use of data that 
implementation of a solution is designed to achieve, the solution is 
unable to fully operationalize the standards and best practices.
    A primary objective of Federal and state PDMP enhancement 
initiatives is integrating PDMP data into health and pharmacy 
information technology (IT). Millions of public dollars are being spent 
on integration technology. This integration removes barriers to easy 
access of PDMP data and allows health care professionals to efficiently 
rely upon the data to inform patient care decisions. Access, use, and 
disclosure rules for PDMP data may differ from those for medication 
history traditionally maintained by health and pharmacy IT. The 
variances may be in one or more of the following categories: (1) 
authorized users of data, (2) methods of accessing data, (3) allowable 
purposes for accessing data, (4) storage and retention of data, (5) 
presentation of data to authorized users, (6) disclosure and use of 
data in health and pharmacy IT, and (7) tracking of requests for data. 
Failure to reconcile these governance rules prior to PDMP data 
integration can impede effective use of PDMP data in the clinical 
workflow. Simultaneously, health and pharmacy IT systems are at greater 
risk of violating idiosyncratic PDMP data usage provisions. 
Policymakers and regulators must proactively modify laws and rules to 
timely support rather than hinder technology and data enhancements 
needed to improve prescribing and dispensing of potentially abused 
substances.
    New or expanded technology and data solutions to address the opioid 
epidemic must strive to break down data silos, not incentivize the 
creation of new silos. Prior Federal efforts strove to encourage an 
interconnected web of health care providers and consolidation of 
patient information. Significant Federal dollars intended to bring 
about the web and consolidation inadvertently incentivized the practice 
of data siloing. Health IT vendors were reluctant to share information 
for fear of losing customers to their competitors. Based on this fear, 
the vendors made the existence of data sharing costly and inconvenient. 
Congress responded by prohibiting and penalizing information blocking. 
The National Academy of Sciences (NAS) reviewed electronic systems 
developed from initiatives to computerize medical records (EMRs). NAS 
found that EMRs ``offer potential improvements to health care 
delivery'' through collection of and quicker access to key patient 
data. \1\ Clinical notes, urine drug tests results, and signed opioid 
treatment agreements may now be included in EMRs.
---------------------------------------------------------------------------
    \1\  National Academy of Sciences, Pain Management and the Opioid 
Epidemic: Balancing Societal and Individual Benefits and Risks of 
Prescription Opioid Use, p.306 (2017).
---------------------------------------------------------------------------
    However, EMRs still have data gaps. Often missing is information 
important to understanding a patient's comprehensive, and sometimes 
complex, relationships with potentially addictive substances. These 
gaps contribute to ongoing pressure for health care professionals and 
officials to use PDMPs, tools originally designed to assist 
investigations of violations of controlled substances laws. The data in 
PDMPs already exist throughout health care systems, but the data are 
maintained in piecemeal fashion. A PDMP has value for health care 
professionals because it provides in a single location a more complete 
picture of a patient's prescription history than can often be found in 
any other single source. The consolidation of patient data has yet to 
be fully realized in the health care sector. As a result, state and 
Federal agencies are spending millions of public dollars to transform 
PDMPs into optimal health care information delivery tools.
    Policymakers must heed the lessons learned from the EMR development 
process. Federally funded technology and data projects to address the 
opioid epidemic must incorporate requirements to effect proper data 
sharing and prevent exclusionary data access primarily used to gain a 
competitive advantage and increase market dominance. Examples of such 
requirements can be found in the Prescription Drug Monitoring Act of 
2017 as introduced, S. 778 (Act). Funding a single hub for sharing PDMP 
data, the Act retains states' ownership rights to determine disclosure 
parameters, and ensures cost efficient data access for patient care and 
public health surveillance activities.
    With the urgent need to save lives and stop other devastating 
consequences of opioid abuse, hundreds of millions of taxpayer dollars 
are and will be expended to expeditiously respond to the need. As 
technology and data projects race forward to make quick progress, the 
projects risk losing focus unless proper guidance is in place. I urge 
Committee Members to take a lead in adopting appropriate quality 
control measures and safeguards to ensure that the projects remain 
aligned with congressional goals for effectively tackling opioid abuse.
                                 ______
                                 
                 [Summary Statement of Sherry L. Green]
    Chairman Alexander, Ranking Member Murray, and Members of the 
Committee, I thank you for this opportunity to share recommendations 
regarding the proper role of policy in data and technology projects 
intended to help address the opioid crisis. My recommendations 
represent quality control principles necessary to ensure that such 
projects remain aligned with the goals of congressional opioid abuse 
prevention strategies, including the Protecting Our Infants Act, the 
Comprehensive Addiction and Recovery Act, and the 21st Century Cures 
Act.

    Health care standards of quality and best practices which 
technology and data projects are to facilitate must remain clear and 
consistent. As the search for tools to address the opioid epidemic 
ramps up, so too do the competing claims of various technology and data 
vendors. Where the standards are uncertain or seemingly in conflict, 
the focus for a technology and data vendor can become doing more, and 
doing more faster than its competitors. Only by consistently repeating 
the standards which technology and data solutions are to facilitate 
will the solutions remain an effective means to a new health care 
practice and approach to addressing the opioid epidemic.

    Legislative and regulatory changes necessary to optimize technology 
and data solutions must keep pace with the adoption of the solutions. 
Technology and data solutions do not operate in a vacuum; they must 
comply with access, use, and disclosure policies for various types of 
data. When those policies fail to support the standards of quality and 
best practices for use of data that implementation of a solution is 
designed to achieve, the solution is unable to fully operationalize the 
standards and best practices. Policymakers and regulators must 
proactively modify laws and rules to timely support rather than hinder 
technology and data enhancements needed to improve patient care and 
public health.

    New or expanded technology and data solutions to address the opioid 
epidemic must strive to break down data silos, not incentivize the 
creation of new silos. Policymakers must heed the lessons learned from 
Federal efforts to computerize medical records. Federal funds for the 
initiative inadvertently incentivized health IT vendors to create data 
silos and Congress had to prohibit information blocking. Consolidation 
of patient data still has yet to be fully realized in the health care 
sector. As a result, health care professionals are required or 
encouraged to use PDMPs, tools originally designed to assist law 
enforcement. A PDMP has value because it provides in a single location 
a more complete prescription history for a patient than exists in other 
single sources. Federally funded technology and data projects must 
include requirements to effect data sharing and prevent exclusionary 
data access used to increase market dominance.

    I urge Committee Members to take a lead in adopting appropriate 
quality control measures and safeguards to ensure that technology and 
data projects remain aligned with congressional goals for effectively 
tackling opioid abuse.
                                 ______
                                 
    The Chairman. Thank you, Ms. Green.
    Dr. Clark, welcome.

   STATEMENT OF H. WESTLEY CLARK, M.D., J.D., M.P.H., DEAN'S 
    EXECUTIVE PROFESSOR, PUBLIC HEALTH PROGRAM, SANTA CLARA 
              UNIVERSITY, SANTA CLARA, CALIFORNIA

    Dr. Clark. Thank you, Mr. Chairman, and Ranking Member, 
Senator Murray, other Members of the HELP Committee.
    Thank you for inviting me to participate in this important 
discussion about the opioid crisis. I have submitted my full 
testimony for the record and will not attempt to repeat it 
here.
    It is clear that the roles of technology and data are 
important in preventing and treating addiction. However, those 
roles need to be approached with care and sensitivity. Data can 
be abused; technology can be misused.
    Under the guise of addressing the opioid crisis, there is 
an assault against individuals who present for substance use 
disorder treatment. There are those who would strip away the 
privacy protections offered by 42 U.S. Code Sec. 290dd-2 and 42 
CFR Part 2. In so doing, they would be decreasing the demand 
for substance use disorder treatment and increasing the demand 
for illegal substances.
    While the theme of this hearing involves opioids, Federal 
substance use disorder confidentiality regulations cover a 
broad range of psychoactive substances including alcohol, 
marijuana, cocaine, methamphetamine, and others. Although 2.1 
million people meet criteria for opioid use disorders, 18 
million people meet criteria for disorders of other substances 
of abuse.
    The core principles underlying the existing substance use 
disorder confidentiality provisions include providing patients 
the opportunity to be informed about who is requesting their 
substance use records and the opportunity to consent to whom 
their information should be disclosed.
    While the technology exists to allow patients a role in 
determining what can happen to their substance use disorder 
treatment records, in terms of to whom that information is 
disseminated, the Electronic Health Record industry and the 
integrated service provider community have shown little 
interest in exploiting that technology. Instead, they argue, 
somewhat disingenuously, that asking for patient consent to 
disclose sensitive information is too burdensome, too 
cumbersome, and unnecessary. They say, ``All you need is 
HIPAA.''
    The Federal Substance Use Confidentiality regulations were 
promulgated because it was recognized that harm could occur to 
those who present for treatment, as treatment records could be 
misused, either willfully or negligently.
    Keep in mind that 42 CFR Part 2 has been changed twice in 
the past year, but that is not enough for those who would 
ignore the potential loss of employment, loss of child custody, 
discrimination, and stigma often associated with substance use.
    Comparing the phenomena associated with substance use with 
such conditions as hypertension, diabetes, asthma, or HIV is 
disingenuous from a confidentiality point of view as those 
conditions are not illegal under the law, and are protected by 
the Americans with Disabilities Act, while active substance use 
is not protected by the ADA.
    Trust is the cornerstone of effective behavioral health 
treatment particularly for the treatment of those substances 
for which no medication is available. Unconsented disclosure of 
sensitive information derived from the therapy records of those 
in substance use disorder treatment may actually precipitate 
relapse and overdoses given the violation of trust associated 
with the unconsented release of information. Without trust, the 
data from patients will be fraught with omissions, evasions, or 
deception. Data analytics will become an illusion.
    It is well known that behavioral health treatment records 
need to contain descriptive narratives about a patient's life 
experiences including information about trauma, marital 
problems, violence, incarceration, sexual encounters, as well 
as substance issues.
    While HIPAA allows for separate psychotherapy notes, 
currently used EHR's actually discourage those separate notes. 
How is a substance use treatment provider going to explain to a 
patient inquiring about the confidentiality of their treatment 
records that HIPAA allows for providing such descriptive 
information when that information includes information going to 
non-healthcare providers, cost management, customer service, 
and business planning among other things?
    The HITECH Act failed mental health and substance use 
disorder treatment providers by providing almost no incentives 
to them while promoting technology to primary care providers 
and hospitals.
    To compensate for that failure, advocates for eviscerating 
or abandoning 42 U.S. Code Sec. 290dd-2 or 42 CFR Part 2 
proposed to abandon the very people we are trying to encourage 
into treatment. Please, do not allow this to happen.
    Incidentally, HIPAA contains a self-pay out of pocket 
provision that allows those with the financial means to 
withhold sensitive information from their health plans. The 42 
CRF Part 2, on the other hand, does not discriminate based on 
economic status, thus permitting respect for the agency and 
autonomy of the rich and poor alike.
    Existing data segmentation strategies, such as consent to 
share, would actually facilitate patient consent under the 42 
CFR Part 2, an information exclusion under HIPAA. However, 
there is little interest in this technology among those who 
should be interested.
    [The prepared statement of Dr. Clark follows:]
                 prepared statement of h. westley clark
    My name is Dr. H. Westley Clark. I am a psychiatrist and addiction 
medicine specialist. I retired from Federal service after proving 
clinical care to our Nation's veterans for 14 years and after directing 
the Center for Substance Abuse Treatment in the Substance Abuse and 
Mental Health Services Administration for 16 years.
    I am currently teaching undergraduates about substances of misuse 
to undergraduates at Santa Clara University, recognizing that the young 
men and women of this Nation are both at risk for substance misuse and 
have the potential to changing the cultural dynamic which puts their 
age cohort at greatest risk for misuse and overdose.
    I am here to advocate for maintaining the integrity of 42 USC 290-
dd and to keeping those Federal regulations that protect individuals 
with substance use disorders who would be discouraged from seeking 
substance use disorder treatment, because they would be subject to 
discrimination and legal consequences in the event that their 
information is improperly used or disclosed.
    As you well know, we are in the midst of the worse opioid epidemic 
that this Nation has ever seen. And, at the same time, less than 10 
percent of people who need treatment seek treatment. Instead of 
recognizing that we need to reassure those in need of treatment that 
they can trust the weakening 42 USC 290-dd and 42 CFR Part 2.
    It is argued that the opioid epidemic justifies modifying 42 CFR 
Part 2 to address the opioid overdose deaths and the misuse of opioids. 
While the issue of opioid misuse is of major importance, we should keep 
in mind that 42 CFR Part 2 does not just apply to opioids.
    Data from the National Survey on Drug Use and Health reveals that 
65 million Americans 12 and Older admit to binge drinking in the past 
month. Of these, 16 million admit to being heavy drinkers. We should 
also be aware that 24 million people admit to being past month users of 
marijuana. \1\
---------------------------------------------------------------------------
    \1\  Source: SAMHSA, Center for Behavioral Health Statistics and 
Quality. National Survey on Drug Use and Health, 2016.
---------------------------------------------------------------------------
    These numbers alone suggest the magnitude of the issues we are 
confronting today, as they exceed the 3.4 million people who admit to 
past month use of pain relievers and the 475,000 who admit to past 
month users of heroin.
    The critical question today is how do we get the 28.6 million 
Americans who are current illegal drug users and the 65 million people 
who are binge drinkers to discuss their substance use with the medical 
community? Sally Satel, psychiatrist, author and commentator asked:

        ``[W]hat should we do about the opioid crisis? First, we must 
        be realistic about who is getting in trouble with opioid pain 
        medications. Contrary to popular belief, it is rarely the 
        people for whom they are prescribed. Most lives do not come 
        undone, let alone end in overdose, after analgesia for a broken 
        leg or a trip to the dentist. There is a subset of patients who 
        are vulnerable to abusing their medication--those with 
        substance use histories or with mental health problems. 
        Ideally, they should inform physicians of their history, and, 
        in turn, their doctors should elicit such information from 
        them.'' \2\
---------------------------------------------------------------------------
    \2\  Satel, Sally, ``The Myth of the Roots of the Opioid Crisis'', 
Politico Magazine, February 21, 2018, https://www.politico.com/
magazine/story/2018/02/21/the-myth-of-the-roots-of-the-opioid-crisis-
217034, accessed 02/24/2018.

    Although the use of alcohol is legal for those over the age of 21, 
the medical community should also communicate with their patients about 
alcohol use. However, as for all psychoactive substances, 
communications between clinician and patient require trust. Trust is 
not possible if the function of disclosure is the release of sensitive 
information into a virtual data storm sewer.
    It is often argued that substance use should be treated like HIV, 
the flu, diabetes or hypertension and therefore should be treated like 
those conditions. Those who make this argument blind themselves to the 
reality that many substances of misuse are illegal, and that disclosure 
of such information can give rise to harm to the individual affected.
    The harms to which a person who admits to substance use may suffer 
includes the loss of employment, the loss of housing, the loss of child 
custody, the loss of benefits, stigma and discrimination, the loss of 
privacy and the loss of autonomy. \3\ Medical records can also be used 
to incriminate a person and subject that person arrest, prosecution, 
and incarceration.
---------------------------------------------------------------------------
    \3\  Lopez, Karla and Reid, Deborah, ``Discrimination Against 
Patients with Substance Use Disorder Remains Prevalent and Harmful: The 
Case for 42 CFR Part 2, ``Health Affairs Blog, April 13, 2017, DOI: 
10.1377/hblog20170413.059618, accessed 02/25/2018.
---------------------------------------------------------------------------
    It is irresponsible to ignore the real harms to which a person with 
a history of substance use could be subject. It is also irresponsible 
to ignore the implication that modern electronic health information has 
for privacy and confidentiality. It is sometimes said that computers 
have eidetic memories--they don't forget. Thus, people in recovery from 
alcohol and drug use who have long since stopped using are still at 
risk for discrimination and stigma.
    The case is often made that the health care delivery systems need 
to know about the substance use history of a patient. You don't hear 
why providers can't simply ask patients themselves about their 
substance use histories. You hear that it is too confusing for 
clinicians to know about 42 CFR Part 2 and to apply the rules. Yet, 
these same clinicians and health care systems spend quite a bit of time 
learning about and executing reimbursement rules, licensing rules, 
administrative rules, quality standard rules, and all the other rules 
that are necessary to get paid for the services delivered to the very 
people whose agency and dignity are now deemed too inconvenient to 
respect.
    No, I rarely hear or read about concern about the harm to the 
patient. Instead, I hear concern for the convenience of the delivery 
system, a concern that creates an adversarial relationship between 
patient and practitioner rather than respect for and trust from the 
patient. What appears to underlie the argument for administrative 
efficiency and systems needs is distrust of the patient, if not 
contempt for the patient.
    Now is the time to welcome people with substance use disorders into 
the health care delivery system, not with the demand that such 
individuals concede their agency, dignity and privacy to the 
administrative convenience of the health care delivery system, but with 
the old adage of ``First, do no harm.''
    Distrust and Contempt for people with substance use disorders has 
led to distortions and misinterpretation of 42 CFR Part 2. Emergency 
room clinicians argue that a patient with an opioid use disorder comes 
into the ED following an overdose and is unresponsive, 42 CFR part 2 
keeps them from getting lifesaving information. Not true, 42 CFR Part 2 
allows those emergency room clinicians to access Part 2 protected 
information kept either by a health information exchange or a substance 
use disorder treatment program in order to treat the patient in the 
emergency status.
    Internists may argue that it is critical not to prescribe an opioid 
to an opioid dependent patient who is on methadone. However, they don't 
establish that asking the patient about their methadone treatment is 
ineffective. Furthermore, they don't establish that checking the PDMP 
is ineffective. If the PDMP is ineffective, they don't argue for 
improving PDMPs by making them real time and regional.
    Family members, concerned about the welfare of their opioid 
dependent adult relative, are not precluded from getting information 
when an unconscious adult is brought into the ER following an opioid 
overdose. Emergency room clinicians under this situation are not 
prohibited from sharing information with those concerned family 
members.
    It is argued that 42 CFR Part 2 perpetuates the stigma of 
addiction. This disingenuous argument ignores the laws, regulations, 
policies and social view about addiction and substance use disorders. 
It is not illegal to be depressed. It is not illegal to have diabetes. 
It is not illegal to have a broken leg. It is illegal to use heroin. 
People with untreated or active diabetes are protected by the Americans 
with Disabilities Act. People with untreated or active substance use 
are not. There are no signs posted at the employment office of 
employers declaring that the workplace is a hypertension free workplace 
and that all new applicants will have their blood pressure checked; 
there are no signs saying that anyone with evidence of hypertension 
shall be denied employment.
    The Department of Health and Human Services has already moved to 
accommodate the modernization of 42 CFR Part 2 through two rounds of 
rulemaking, including a 2017 Final Rule and a 2018 Final Rule. However, 
the EHR community and a number of health systems remain restless, 
impatient and intolerant of those with substance use disorders, 
suggesting that information sharing is more important than the people 
about whom that information is shared. Thus, the regulatory efforts to 
allow patient to provide a general disclosure for substance use 
disorder information, to offer some flexibility in transmitting 
substance use data electronically, and to clarify the circumstances in 
which providers can disclose patient information to contractors and 
subcontractors for payment and healthcare operations is not enough. The 
critics of 42 CFR seek to expose those with substance use disorders who 
seek treatment, making the exercise of treatment a dangerous 
proposition.
                   Patient Attitudes Toward Treatment
    We spend millions of dollars collecting information about the 
substance use patterns of people in the US. Perhaps we should be 
concerned about the reality that 89 percent of people, who meet 
criteria for needing substance use disorder treatment, did not receive 
such treatment. \4\
---------------------------------------------------------------------------
    \4\  Source: SAMHSA, Center for Behavioral Health Statistics and 
Quality, National Survey on Drug Use and Health, 2015 and 2016.
---------------------------------------------------------------------------
    Of the 28.6 million people who misused illicit drugs and the 65 
million people who were binge drinkers in the past month, only 3.8 
million people received treatment in the past year. Of course, mere use 
does not equate with dependence or needing treatment. However, NSDUH 
data indicate that over 20 million people 12 or older met criteria for 
a substance use disorder in the past year in 2016, with 2.1 million 
meeting criteria for an opioid use disorder.
    What is equally interesting is that of the people who met criteria 
for needing treatment and did not receive treatment, 95.5 percent 
perceived no need for treatment. In short, 18.7 million people needed 
but did not receive treatment; of these, 17.9 million perceived no need 
for treatment.
    Now comes the critics of 42 CFR Part 2, under the flag of bringing 
integrated treatment to those in need, claiming that it is 42 CFR Part 
2 that operates as a barrier to effective and efficient treatment of 
opioid use disorders, claiming that there is no need for special 
concerns about substance use disorders, today, never mentioning how 
they will explain to those actually seeking treatment and those in need 
of treatment the ramifications of attenuating 42 CFR Part 2.
       Changing 42 CFR Part 2 and the Response of Substance Users
    It is important to recognize that 42 CFR Part 2 does not apply to 
most clinicians or most clinical settings. In fact, 42 CFR Part 2 only 
applies to programs that hold themselves out ``as providing, and 
provides, alcohol or drug abuse diagnosis treatment, referral for 
treatment or prevention.'' Of course, 42 CFR Part 2 governs substance 
use disorder patient records for those patients who receive, diagnosis, 
referral or treatment from (a) an identified unit of a general medical 
facility that holds itself out as providing, and provides alcohol or 
drug use disorder diagnosis, treatment or referral for treatment or (b) 
medical personnel or other staff in the general medical care facility 
whose primary function is to provide those services.
    It is the patient records of a substance use disorder program 
(which includes the substance use patient records clinicians who hold 
themselves out as treating people with substance use disorders in even 
in non-specialty settings), that are controlled by 42 CFR Part 2. This 
creates a responsibility for the substance use disorder program to 
explain to the patient the meaning of confidentiality as it applies to 
information disclosed to the treatment program.
    For the millions of people whose substance use does not meet 
criteria for protection under 42 CFR Part 2, HIPAA controls. HIPAA 
regulations allow for unconsented disclosure of patient information 
for, among other things, healthcare operations.

    Healthcare operations include:

          Underwriting, enrollment, premium rating, and other 
        activities related to the creation, renewal, or replacement of 
        a contract of health insurance or health benefits, and ceding, 
        securing, or placing a contract for reinsurance of risk 
        relating to claims for health care (including stop-loss 
        insurance and excess of loss insurance)

          Reviewing the competence or qualifications of health 
        care professionals, evaluating practitioner and provider 
        performance, health plan performance, conducting training 
        programs in which students, trainees, or practitioners in areas 
        of health care learn under supervision to practice or improve 
        their skills as health care providers, training of non-health 
        care professionals, accreditation, certification, licensing, or 
        credentialing activities;

          Business planning and development, such as conducting 
        cost-management and planning-related analyses related to 
        managing and operating the entity, including formulary 
        development and administration, development or improvement of 
        methods of payment or coverage policies

          Business management and general administrative 
        activities of the entity, including, but not limited to:

        (i) Management activities relating to implementation of and 
        compliance with the requirements of this subchapter;

        (ii) Customer service, including the provision of data analyses 
        for policy holders, plan sponsors, or other customers, provided 
        that protected health information is not disclosed to such 
        policy holder, plan sponsor, or customer.

        (iii) Resolution of internal grievances;

        (iv) The sale, transfer, merger, or consolidation of all or 
        part of the covered entity with another covered entity, or an 
        entity that following such activity will become a covered 
        entity and due diligence related to such activity; and

        (v) Consistent with the applicable requirements of Sec.  
        164.514, creating de-identified health information or a limited 
        data set, and fundraising for the benefit of the covered 
        entity.

    Do non-42 CFR Part 2 covered providers explain the width and depth 
of the health care operations provision under HIPAA? Would patients 
exempted from 42 CFR Part 2 protections feel that disclosing histories 
of substance use is wise under HIPAA, even if experimental or rare use 
of psychoactive substances is involved? Much of the literature favoring 
weakening 42 CFR Part 2 or aligning it much more substantively does not 
discuss this perspective.
    Moving from HIPAA into those programs whose records are controlled 
by 42 CFR Part 2, it is clear that those with moderate to severe 
substance use disorders requiring treatment already do not believe that 
treatment is warranted. How are we going to encourage them to 
participate in treatment when we propose to broadcast their personal 
information through network of uncertainty entities with uncertain 
purpose?
    Unfortunately, there are more serious consequences to voiding the 
patient's right to consent to the disclosure of sensitive information. 
The unconsented disclosure of sensitive information resulting in harm 
to the patient could easily give rise to suicide, relapse to substance 
use or overdose; these are tragic events that we should be avoiding 
rather than pretending that the agency and dignity of the patient has 
no value and can be compromised for the convenience of EHR vendors, 
data miners and health care operations. Furthermore, we should 
recognize that many in substance use disorder treatment are at risk for 
depression, anxiety and other psychiatric disorders, any of which would 
be made worse by a breach of trust by substance use disorder treatment 
programs and the health care delivery system.
                         Blaming the Vulnerable
    The Health Information Technology for Economic and Clinical Health 
Act (HITECH Act) was enacted under Title XIII of the American Recovery 
and Reinvestment Act of 2009. \5\ It provided billions of dollars of 
incentives to an array of primary care hospitals and to physicians to 
adopt electronic health records and to promote the exchange of health 
information. However, that same act essentially ignored the behavioral 
health community; as a result, there were no incentives available for 
substance use disorder treatment programs to adopt electronic health 
records. In addition, there were no incentives to the electronic health 
record industry to develop software and protocols specific to the 
behavioral health community and the sensitive information generated by 
behavioral health providers, information of little use to most primary 
care providers.
---------------------------------------------------------------------------
    \5\  The American Recovery and Reinvestment Act of 2009, (Public 
Law 111-5).
---------------------------------------------------------------------------
    At the time of the unfolding of the HITECH Act, I was the Health 
Information Technology Strategic Initiative Lead for SAMHSA. My team 
and I met with a number of software vendors in an effort to address the 
unique needs of the behavioral health community and to compensate for 
the omission of behavioral health from the promulgated incentives 
provided to general medicine. We met with little success.
    However, in order to compensate for excluding behavior health from 
the incentives, standards, and designs for the evolving EHR systems, 
information exchanges, and the growing recognition that comprehensive 
health care required addressing behavioral health, efforts were mounted 
to promote the fiction that behavioral health patient information 
contained nothing unique and distinct from the general health care 
environment.
    The notion that all health care information is equivalent runs 
counter to the historical status recognized in the psychotherapist-
patient privilege which was justified on the grounds that some personal 
health information was more sensitive than others. Discussions of 
mental health, substance use, and sexual health are inhibited unless 
the patient has certain reassurances that highly sensitive personal 
health information would remain between themselves and their health 
care providers. Indeed, ``the prevailing legal default and ethical norm 
in Western nations both strongly favor the preservation of patient 
confidence in the absence of compelling grounds to act otherwise.'' 
1A\6\
---------------------------------------------------------------------------
    \6\  Shenoy, A and Appel, JM, ``Safeguarding Confidentiality in 
Electronic Health Record'', Cambridge Quarterly of Healthcare Ethics 
(2017), 26, 337-341.
---------------------------------------------------------------------------
    As Shenoy and Appel point out, the behavioral health record ``often 
combines data related to the patient's present symptoms, with a 
descriptive narrative of the patient's life experience, including 
sensitive details of psychological trauma, domestic violence, 
incarceration, sexual encounters, and substance abuse. Much of this 
information is of great value to a therapist, but not always of 
clinical use to many other medical providers. The stigma attached to 
mental healthcare among some individuals and in certain cultural 
communities even leads some patients to avoid using their insurance for 
psychiatric care in order to protect their privacy.'' \7\
---------------------------------------------------------------------------
    \7\  Ibid.
---------------------------------------------------------------------------
    At SAMHSA, we recognized the continued sensitivity of behavioral 
health information, especially for substance use in particular. As a 
result, we developed an open source codebase through a contract that 
would provide an inexpensive software application for the behavioral 
health community. \8\ Unfortunately, due to complaints of unfair 
competition we discontinued our efforts.
---------------------------------------------------------------------------
    \8\  http://www.feisystems.com/what-we-do/learn-about-wits/why-
choose-wits-2/.
---------------------------------------------------------------------------
    The HITECH Act with its focus on meaningful use and information 
exchange did not change the unique character of behavioral health 
information. As a result, we developed Consent2Share, an open-source 
data segmentation platform that could be incorporated into existing 
electronic health records to allow patients to be able to consent to 
the disclosure of highly sensitive patient information. \9\
---------------------------------------------------------------------------
    \9\  Department of Health and Human Services: 42 CFR Part 2: 
Confidentiality of substance use disorder patient records; proposed 
rule. Federal Register 81: 6988-7024, 2016.
---------------------------------------------------------------------------
    Consent2Share was developed evolved within the Data Segmentation 
for Privacy (DS4P) initiative within ONC's Standards and 
Interoperability (S&I) Framework to improve the interoperability of the 
plethora of EHRs containing sensitive information that must be 
protected. The DS4P initiative met its two goals, which were to: 
Demonstrate how standards can be used to support current privacy 
policies, including 42 CFR Part 2, for sharing sensitive health 
information across organizational boundaries; and develop standards 
that will enable sensitive electronic health information to flow more 
freely to authorized users while improving the ability of health IT 
systems to implement current privacy protection requirements for 
certain Types of health care data, such as substance use disorder 
patient records.
    Unfortunately, the EHR vendor community felt no need to support 
data segmentation, dismissing the importance of privacy and 
confidentiality to patients. Furthermore, health information exchanges 
chose to ignore the importance of privacy and confidentiality to the 
patients by choosing not to embrace the utility of data segmentation 
and patient choice. Naturally, without data segmentation and consent 
management capacities, substance abuse treatment programs operating 
under 42 CFR Part 2 requirements have diminished capacities to share 
information with integrated treatment models that ignore patient 
choice.
    In short, SAMHSA was able to demonstrate that patient choice could 
be respected without compromising the agility and flexibility of 
required for integrated information exchange. However, for matters of 
mere convenience and low market demand, most EHR vendors and health 
information exchanges chose to support the less expensive and ethically 
problematic position of eviscerating 42 CFR Part 2.
            Economic Disparities, HIPAA, and Confidentiality
    What is remarkable about the industry and provider objections to 
having patients weigh in on whether their private medical information 
should be disclosed is the loophole in HIPAA that allows rich people or 
middle people to have the right to restrict certain disclosures of 
protected health information to a health plan where the individual pays 
out of pocket in full for the health care or service received. \10\ 
Health care providers, under HIPAA, are required to include such a 
statement in the notice of privacy practices provided to the patient. 
Thus, if a patient is rich and can pay for their own treatment in full, 
including substance use disorder treatment or if they are middle class 
and can mortgage their home to pay for their treatment in full, they 
can avoid disclosing the fact that they are in substance use disorder 
treatment to their health plan. What is amazing is that providers who 
are committed to doing no harm are willing to sacrifice poor whites, 
poor blacks, poor Hispanics, poor Native Americans, poor Alaskan 
Natives, poor Hawaiians, and poor Asians in the service of a fiction of 
needing highly sensitive personal information without a patient's 
consent when they could most likely receive that information simply by 
asking the patient. In situations where a patient refuses consent to 
disclose sensitive information to entities outside of the treatment 
situation, that should be the patient's prerogative.
---------------------------------------------------------------------------
    \10\  Department of Health and Human Services; 45 CFR Parts 160 and 
164: Modifications to the HIPAA Privacy, Security, Enforcement, and 
Breach Notification Rules Under the Health Information Technology for 
Economic and Clinical Health Act and the Genetic Information 
Nondiscrimination Act; Other Modifications to the HIPAA Rules. Federal 
Register 78 (17: 5566-5702)
---------------------------------------------------------------------------
    Given the well documented harm that can happen to a person who is 
an admitted substance user, it should not be EHR vendors or health 
systems that should decide what sensitive information should be 
disclosed outside of a substance use treatment process. Financial 
ability should not be the deciding factor on whether a person retains a 
modicum of control over their personal information.
   Increased Liability for Substance Use Disorder Treatment Programs
    Substance Use Disorder treatment programs have a duty to inform 
patients about the limits of confidentiality. Given the spectrum of 
entities under the rubric of healthcare operations, it would be 
difficult for a substance use disorder treatment program to accomplish 
this with any degree of effectiveness; this would expose the covered 
program to liability.
    Given that the potential harms from inappropriate disclosure of 
sensitive information garnered during substance use disorder treatment 
is real, the disclosure of that information may give rise to legal 
claims including lawsuits for some form of negligence. Unfortunately, 
since substance abuse treatment programs will be the entities releasing 
information under the proposed modified 42 CFR Part 2, undoubtedly they 
will bear the brunt of the legal burden. Increased liability insurance, 
legal costs, and impaired reputations will ensue. After all, once 
sensitive information is released into the entity that releases that 
information has no control over its distribution. The question would 
become should substance abuse treatment program that released the 
information have known that it contained information that could be used 
to the detriment of their current or past patient.
    Substance use disorder treatment programs caught up in lawsuits may 
have to withdraw from the treatment marketplace. Treatment programs 
that close under the weight of malpractice claims will only diminish 
the number of available treatment slots. The cost of care will also 
increase as treatment programs have to compensate for the increased 
administrative costs of doing business.
                               Conclusion
    We cannot adequately address the current opioid epidemic if we 
remove the protections that 42 CFR part 2 and its authorizing 
legislation, 42 USC Sec.  290dd-2, offers. We cannot treat those 
experiencing substance use disorders with contempt by weakening the 
protections that they currently have. We cannot treat those who 
experience substance use disorders as a means to an end, attempting to 
compensate for the lack of public investment in electronic health 
records for the behavioral health treatment communities following the 
HITECH Act's focus on primary care.
    Efforts to balance the health information technology requirements 
of integrated systems while preserving a patient experiencing a 
substance use disorder's right to consent to the disclosure of their 
substance use treatment history and sensitive matters subsumed under 
that history have been thwarted by the EHR industry and by health 
information exchanges. The claim that it would cost too much is 
overshadowed by the existence of open source strategies that could 
accomplish the necessary consent management strategies and by the 
inherent right of a person to determine what happens to sensitive 
information.
    In truth, 42 CFR Part 2 has been changed in 2017 and 2018. Now is 
the time to leave it alone, to let the health care delivery system gain 
a modicum of expertise to those changes, and to allow the information 
technology industry an opportunity to further pursue technological 
accommodations to existing information systems to permit patient 
consent to sensitive information.
                                 ______
                                 
                [Summary Statement of H. Westley Clark]
    1. 42 USC 290-dd, the authorizing legislation for substance use 
disorder (SUD) treatment should remain as is.
    2. Because two changes to 42 CFR Part 2, the implementing 
regulations of 42 USC 290-dd, have already occurred, once in 2017 and 
once in 2018, no further changes should be executed at this time.
    3. The promulgated changes to 42 CFR Part 2 promoted a closer 
alignment with HIPAA.
    4. The public policy purpose of 42 USC 290-dd and 42 CFR Part 2 is 
to encourage people in need of SUD disorder treatment to enter such 
treatment, fostering (a) an increased demand for SUD treatment and (b) 
a decrease demand for illegal substances. Although new technologies 
have arisen, including electronic health records (EHRs), and new 
service models have evolved, the original purpose of 42 USC 290-dd and 
42 CFR Part 2 has not changed. Additional efforts to align 42 CFR Part 
2 with HIPAA may eviscerate the purpose of 42 CFR Part 2 and also 
impose burdens on those who are in Recovery from SUDs.
    5. Epidemiological data demonstrate that most people in need of SUD 
treatment do not receive it and do not perceive a need for such 
treatment. Violating their confidentiality will not encourage a demand 
for treatment for those with active SUDs.
    6. While the opioid epidemic is a major public health concern, 42 
USC 290-dd and 42 CFR Part 2 also encourage those with other SUDs to 
enter treatment; this includes individuals with alcohol dependence and 
dependence on substances subject to the Controlled Substances Act such 
as, marijuana, cocaine, methamphetamine and other psychoactive 
substances.
    7. The core principles underlying the existing substance use 
disorder confidentiality provisions are providing patients the 
opportunity to be informed about whom is requesting their substance use 
records and the opportunity consent to whom their information should be 
disclosed.
    8. Despite new EHRs and integrated service models, the information 
disclosed by patients in substance use disorder treatment is uniquely 
sensitive, often involving illegal acts, psychological trauma, domestic 
violence, and potentially compromising activities the disclosure of 
which can result in substantial harm to patients.
    9. Comparing the phenomena associated with substance use with such 
conditions as hypertension, diabetes, asthma, or HIV is disingenuous as 
those conditions are not illegal under the law and are protected by the 
Americans with Disability Act, while active substance use is.
    10. Trust is the corner stone of effective SUD treatment, 
particularly for the treatment of those substances for which no 
medication is available. Unconsented disclosure of sensitive 
information derived from the records of those in substance use disorder 
treatments may precipitate relapse and overdoses, given the violation 
of trust associated with the unconsented release of such information.
    11. Acquiring patient consent to disclose sensitive information 
preserves trust and permits a balancing of policy interests associated 
with facilitating integrated care.
    12. The technology exists, although resisted by EHR vendors and 
some health care systems, to facilitate patient consent to the 
disclosure of sensitive information. The lack of interest in this 
technology may be a product of the lack of incentives within the HITECH 
Act for behavioral health treatment providers; patients interested in 
SUD treatment should not be punished by this policy omission.
    13. Further weakening of 42 USC 290-dd and 42 CFR Part 2 will 
increase the administrative costs of substance use disorder treatment 
programs due to liability issues associated with inappropriate 
disclosures.
    14. HIPAA contains a self-pay, out-of-pocket provision, that allows 
those with the means to withhold sensitive information from their 
health plans. 42 CFR Part 2, on the other hand, does not discriminate 
based on economic status, thus permitting respect for the agency and 
autonomy of the rich and the poor alike.
                                 ______
                                 
    The Chairman. Thank you, Dr. Clark.
    Mr. Shah, welcome.

  STATEMENT OF SANKET J. SHAH, CLINICAL ASSISTANT PROFESSOR, 
    HEALTH INFORMATICS, UNIVERSITY OF ILLINOIS AT CHICAGO, 
                       HINSDALE, ILLINOIS

    Mr. Shah. Good morning, Chairman Alexander, Ranking Member 
Murray, and Members of the HELP Committee.
    My name is Sanket Shah and today, I am going to provide you 
my view on how technology, specifically analytics, can help 
curb the overuse, misuse, and abuse of opioids.
    Healthcare data and analytics can play a key role in 
helping to combat this national crisis. Descriptive, 
diagnostic, predictive, and prescriptive analytics make it 
possible to identify individuals who are at-risk of becoming 
opioid dependent. Armed with this information and the right 
technologies, healthcare providers and communities can make 
better informed decisions and understand the risk of possible 
dependency.
    The first area we all should start with is descriptive 
analytics. Descriptive analytics identify what is happening and 
where.
    According to a 2017 study published by the Blue Cross Blue 
Shield Association, opioid use disorder diagnoses increased 493 
percent from 2010 through 2016. The same study identified that 
women aged 45 and older have higher rates of opioid use 
disorder than males, while males under the age of 45 have 
higher rates of opioid use disorder than females. We also know 
that females fill more opioid prescriptions across all age 
groups than males.
    Once we understand what is happening, our focus must shift 
to why it is happening. This is where diagnostic analytics come 
into play.
    For instance, we know potential determinants for opioid 
dependency include gender, age, whether the patient sought 
treatment for an acute injury or a chronic condition, and the 
size and dosage of the prescription.
    We also know that many patients engage in doctor and 
pharmacy shopping practices to obtain harmful, large quantities 
of opioids from various sources.
    According to a report published by the Inspector General of 
the United States Department of Health and Human Services, one 
such egregious case in Illinois revealed a Medicare enrollee 
received 73 prescriptions for opioid drugs from 11 prescribers 
and filled them at 20 different pharmacies.
    When you couple these factors with the lack of effective 
risk assessment and decision support tools available to 
providers, we miss the early warning signs for potential opioid 
dependency.
    Here we are. We already know the what and the why. The role 
of technology and analytics can help prevent addiction. Here is 
where we must focus on predictive analytics. Predictive 
analytics enables us to leverage data to anticipate what is to 
come.
    For example, according to a study published in the 
``British Medical Journal,'' the duration of opioid treatment 
is a far more potent predictor of abuse and overdose than just 
dosage alone. In fact, each additional week of opioid use 
increased the risk of dependence, abuse, or overdose by nearly 
20 percent. Each additional refill boosted the risk by 44 
percent with the first refill doubling that risk.
    To truly have predictive analytics, we need more data 
sources. Currently, we find ourselves in data silos across the 
public and private healthcare sectors.
    My recommendation is to open the lines of communication and 
pathways to share data for a holistic view to help combat this 
epidemic.
    The Federal Government has the means and infrastructure to 
create an integrated data environment which we can source from 
at the local and state levels. Having access to such a vast 
data repository would enable us to create robust predictive 
analytics that leverage multiple sources, including social 
determinants of health, medical, and family history, and also 
true episodes of care.
    A secure and encrypted data repository would empower our 
healthcare informaticists to administer and deploy innovative 
technologies to enhance our predictive capabilities. We can 
collaborate on advanced machine learning algorithms for deeper 
pattern analyses from both the provider and patient fronts.
    The insights gained could be tremendous. We all can 
potentially benefit from these new technologies by knowing 
which patients might respond better to non-pharmacologic, 
multimodal therapies, or targeted care management programs.
    To accomplish this, we simply need access to more substance 
abuse data. I ask you all to consider and support the 
Prescription Drug Monitoring Act of 2017 which requires any 
state that receives Federal grant funding to establish a 
Prescription Drug Monitoring Program to share their data with 
other states.
    In addition, I also ask you to consider supporting the 
Protecting Jessica Grubb's Legacy Act, which calls for 
modernizing Part 2, and allows for closer alignment with HIPAA 
regulations, and sharing of substance abuse disorder records 
for true, accurate diagnoses and effective treatment.
    Our predictive analytics have identified at-risk 
individuals for developing an addiction. We can use 
prescriptive analytics to offer up actionable insights. 
Providers and health plans can predict what may happen and make 
the necessary changes to true, proper treatment plans.
    Armed with actionable insights, new treatment models and 
alerts can be developed to deemphasize opioid medication use 
for at-risk individuals. This includes the right decision 
support tools at the point of care.
    Ultimately, technology alone will not help curb this 
epidemic. We must also use the information and insights we have 
gained to educate our providers, patients, and communities 
about proper adherence and potential risks of opioid use.
    America's prescription opioid epidemic continues to be a 
public health crisis. However, using descriptive, diagnostic, 
predictive, and prescriptive analytics, may provide an 
opportunity to identify at-risk individuals and change the 
course to help address this epidemic.
    We have the data. I ask your help to share that data.
    Thank you.
    [The prepared statement of Mr. Shah follows:]
                  prepared statement of sanket j. shah
    Good morning Chairman Alexander, Ranking Member Murray, and Members 
of the HELP Committee. My name is Sanket Shah and today I am going to 
provide you my view on how the role of technology, and more 
specifically, analytics may help curb the overuse, misuse, and abuse of 
opioids. Healthcare data and analytics can play a key role in helping 
to combat this national crisis. Descriptive, diagnostic, predictive, 
and prescriptive analytics make it possible to identify individuals who 
are at risk of becoming opioid dependent. Armed with this information 
and the right technologies, healthcare providers and communities can be 
better informed about the risk of possible dependency.
    The first area we all should start with is descriptive analytics. 
Descriptive analytics identify what's happening and where. According to 
a 2017 study published by the Blue Cross Blue Shield Association, 
opioid use disorder diagnoses increased 493 percent from 2010 through 
2016. The same study also has identified that women aged 45 and older 
have higher rates of opioid use disorder than males, while males under 
the age of 45 have higher rates of opioid use disorder than females. We 
also know females fill more opioid prescriptions than males across all 
age groups. \1\
---------------------------------------------------------------------------
    \1\ ``America's Opioid Epidemic.'' Blue Cross Blue Shield 
Association (BCBSA), 29 June 2017, www.bcbs.com/the-health-of-america/
reports/americas-opioid-epidemic-and-its-effect-on-the-nations-
commercially insured.
---------------------------------------------------------------------------
    Once we understand what is happening, our focus must shift to why 
it's happening. This is where diagnostic analytics come into play. For 
instance, we know potential determinants for opioid dependency include 
gender, age, whether the patient sought treatment for an acute injury 
or a chronic condition, and the size of the dosage and duration of the 
prescription. \2\ We also know that many patients engage in doctor and 
pharmacy shopping practices to obtain harmful quantities of opioids 
from various sources. According to a report published by the Inspector 
General of the United States Department of Health and Human Services, 
one such egregious case in Illinois revealed a Medicare enrollee 
received 73 prescriptions for opioid drugs from 11 prescribers and 
filled them at 20 different pharmacies. \3\ When you couple these 
factors with the lack of effective risk assessment and decision support 
tools available to providers, we miss the early warning signs for 
potential dependency.
---------------------------------------------------------------------------
    \2\  America's Opioid Epidemic. BCBSA.
    \3\  ``Opioids in Medicare Part D: Concerns about Extreme Use and 
Questionable Prescribing.'' HHS OIG Data Brief, July 2017, oig.hhs.gov/
oei/reports/oei-02-17-00250.pdf.
---------------------------------------------------------------------------
    Here we are, we already know the what and the why. The role of 
technology and analytics can help prevent addiction. Here is where we 
must focus on predictive analytics. Predictive analytics enables us to 
leverage data to anticipate what is to come. For example, according to 
a study published in the British Medical Journal, the duration of 
opioid treatment is a far more potent predictor of abuse and overdose 
than just dosage. In fact, each additional week of opioid use increased 
the risk of dependence, abuse, or overdose by nearly 20 percent. Each 
additional refill boosted the risk by 44 percent with the first refill 
more than doubling the risk. \4\
---------------------------------------------------------------------------
    \4\  Brat, Gabriel A, et al. ``Postsurgical Prescriptions for 
Opioid Naive Patients and Association with Overdose and Misuse: 
Retrospective Cohort Study.'' Bmj, Dec. 2017, doi:10.1136/bmj.j5790.
---------------------------------------------------------------------------
    To truly have accurate predictive analytics we need more data 
sources. Currently, we find ourselves in data silos across the public 
and private healthcare sectors. My recommendation is to open the lines 
of communication and pathways to share data for a ``holistic view'' to 
help combat this epidemic. The Federal Government has the means and 
infrastructure to create an integrated data environment which we can 
source from at local and state levels. Having access to such a vast 
data repository will enable the creation of robust predictive analytics 
that leverages multiple variables such as social determinants of 
health, family and medical history, and access to complete episodes of 
care. A secure and encrypted data repository would empower our 
healthcare informaticists to administer and deploy innovative 
technologies to enhance our predictive capabilities. We can collaborate 
on advanced machine learning algorithms for deeper pattern analyses 
from both the provider and patient fronts. The insights gained could be 
tremendous. We all can potentially benefit by knowing which patients 
might respond better to non-pharmacologic, multi-modal therapies, or 
targeted care management programs.
    To accomplish this, we simply need access to more substance abuse 
data. I ask you all to consider and support the Prescription Drug 
Monitoring Act of 2017 (S. 778) which requires any state that receives 
Federal grant funding to establish a prescription drug monitoring 
program to share their data with other states. In addition, this act 
also contains components to help fund a data sharing hub which I spoke 
of earlier. I also ask you all to consider supporting the Protecting 
Jessica Grubb's Legacy Act (S. 1850) which calls for modernizing Part 2 
to align with HIPAA regulations and will grant appropriate sharing of 
substance use disorder records to ensure persons with opioid use 
disorder and other substance use disorders receive accurate diagnoses 
and effective treatment.
    Once predictive analytics have identified at-risk individuals for 
developing an addiction, we can use prescriptive analytics to offer up 
actionable insights. Providers and health plans can predict what may 
happen and make the necessary changes to treatment plans. Armed with 
actionable insights, new treatment models and alerts can be developed 
to de-emphasize opioid medication use for at-risk individuals. This 
includes the right decision support tools for our providers at the 
point of care.
    Ultimately, technology alone will not be able to curb this 
epidemic. We must also use the information and insights we have gained 
to continue to educate our providers, patients, and communities on the 
proper adherence and potential risks of opioid use. America's 
prescription opioid epidemic continues to be a public health crisis. 
Using descriptive, diagnostic, predictive, and prescriptive analytics, 
we have an opportunity to identify at-risk individuals and change the 
course to help address the epidemic.
    We have the data. We need your help to share it.

    List of recommendations to consider:

          Pass S. 778--Prescription Drug Monitoring Act of 2017

          Pass S. 1850--Protecting Jessica Grubb's Legacy Act

          Create and enable authorized access to an integrated, 
        secure data repository for opioid prescriptions, treatments, 
        overdoses, and individuals at risk
                                 ______
                                 
                 [Summary Statement of Sanket J. Shah]
    Using healthcare data and analytics can play a key role in helping 
to combat this national crisis. Descriptive, diagnostic, predictive, 
and prescriptive analytics make it possible to identify individuals who 
are at risk of becoming opioid dependent. Armed with this information 
and the right technologies, healthcare providers and communities can be 
better informed about the risk of possible dependency.
    We know potential determinants for opioid dependency include 
gender, age, whether the patient sought treatment for an acute injury 
or a chronic condition, and the size of the dosage and duration of the 
prescription. \1\
---------------------------------------------------------------------------
    \1\  America's Opioid Epidemic. BCBSA.
---------------------------------------------------------------------------
    To truly have accurate predictive analytics we need more data 
sources. Currently, we find ourselves in data silos across the public 
and private healthcare sectors. My recommendation is to open the lines 
of communication and pathways to share data for a ``holistic view'' to 
help combat this epidemic. The Federal Government has the means and 
infrastructure to create an integrated data environment which we can 
source from at local and state levels. Having access to such a vast 
data repository will enable the creation of robust predictive analytics 
that leverages multiple variables such as social determinants of 
health, family and medical history, and access to complete episodes of 
care. A secure and encrypted data repository would enable our 
healthcare informaticists to administer and deploy innovative 
technologies to enhance our predictive capabilities. This includes 
advanced machine learning algorithms for deeper pattern analyses from 
both the provider and patient fronts.
    To accomplish this, we simply need access to more substance abuse 
data. I ask you all to consider and support the Prescription Drug 
Monitoring Act of 2017 (S. 778) which requires any state that receives 
Federal grant funding to establish a prescription drug monitoring 
program to share their data with other states. In addition, this act 
also contains components to help fund a data sharing hub which I spoke 
of earlier. I also ask you all to consider supporting the Protecting 
Jessica Grubb's Legacy Act (S. 1850) which calls for modernizing Part 2 
to align with HIPAA regulations and will grant appropriate sharing of 
substance use disorder records to ensure persons with opioid use 
disorder and other substance use disorders receive accurate diagnoses 
and effective treatment.
    Once predictive analytics have identified at-risk individuals for 
developing an addiction, we can use prescriptive analytics to offer up 
actionable insights. Providers and health plans can predict what may 
happen and make the necessary changes to treatment plans.
    Using descriptive, diagnostic, predictive, and prescriptive 
analytics, we have an opportunity to identify at-risk individuals and 
change the course to help address the epidemic.

    List of recommendations to consider:

          Pass S. 778--Prescription Drug Monitoring Act of 2017

          Pass S. 1850--Protecting Jessica Grubb's Legacy Act

          Create and enable authorized access to an integrated, 
        secure data repository for opioid prescriptions, treatments, 
        overdoses, and individuals at risk
                                 ______
                                 
    The Chairman. Thank you, Mr. Shah, and thanks to each of 
you.
    We will now have a round of 5 minute questions from 
Senators. We have a number of Senators, so if you will try to 
keep the questions and answers within 5 minutes, everyone can 
participate.
    Senator Young.
    Senator Young. Mr. Shah, I enjoyed your testimony where you 
spoke, quite a bit, about the potential descriptive, 
diagnostic, predictive and prescriptive analytics, and their 
potential to make it possible to identify individuals who are 
at-risk of becoming opioid dependent.
    I see limitless opportunities to harness these tools. Not 
just in the area of healthcare and improving health outcomes, 
but proactive policing. We are starting to see it in various 
geographies. Our office is working on child abuse prevention 
through the utilization of some of these tools.
    What can the Federal Government do now to encourage 
exploration, more exploitation of these analytic tools while 
ensuring patient protections?
    Mr. Shah. Thank you for the question.
    Well, there are a couple of things I mentioned in my 
testimony here. I think that the Prescription Drug Monitoring 
Act of 2017, first of all, allows opening up the doors, if you 
will, for states to access data from states that are currently 
employed in the Prescription Drug Monitoring Programs. So 
sharing of that data will enable opening up some insights that 
we can share at local, state, and Federal levels.
    In addition to that, I think part of the question also 
revolves around privacy and ensuring that we have our 
communities protected. When you are looking at this epidemic 
and you are looking at patient safety, it is imperative that we 
do leverage substance abuse disorder records to identify 
patterns, identify deeper analytics, and then share that 
amongst communities.
    As it relates to safety and privacy, part of that includes 
protecting the data and accessing it in a protected environment 
and in an encrypted environment, if you will. So I think that 
would address some of the concerns from the public as it 
relates to sharing some of their data. Again, the focus here is 
to really handle the data with care and then ultimately benefit 
these communities across the U.S.
    Senator Young. Yes, there are some serious privacy 
concerns. We need to make sure we do our very best to address, 
and not confuse, wherever possible correlation with causation, 
as they say, as well. Right?
    I see some affirmative nods from our panelists.
    Mr. Shah, specifically talking about predictive analytics, 
which you brought up, analytics that enable us to leverage data 
to anticipate what is to come.
    In order to use this tool, you claim we need more data 
sources and open communication. We need to break down the data 
silos across the private and public healthcare sectors, and 
then connect the dots, as it were.
    Can you elaborate on what those data silos are and then 
tell me why we cannot tear down these silos right now? Does 
this have, in part, to do with market power and vendor 
interest? Or are there other legal barriers or are there other 
dynamics that you would like to speak to, sir?
    Mr. Shah. Sure.
    Well, there are a lot of stakeholders, and data is very 
important to these individual stakeholders. You have the 
provider systems. You have the payer systems. You have 
pharmaceutical companies. And then, you also have consumer-
created data as well.
    A part of the barrier here is just identifying an 
environment and creating an environment to access the 
information, collaborate.
    Part of the concern here is that most organizations are 
very protective of their data because it is certainly an asset 
for that particular organization.
    We are starting to get there, as it relates to sharing some 
of our important data from a consumer perspective and it is 
also from a private industry perspective. But the challenges 
remain and I think we need assistance to help break down those 
barriers.
    Senator Young. What options, if you have some in mind, do 
we have for action at the Federal level to catalyze or 
incentivize change in this area that facilitates more data 
sharing?
    Mr. Shah. One suggestion I would have is to, again, part of 
that Prescription Drug Monitoring Act of 2017 is to create an 
integrated data source that we can all pull from. It would be a 
secure data source leveraging public data that is available.
    Senator Young. Yes.
    Mr. Shah. Ultimately with the goal to, hopefully, bring in 
some of the private data as well so we can start to leverage 
this for not only research and development purposes, but also, 
as you mentioned, for some predictive capabilities across 
various different industries.
    Senator Young. Would you sit down with me, obviously after 
this hearing, but to dive a little deeper on this issue? I am 
not sure, in terms of methodology, how we get there.
    Maybe the National Academy of Sciences could help. We could 
commission them to establish data standards that, over a period 
of years, would be adopted.
    Mr. Shah. Sure.
    Senator Young. The private sector would understand that. 
There, no doubt, would be resistance to this idea. I am Okay 
with resistance, if it is the right thing to do.
    Would you, or other stakeholders, other panelists, or 
people watching these proceedings, I just welcome them to work 
with our office on this?
    Thanks so much for being here, all of you.
    The Chairman. Thank you, Senator Young.
    Senator Murray.
    Senator Murray. Well, thank you, to all of our panelists 
today.
    Dr. Mahon, let me start with you.
    Throughout our hearings, we have heard about the need for 
more options for states' opioid disposal. I was particularly 
interested in the drug disposal bags you mentioned during your 
testimony, which offer, I assume, individuals a way to dispose 
of dangerous drugs on their own.
    Tell us a little bit about those bags. Who provides them? 
How do they work and whether they could be made more widely 
available?
    Dr. Mahon. Absolutely.
    As part of our disposal bag that we provide to all of our 
patients, any time a patient fills a first-time opioid 
prescription, we mail them a disposal bag. Express Scripts will 
mail the member a disposal bag to their home with instructions 
on how to properly, safely dispose their unused medications.
    Then we also subsequently offer the consultation by the 
specialized pharmacists who can continue to have the 
conversation with the member on any questions that they may 
have about proper disposal.
    The way the bags work, they have activated carbon inside 
the bag where the member just simply pours a little bit of 
water and their pills in the bag. It immediately renders the 
drug deactivated and the bags themselves are biodegradable. 
They are safe to be disposed in the member's home.
    We really feel that that is the best mechanism to allow 
members safe disposal in their home.
    Senator Murray. Are you finding that people use them?
    Dr. Mahon. Yes. We even are having members respond back to 
us. They are writing us letters and notes back.
    One, thanking us for the educational letters we are sending 
them and basically saying, ``Thanks for the heads up.''
    Two, ``Thank you for letting me or giving me large enough 
of a bag where I can dispose not just my own medications, but 
my family's unused medications,'' to be able to properly get 
them out of the homes.
    Senator Murray. Is there any way we can make these more 
widely available? Do you have any ideas?
    Dr. Mahon. What we have seen in our program, what works 
really well, is having a targeted, data-driven approach. So not 
just giving the disposal bags at random times during the year, 
but really, when it matters the most: when the patient is just 
leaving the pharmacy counter and they are going home.
    We know that they have gone home with that medication. They 
are probably going to have some additional meds left; giving it 
to them at that time, and then having that correlation with 
that specialized pharmacist that is coaching them through the 
process, really helps.
    Senator Murray. Interesting. Thank you.
    Ms. Green, you talked about the potential use of PDMP's to 
combat the opioid crisis, but also the need for data standards, 
so these programs can be effective.
    Can you provide some examples to us about how data and 
requirements for maintaining data can be different across 
PDMP's, and how that can lead to systems that are not able to 
communicate or share information effectively?
    Ms. Green. Certainly.
    Right now, we have a number of standards about who can 
access the data and under what conditions, and those can vary 
from state to state.
    For example, one of the best options we have with PDMP's is 
to facilitate widespread integration of the data into 
Electronic Health Records. And yet, we have a number of states 
that do not allow the PDMP report to actually be placed into 
the Electronic Medical Record, and they have different 
standards for access and use of that data, once it is actually 
in the medical record.
    We need consistent standards for placing the PDMP report in 
a medical record and upon placement, allowing all of the same 
standards for use and disclosure that apply to other 
information in that medical record to apply to the PDMP data.
    Those are the kinds of standards we now need to be 
reconciling and making more uniform.
    Senator Murray. Do we need some single data hub to ensure 
interoperability or is there another way to do it?
    Ms. Green. We already have 44 states that are actually 
sharing data and the others that are in the process of making 
legal changes.
    But I would suggest, yes, we need a federally funded, 
single hub that is based on the interests and the needs of 
states and the public officials at the Federal level because 
that is what will generate the reconciling and the uniformity 
of some of these standards so that we can maintain the 
interest.
    Some of the differences are occurring in different states 
because of some of the particular interests that have been 
affecting the laws and policies that go through some of the 
states.
    If we can create uniformity at the Federal level, have it 
be federally funded, then we can create the kind of uniformity 
and standards that we are all looking to do at this point in 
time. And that would be under the Prescription Drug Monitoring 
Act of 2017.
    Senator Murray. And Dr. Clark, real quick, I am encouraged 
by data sharing, but I am concerned about individual privacy, 
especially when people with substance use disorders are worried 
that their treatment could impact their job or family, and may 
discourage them from getting care.
    What issues do we need to be aware of when it comes to 
protecting patient privacy?
    Dr. Clark. First, we need to distinguish between those 
people who are at-risk from those people who already have 
developed a substance use disorder problem and are seeking 
treatment. Those are two different populations of people.
    Predictive analytics work very well for those people who 
are at-risk, but for those people who are attempting to present 
their treatment, we need to keep in mind they are much more 
vulnerable than the former group.
    What we do not want to do is to discourage people, who are 
trying to get treatment, from postponing that treatment. 
Because if they postpone treatment, then they continue their 
demand for illicit substances or the misuse of alcohol while 
their children, and their employer, and others are trying to 
cope with them.
    As Sally Satel, M.D. pointed out, ``We must be realistic 
about who is getting in trouble with opioid pain medications. 
Contrary to popular belief,'' she said, ``It is rarely the 
people for whom they are prescribed. Most lives do not come 
undone, let alone end in overdose after analgesia for a 
specific problem.'' We should not let these different 
populations of people----
    Perhaps that is what we could use predictive analytics for, 
is to differentiate the populations so that we could have 
appropriate standards with that.
    Senator Murray. Thank you.
    The Chairman. Thanks, Senator Murray.
    Senator Cassidy.
    Senator Cassidy. Thank you all.
    Ms. Green, I came in on the tail end of Senator Murray's 
criticism of PDMP data, but what I heard, I am thinking, 
``Right on.''
    You say 44 states--and I am not challenging you, I am just 
exploring--44 states share data, but do they share all of each 
other or just interstate compact?
    Ms. Green. It depends on the states. Some states share 
widely across, some states have decided to only share within 
their region, and the reason for that----
    Senator Cassidy. That is fine.
    Ms. Green. Yes, Okay.
    Senator Cassidy. We have limited time. I do not mean to be 
rude.
    Ms. Green. That is all right.
    Senator Cassidy. As I know, and I have mentioned before, 
Missouri, for example, does not have PDMP.
    But it is also my understanding that a lot of states do not 
real time put up their data. There is a delay and that most 
states do not allow it to be shared with law enforcement. And I 
read--I do not know if it is up to date--that in 17 states, the 
Veterans Administration does not share data with the PDMP.
    Is all of that correct as far as you know?
    Ms. Green. Not quite.
    Law enforcement is able to access in every state, but the 
standard is different. Twenty-eight states allow it based on 
bone fide investigation, others use a court ordered subpoena.
    Senator Cassidy. But in terms of Mr. Shah's hope that it 
can be used proactively, relatively few allow a proactive 
sharing.
    Ms. Green. That is correct because PDMP laws are very 
specific about who can access the system for what purpose, and 
that purpose of proactive sharing would have to actually be 
written into the laws.
    Senator Cassidy. Now, Ms. Mahon, after Hurricane Katrina, 
it was amazing. All these patients had been displaced on 
chemotherapy, everything else. And somebody flipped a switch, 
and every doctor treating a Katrina evacuee had access to their 
prescription data. It was amazing.
    A doctor in Oklahoma City seeing one of my patients could 
access my patient's records and would call me, but he then knew 
the Interferon my patient was on, and the complications, et 
cetera, etc.
    It tells me--and I have actually seen an even more robust 
data base than this--that we are futzing around with PDMP's, 
but someone like you, Express Scripts, already has a more 
robust dataset than PDMP's.
    Is that a fair statement?
    Dr. Mahon. We have the dataset of the information if the 
member is using their pharmacy benefit card at the pharmacy 
counter. So if the patient chooses not to use their pharmacy 
benefit, and they are paying cash for their prescription, those 
are the claims that we would not see.
    Senator Cassidy. When I say I have seen a more robust 
version, I have seen a robust version of a dataset which now 
includes the cash payment. Every Veterans Administration is 
real time as in when the pharmacy fills it, it is uploaded. 
This is a mechanism to avoid for a compliance with anti-
kickback regulations with government programs.
    We are, again, futzing around with PDMP with all its 
limitations and we have before us a robust dataset that is even 
better than what I saw after Hurricane Katrina.
    Mr. Shah, this seems like nirvana for someone like you. You 
can sort by provider. You can sort by patient and, I presume, 
geolocate within this dataset.
    Would you agree?
    Mr. Shah. Yes, absolutely. I would definitely agree with 
that, especially from an analytic point of view.
    When we have a conformed dataset like that, you can slice 
and dice from various different factors, get to the root cause 
drivers of certain trends.
    Certainly to your point, Senator Cassidy, it absolutely 
would be ideal.
    Senator Cassidy. To believe that this dataset exists and is 
compliant with current law, then we should be exploring how to 
use it.
    By the way, I have a bill in for PDMP because right now, 
that is the only weapon you have to fight this war against 
opioid abuse. You need to use it, but I keep on saying, we 
should go nuclear. We should be using this dataset, not the 
PDMP, and I just say to assert.
    Dr. Clark, you probably gathered I am a doctor, 
hepatologist, treated a lot of patients with liver disease, and 
some patients with liver disease got theirs from shooting 
drugs. Some are actively shooting drugs.
    I say this not to challenge, but again to explore. If you 
looked at the history of somebody I was treating, I would have 
everything. He shot drugs in the past or he is still shooting 
drugs. When did you stop? Did you share a needle? Sexually 
transmitted disease. Have you had a blood transfusion? Did your 
mother have hepatitis when you were born? All this stuff. What 
is your mental health history? Because if I am going to treat 
him with Interferon, that is what we were using at the time, 
mental health was very important.
    Now all that was covered under HIPAA, not under the mental 
health provisions. So your sensitivity, I thoroughly respect. 
My gosh. If somebody had come for those records, I would have 
barred the door and used whatever weapons I could to keep 
somebody from using those. But at the same time, you sense that 
there should be a greater sensitivity. But I would, again, 
point out that I had all that information whether the person 
was bipolar, for example.
    Knowing that you are the advocate, and not challenging you, 
but just to explore. Why would HIPAA not be adequate when, 
frankly, I am already getting that data and I am only covered 
by HIPAA because I am not a mental health professional?
    Dr. Clark. I think we need to be careful about a 
presumption of dysfunction. We also need to distinguish 
monitoring from surveillance.
    If we are not careful, we will wind up viewing patients as 
dysfunctional by definition and we will be using healthcare 
records for purposes of surveillance for purposes beyond----
    Senator Cassidy. Okay. Now that is totally foreign from 
what I was saying.
    Dr. Clark. No, it is not, sir. I am not just trying to 
argue with you.
    I am just saying when you start creating larger scale data 
bases for the purposes of predictive analytics, you need to be 
concerned about surveillance. When you start regarding every 
patient as being dysfunctional from the outset, then you need 
to be concerned about surveillance, and that is what concerns 
me.
    One of the things about 42 CFR Part 2 is you inform the 
patient. You ask the patient. Dr. Mahon's perspective----
    Senator Cassidy. You have to be brief, because the Chairman 
is about to start rapping on us.
    Dr. Clark. Yes. In Dr. Mahon's perspective, you have the 
patient actively involved.
    The Chairman. We are over time.
    Dr. Clark. The patient's consent is a guarantee.
    Senator Cassidy. I yield back. Thank you for your 
forbearance.
    The Chairman. Thank you, Dr. Cassidy. I am sure we will get 
back into that subject.
    Senator Bennet.
    Senator Bennet. Thank you, Mr. Chairman.
    Thank you to the panel for sharing your expertise on this 
important issue.
    One Coloradan dies every 17 hours from an opioid overdose. 
Our state, as a result, is already trying to make strides to 
prevent the spread of opioids with technology.
    When I recently visited the University of Colorado Health 
Emergency Room, I saw how they used Electronic Health Records, 
in conjunction with their prescription drug monitoring program, 
to obtain the information they need in just one click. Patients 
leaving the E.R., as a result of this, are leaving with 
prescriptions for pain medications have dropped from 20 percent 
of patients to 12 percent over a 3-year period.
    Last year, the state also passed a bill to expand provider 
access to PDMP, so they can better identify patients at high 
risk of doctor shopping or opioid diversion.
    Today, I will be introducing the Every Prescription 
Conveyed Securely Act along with Senators Warren, Heller, and 
Toomey. The bill directs providers to use electronic 
prescribing for opioids and other controlled substances 
prescribed under Medicare Part D. The aim is to reduce 
fraudulent prescriptions, give providers and pharmacists real 
time information on opioid use, and streamline prescribing for 
both the provider and the patient.
    I wonder, Dr. Mahon, if you could say a word about that 
bill and what you have seen in states that have required 
doctors to e-prescribe controlled substances?
    Dr. Mahon. Absolutely, and thank you very much for the work 
on this, Senator.
    What we have seen specifically, there are seven states 
today that currently do utilize electronic prescribing. In 
particular, the State of New York mandated not just electronic 
prescribing for all prescriptions, but in particular, 
controlled substances.
    What we have seen in our data, looking at the 2017 data in 
our claims that 89.8 percent of all controlled substances are 
now being e-prescribed in the State of New York compared to 
only 21 percent nationally.
    That really tells us that having this information is going 
to be critical and crucial in helping identify, track, and 
manage these controlled substances that, today, we certainly do 
not have the ability to do so.
    Senator Bennet. Thank you.
    Before I go to Mr. Shah, Mr. Chairman, I wonder whether I 
could enter into the record a letter of support for that bill 
from a number of organizations, including one represented here.
    The Chairman. It will be.
    Senator Bennet. Thank you, Mr. Chairman.
    The Chairman. Thank you.
    Senator Bennet. Mr. Shah, in your testimony, you spoke 
about the problems associated with doctor and pharmacy 
shopping. You gave an example about a Medicare enrollee that 
received 73 prescriptions, I think it was----
    Mr. Shah. Yes.
    Senator Bennet [continuing]. For opioid drugs from 11 
prescribers and filled them at 20 different pharmacies.
    Do you think by expanding electronic prescribing, would 
that help to reduce these kinds of fraudulent transactions?
    Mr. Shah. Yes, absolutely. I mean, there are two factors to 
that.
    One, prescribers are able to pull up information about 
where this particular patient has gone or how many times they 
are refilling it in near real time.
    Quite importantly, I think what Senator Murray had talked 
about a little earlier, is in some states having providers 
actually look at the PDMP data bases to identify, ``There is 
something going on with a particular individual. They just had 
a refill 7 days ago. Maybe I want to consider an alternative 
pain therapy or some member education or patient education 
options.''
    To answer your question, I think that would certainly be 
something that would help curtail this doctor, pharmacy 
shopping that we are seeing amongst our communities.
    Senator Bennet. Thank you. I appreciate that testimony.
    I hope, Dr. Clark, that we will be able to overcome our own 
presumption of dysfunction in the U.S. Congress and actually 
move some of this legislation forward.
    Grateful to the panel.
    The Chairman. Thank you very much, Senator Bennet.
    Senator Isakson.
    Senator Isakson. Mr. Shah, is there any convincing evidence 
and data that shows how many people overdose on opioids that 
were prescribed and how many overdosed on opioids that were 
obtained--it might have been prescribed to somebody--but were 
obtained illegally in some other mechanism?
    Mr. Shah. Sir, your question, I believe is there any 
evidence on where individuals that have obtained----
    Senator Isakson. Get it illegally.
    Mr. Shah. I do not have any of those figures with me right 
now, but I know there are studies of individuals that have 
obtained prescriptions through legal means, quite frankly, and 
then there are also figures in articles that I have read where 
individuals are stealing prescriptions or obtaining them 
illegally.
    I do not, unfortunately, have the statistics to answer your 
question right now.
    Senator Isakson. But it is true that a lot of legally 
obtained opioids are, in fact, sold or otherwise distributed to 
people.
    Mr. Shah. Absolutely. Yes. One hundred percent.
    Senator Isakson. I will tell you a number. My staff is 
responsible for the credibility of this number. I am not taking 
responsibility for it, although I have a great staff and he is 
behind me.
    [Laughter.]
    Senator Isakson. I am sure he is right, but I could not 
believe it when I read my homework last night.
    In the State of Georgia, we have over 541 million doses of 
opioid circulating in the state. That is 54 doses for each man, 
woman, and child living in our state. That is a state of 10.5 
million people.
    How does that sound to you in terms of accuracy?
    Mr. Shah. It does sounds accurate, yes.
    Senator Isakson. It is way too much and it is way too many.
    A year ago Monday, I had some major surgery. When I became 
conscious and started recognizing what was going on post-
surgery, I got into an educational session with my physician 
and the pharmacist on opioids, Oxycodone, Hydrocodone, all 
kinds of things like that.
    Because of being a United States senator, and having a 
great Chairman like Lamar Alexander, and people who focus on 
things like the opioids, and people like you testifying--even 
though I was still recovering from the surgery and anesthetic 
that I had--I knew that was dangerous stuff.
    I asked him, ``Are there alternatives to Hydrocodone and 
other things that I could go ahead and take? I do not want to 
run the risk of getting anything bad.''
    Basically, I was prescribed 6 Tylenols a day and dealt with 
my pain with 6 Tylenols a day. My discomfort, it did not deal 
with all the pain.
    Do we do a good enough job of aggressively counseling 
patients who have procedures that are painful by recommending 
alternative sources other than opioids?
    Mr. Shah. I have a personal example to share with you as 
well. To answer your question, I think not enough of a good 
job, but we are getting there.
    My father actually had knee surgery in Chicago where, a 
very similar situation. An immigrant here, does not speak a lot 
of English that well, and the provider had come in and just 
said, ``Here are your prescriptions.'' They happened to be 
opioid prescriptions, of course, for pain management after the 
knee surgery without much follow-up or education in that sense. 
So, I think, to answer your question.
    Then, there are all these alternatives, but we have to be 
careful with those alternatives because of the CDC, I believe, 
had released a study here on morphine equivalents which, in 
high dosages, can be just as detrimental as what you are seeing 
with traditional opioids like you said with the Hydrocodone, 
OxyContin, and Percocet, and whatnot.
    We are getting there. It is still about education. I think 
we continue the education of the providers and our member 
communities as well.
    Senator Isakson. Well, everybody's testimony has been 
fantastic, and I appreciate it a lot, but you have all really, 
in one way or another, endorsed the more data, the better. 
Better for predictive analytics, which is a great term. It 
sounds like I went to Georgia Tech when I say that. I did not 
go there, but it sounds good anyway.
    We can better cast when somebody would be at-risk and 
hopefully get them into a preventive environment before they 
ever are tempted with drugs would help us an awful lot. And I 
think it would be one of the roles this Committee can take.
    Thank you, Mr. Chairman.
    The Chairman. Thank you, Senator Isakson.
    Senator Kaine.
    Senator Kaine. Thank you, Mr. Chairman.
    Thank you to the witnesses.
    Here is a question that was suggested to us from my Medical 
Society of Virginia physicians. They indicate that in Virginia, 
Federal requirements for electronic prescribing software for 
opioid medications and other controlled substances are very 
onerous, expensive, and they are not easily compatible with 
existing e-prescription software so that the Federal 
requirement does not match the existing software.
    The barriers make many small and medium sized providers 
reluctant to e-prescribe controlled substances, leading to gaps 
in data and less secure prescriptions.
    What regulatory barriers exist to making opioid related e-
prescribing software more affordable and accessible to 
providers? What is Express Scripts, or other EHR vendors, doing 
to make the opioid e-prescription software more affordable and 
more integrated with existing e-prescribing platforms?
    Dr. Mahon. Thank you for the question, Senator.
    I will start this off with just really understanding the 
technology. While certainly there may be some initial humps and 
hurdles you have to get through, once the providers are on an 
e-prescribing platform, it really is pretty simple to go 
through the authentication processes.
    This technology has advanced, especially just in the last 
12 months. We went from having to require a key fob 
authentication to now just requiring your fingerprints to be 
scanned to now just even using a smart application to get 
through the system. So it really is just moving with the 
technology, where it is going, and really allowing providers to 
get through the process much more simply.
    In terms of other barriers or it could be rural barriers, 
lack of Internet access or things like that, we do believe that 
those can be overcome as well.
    Then in terms of the expense or the cost piece of getting 
approved for e-prescribing, it is relatively inexpensive. It is 
typically around $100 to get e-prescribing software and 
licenses once you are on an e-prescribing platform.
    Overall, we definitely see the technology is evolving that 
this is really going to become a pretty simple process.
    Senator Kaine. Do the other witnesses have anything 
additional to add on that?
    Mr. Shah.
    Mr. Shah. Yes, I will speak to the data interchange issues 
that you had alluded to.
    To address that, there is certainly technological advances 
that are happening every day. Quite frankly, we can work with 
some of these private vendors through API's to get data 
interchanged in a conformed format. But that requires 
collaboration, quite frankly.
    To answer your question about some of the challenges, yes, 
that is part of the sharing, getting that data from various 
different sources unstructured or not. We have the means 
through technology to integrate and assimilate that data in a 
conformed fashion.
    I think that is a barrier that can be easily overcome, but 
it will require some collaboration, as I mentioned earlier.
    Senator Kaine. Let me ask one other question, and actually, 
Dr. Clark, this might be a good question for you, but others 
could weigh in too.
    In the immediate preceding Congress, Senators Toomey, 
Portman, Brown, and I acted together to include a provision to 
establish a lock-in program for Medicare Part D for controlled 
substances, similar to the lock-in program that was already 
existing in Medicare.
    These lock-in programs, as you know, prevent inappropriate 
prescriptions from crossing the pharmacy counter by locking in 
certain beneficiaries at-risk of drug abuse into choosing, and 
they get to choose, but then once they choose, they choose a 
sole prescriber and pharmacy.
    Those who worked on the provision, we were having some real 
concerns about the way CMS is approaching it, that they are 
taking a very limited approach to the kinds of at-risk 
beneficiaries that should be included in the lock-in program in 
Medicare Part D.
    I wonder if you would talk about what you think are the 
benefits and disadvantages of lock-in programs, and what we 
might do to make sure that we do it in an effective way.
    Dr. Clark. Well, sir, first to your earlier point, I wanted 
to point out that we want to make sure that technology does not 
change so fast that the practitioners in the field cannot keep 
up with it, and that is an essential point. As a physician, 
that is something that you struggle with.
    That gets to the next question about the lock-in program.
    As a senior citizen, I want to make sure I can have access 
to geographically accessible and administratively accessible 
prescription programs. I do not object to lock-in programs per 
se, but we want to make sure that senior citizens are not 
precluded from accessing care. I have a 99-year-old mother. She 
has limited physical abilities. I am 72 this year myself.
    If, in fact, we create programs that prevent people having 
access, then basically we get back to my earlier assumption of 
the presumption of dysfunction and senior citizens will bear 
the burden of that.
    Senator Kaine. It is not just senior citizens, because 
Medicaid is a similar program.
    But are there other comments on lock-in programs quickly?
    Ms. Green. Yes, if I can quickly say this.
    Lock-in programs are great for reducing the use of specific 
drugs, but we need a corresponding increase in support by 
public and private payers for non-opioid alternatives for 
treatment.
    Senator Kaine. Great. Thank you very much.
    Thanks, Mr. Chairman.
    The Chairman. Thanks, Senator Kaine.
    Senator Roberts.
    Senator Roberts. Thank you, Mr. Chairman.
    Many thanks to you and Senator Murray. I think this is the 
fifth hearing that we have had here. Thank you for your 
leadership.
    I am the co-Chairman of the Senate Rural Healthcare Caucus 
and I am particularly concerned with the growing rise of the 
opioid epidemic and substance abuse in rural areas.
    Senator Donnelly and I introduced legislation last year 
that would help our rural areas develop substance abuse 
prevention, treatment, and recovery facilities.
    Tomorrow, this Committee will mark-up legislation that I 
introduced to reauthorize the State Offices of Rural Health. 
These offices, obviously, play a very critical role in 
strengthening the rural healthcare delivery system by 
collecting and disseminating data that helped inform the Senate 
to make good decisions.
    Ms. Green, how can we ensure that Federal, state, and local 
efforts to combat the opioid crisis take into account the 
unique situation faced by rural Americans?
    Ms. Green. We need to increase our efforts for tele-health 
in projects like Project ECHO out of New Mexico, which is 
specifically designed to match up practitioners in rural areas 
with specialists on opioid addiction at various specialty hubs.
    That would be one of the best efforts we can make to ensure 
that there are appropriate resources available to address the 
opioid problem.
    Senator Roberts. What role can state offices of rural 
health play in ensuring that the opioid policies are based on 
reliable data and sound evidence?
    Ms. Green. In terms of getting reliable data, one of the 
best things we can do in making sure we have reliable data is 
to ensure that there is effective coordination at the state 
level, as well as the Federal level, for the types of data, 
patient data that we should be consolidating and making 
available to different practitioners at different levels--
local, state, and Federal--to be able to address the opioid 
crisis.
    We do not have enough consolidation of patient data, and 
there is no comprehensive patient prescription history or 
patient history in any one single source yet.
    Senator Roberts. We are at the base of the mountain.
    On the Senate Finance Committee, I am interested in the 
potential for the electronic authorization--I may be going over 
plowed ground here--or e-prior authorization within Medicare 
Part D to strike a proper balance between limiting the 
unnecessary dispensing of opioids and avoiding overly 
burdensome requirements.
    Dr. Mahon, you mentioned in your testimony some successful 
opioid management initiatives, including the use of prior 
authorization for certain prescriptions. But I have 
consistently heard from constituents with concerns about the 
burdens this practice can place on physicians and patients.
    How can e-prior authorization be used by Medicare Part D 
and Medicare Advantage plans to help uphold responsible 
dispensing of opioids while reducing the physician and patient 
burden?
    Dr. Mahon. Thank you for the question, Senator.
    Certainly, as we have seen in our data in our program that 
if a physician absolutely needs to provide more medication to 
any given member, they have the ability to request an 
authorization, a prior authorization of that drug.
    Where our electronic prescribing is going to be critical is 
prior to the point of prescribing, the physician will already 
know if this particular drug requires the quantity level or a 
prior authorization. So they are able to go through that 
process pretty quickly without having to do it after the fact, 
which is what is really important as we continue to manage this 
opioid epidemic.
    But in particular, what we saw in our data was that only 4 
percent of all providers who initially wrote a prescription for 
a 30-day supply--subsequently got reduced to a 7-day supply--
requested an authorization to give the member more than seven.
    Senator Roberts. Four percent.
    Dr. Mahon. Only 4 percent of doctors requested an 
authorization where they basically said, ``Thank you. This 
patient only needs a 7-day supply and it is sufficient.''
    Senator Roberts. Well, I thank you for that.
    The FDA has approved several Abuse Deterrent Formulations, 
or the ADF's, in recent years, which have shown effectiveness 
in reducing opioid misuse and abuse. However, many patients 
still lack the proper access to these drugs due to all sorts of 
factors.
    What policy should the Federal Government consider to 
increase patient access to ADF's and what is the relationship 
between the ADF's and prior authorization?
    Dr. Mahon. From our standpoint, we do caution you to avoid 
allowing more ADF's into the system. The reason for that is 
these medications are not necessarily abuse-proof. So they 
could be an abuse deterrent, but they are not abuse-proof and 
that is really the big difference that individuals can still 
continue to overdose on these medications.
    There are many different techniques online that you can 
just go and look. How do I boil, or cook, or do whatever you 
can do from these medications?
    It is really critical that we do not rely just on ADF 
medications as a single tool. We need to rely on this epidemic 
to really solve it from multiple angles. Educating the 
physicians, working with the pharmacies, consulting the members 
to really solve the problem, and ADF's alone are not going to 
be able to do that.
    Senator Roberts. I appreciate it.
    I yield back, Mr. Chairman. Thank you very much.
    The Chairman. Thank you, Senator Roberts.
    Senator Smith.
    Senator Smith. Thank you very much, Chairman Alexander and 
Ranking Member Murray.
    I really appreciate this meeting today and all of your 
testimony.
    We all know, because we have been sitting through these 
conversations, that roughly 75 percent of people who began 
abusing opioids in the 2000's got there through an opioid 
prescription. We are so focused on trying to figure out how we 
can tackle this by improving the use of data.
    Chairman Alexander, I have a letter from Prime 
Therapeutics, which is a company headquartered in Minnesota, 
which is also doing some really good work on this.
    I would like to ask consent to enter this into the record 
on their behalf.
    The Chairman. It will be included.
    Senator Smith. This letter identifies that Prime has 
figured out how to achieve a 71 percent decline in the number 
of high risk opioid users through their work which, I think, 
gives us some hope in this difficult time.
    But Ms. Green, I wanted to ask you, and really, anybody on 
the panel, just one more question about this issue of what we 
can do at the Federal level to facilitate the sharing of 
information.
    This is fresh in my mind because I have talked to so many 
providers in Minnesota and have really found that our PDMP is 
cumbersome, that they have to log-in and then log-out. We do 
not require people to do this, in part, because it is so 
cumbersome. I am sure there are lots of reasons for this.
    But what are the top one or two things that we could do to 
facilitate this data sharing?
    Ms. Green. In every Federal grant and contract that is let 
out, including CDC, the Justice Department, SAMHSA, they all 
address trying to improve PDMP's.
    We need to include consistent standards in all of those 
grants and contracts that indicate, ``The following standards 
must be met. There must be comprehensive data sharing, and 
there must be data made available for public health 
surveillance purposes,'' and all the other purposes that we 
need that data made available for, because that is the funding 
that is primarily being used now to improve PDMP's.
    Senator Smith. Thank you.
    Does anybody else want to make any other comment on that?
    Dr. Clark?
    Dr. Clark. Yes, I want to echo what she said because, in 
fact, if we can enhance our PDMP's, which is something we have 
been struggling with almost now for 10 years, if we can enhance 
the technology.
    We do not have the kind of information that people argue 
that we need. What we need to know is that the epidemic is 
actually subsiding. And so, we want to make sure we do not put 
something in place that, as they say, closes the barn door 
after the horse is gone.
    Senator Smith. Yes.
    Dr. Clark. PDMP's have been around. We have known for 10 
years what some of the limitations are. We just have not done 
much about those limitations. It would be useful if we did 
something about those limitations.
    Senator Smith. I would like to just change direction a 
little bit here.
    We have been hearing a lot how technology can help prevent 
opioid abuse, which is so important. But I am also really 
interested in how technology can help us provide good care to 
people who are in the process of recovery.
    I have had a chance to talk with some folks in Minnesota 
that are working on this. Particularly, we have an example in 
Little Falls, Minnesota, rural Minnesota, which is how they 
describe it as sort of a hub and spoke strategy, which is 
really about coordinating care and providing all of the 
services that somebody needs when they are in the middle of 
recovery.
    I would be very interested to know from this panel, maybe 
Dr. Mahon, you could address this because you talked about 
patient engagement.
    How can we better use technology to coordinate care as we 
address this epidemic?
    Dr. Mahon. Absolutely. Thank you for the question, Senator.
    What we have seen as part of our program, it becomes really 
critical to collaborate with the provider. So one of the many 
things that Express Scripts is doing is we are actually 
communicating with the doctors via the Electronic Health Record 
and sending them information at the point of care when the 
individual is about to receive the next prescription.
    Giving them information, not just on what the morphine 
equivalent dose is of that individual, which is really, really 
important, but then also giving them recommendations on 
subsequent therapy or care.
    Whether it is if we see somebody is on 600 mgs of morphine 
equivalent, this individual should probably be on a Naloxone, 
or giving them additional information on how to properly 
counsel them on where else services are available in-network, 
whether it is a treatment facility and relaying and sharing 
that information within the Electronic Health Record becomes 
really, really important. So we definitely see that.
    Then the second piece is using telemedicine and virtual 
pharmacists, which is what we are leveraging to actually be 
communicating with the members as well, and then sharing that 
information back to the pharmacy. So the way we view it is 
really encapsulating the member within the care continuum of 
the pharmacy, their home, and the physician office. It becomes 
really important.
    Senator Smith. Right. Thank you very much.
    Mr. Shah, I can see you are eager to respond, but I am out 
of time. So we will look forward to hearing your comments 
later.
    Thank you.
    The Chairman. Thank you, Senator Smith.
    Let me operate here as a little bit of a skeptic on the 
Federal Government's ability to deal with lots of data. I think 
of HealthCare.Gov. I think of the actual mess with Electronic 
Healthcare Records that we got into. We ended up having six 
hearings on that that we did not plan to have during our 21st 
Century Cures.
    Vanderbilt University has been trying for years with health 
records that Meaningful Use One was helpful; Meaningful Use Two 
was Okay; Meaningful Use Three was terrifying. And when the 
government tried to combine that with the ability of getting 
everybody to change to a method of payment that had to do with 
outcome, it became an even bigger difficulty.
    I have two questions I would just like to ask you to 
comment. One involved e-prescribing. One involves the Federal 
data hub.
    We have a million doctors out there, many of whom are still 
having a hard time with Electronic Health Records. The best way 
to get your healthcare record from one hospital to another is 
to put it in your suitcase, and carry it over there, and give 
it to the doctor, even though we spent more than $30 billion 
encouraging people to do that.
    Would it not be a good idea on e-prescribing, to the extent 
we move ahead, to do it slowly, to try it out? You have, I 
think, seven states are already doing it. If the technology is 
evolving, why should we not let it evolve before we impose it 
on doctors who are already struggling with lots of burdens?
    What suggestions would you have about making the movement 
to e-prescribing, if that is where we need to go, as something 
that we can do in an evolving way? If McDonald's was 
introducing a new sausage gravy, they would do it in Dallas and 
Nashville and see how it tasted before they imposed it in 
14,000 stores. So that is one.
    As far as the prescription drug data base, do you really 
want that at the Department of Justice or do you really want it 
at all?
    Why not, instead, establish the standards that states can 
use and why not leave room for Amazon, Google, and Delta? We 
can do our airline reservations just instantly with these 
things. We can order from Amazon, why not let the private 
sector come in and offer a way to fill in whatever gap remains 
after we have gradually improved the state PDMP's?
    Count me as a skeptic. Express Scripts is not a Government 
company, and if it were, it would probably not be nearly as 
successful.
    What would you do about evolving e-prescription and 
improving the state PDMP's, but leaving room for the private 
sector to fill in the gaps?
    Ms. Mahon.
    Dr. Mahon. Yes, I will start first.
    We certainly understand that there are limitations of 
prescribers and I think we can definitely look at what the 
State of New York did when they launched the mandated 
controlled substance prescribing. They did leave out some 
exceptions: long term care facilities, certain situations. So 
because we have an epidemic, I do not think we have time to 
wait.
    Now, what we can do is carefully figure out what are some 
of those areas that we could leave some exceptions and give 
them extended dates of when they should be ready to be mandated 
for controlled substance prescribing.
    But certainly, the rest of the physician practices that do 
have the ability to do so, we should move on them now.
    Ms. Green. The private sector can assist with certain 
innovations, but their focus is ultimately market dominance and 
competition. That is the nature of the laissez-faire economic 
system we have and this is part of the problem we have had with 
some of the data systems already. That starts to take hold.
    What I am interested in is, yes, the standards but also I 
want the Federal Government to maintain those standards that 
are appropriate for the public interest.
    The Chairman. But standards do not require establishing an 
entire Federal hub of data in the Justice Department, does it?
    Ms. Green. Well, but it would require it if it were 
federally funded.
    The Chairman. What makes you think the Federal Government 
has that competence?
    Ms. Green. Well, first of all, I would say that I never 
assume the Federal Government has competence in anything.
    The Chairman. Yes, well then, why would you suggest a 
Federal hub for all this data in the Justice Department?
    Ms. Green. Because the Federal Government is not going to 
be maintaining the hub; they are going to be setting out the 
Federal funding for it and the standards are going to be within 
the Federal funding. So they are not actually going to be, 
there is not going to be a Federal agency that is actually 
doing the hub.
    The Chairman. They did a great job with HealthCare.Gov.
    Ms. Green. I agree with you that there were mistakes there, 
but what I am suggesting----
    The Chairman. It was a disaster.
    Ms. Green. There were mistakes there, but what I would 
suggest is that we learn from those mistakes.
    The Chairman. It took a year and a half to fix it.
    Ms. Green. Then what I would suggest is that this body is 
the perfect body to heed the lessons learned from health-dot-
com or health-dot-gov and to translate those into the standards 
that are actually put in the hub.
    The hub that is in the Prescription Drug Monitoring Act 
does not suggest that the Federal Government is actually going 
to run that. What it does is it says the Federal Government----
    I have no particular concerns if it is the Department of 
Justice or HHS, but what I am concerned about is that Federal 
funding is a mechanism by which to ensure that the standards 
for public interest actually are maintained.
    The Chairman. Thank you, Ms. Green.
    I am out of my own time and I do not want to set a bad 
example.
    Senator Warren.
    Senator Warren. Thank you, Mr. Chairman.
    I am actually going to pick up where you are. My view is we 
do need data, and we need the data in order to inform better 
decisions. The Federal Government collects all kinds of data 
right now.
    We get the monthly data reports on how many people are 
employed and whether the unemployment rate is going up or going 
down.
    We collect data on education, how many people are in 
school, how many college graduates there are, how much student 
loan debt is outstanding.
    We collect data on highways, and transit, and all kinds of 
things because we think it informs making better decisions 
going forward. Our alternative is to grope around blindly and 
it seems to me that is not the best way to make a decision.
    With e-prescriptions, so that the doctor instead of handing 
you a piece of paper, which you could lose, which you could 
alter, which you could photocopy, particularly where an opioid 
is concerned, a doctor who says, ``Wait. I am going into a 
secure system. I am doing my authorization. This is how much I 
am prescribing,'' it cannot be altered at that point. As Dr. 
Mahon says, someone can then come back and even say to the 
doctor, ``Really? That is how much you want to prescribe for 
somebody who only weights 70 pounds? Or you may want to 
consider other concerns or ways to help protect this client.''
    I see this exactly the other way that e-prescribing is 
something that is enormously valuable.
    We are in the middle of a crisis and that, right now, the 
latest data I have for 2016--I think some of our panelists may 
have later data--that only about 14 percent of doctors are e-
prescribing controlled substances, and that these are the 
prescriptions that are most likely to be misused and diverted. 
And yet, we are sending them to pharmacies on little pieces of 
paper.
    I want to ask the question the other way. Dr. Mahon, you 
work at Express Scripts, which is one of the largest Pharmacy 
Benefit Mangers in the country.
    If more doctors were using e-prescribing to write 
prescriptions for opioids, would we have a better understanding 
of the opioid crisis in America?
    Dr. Mahon. Absolutely. One quick comment.
    The numbers did go up last year. Now we are at 20 percent 
of physicians that are e-prescribing.
    Senator Warren. Great.
    Dr. Mahon. But still, not enough.
    Really, absolutely, we would because one of the biggest 
gaps that we have today, as a Pharmacy Benefit Manager, is we 
do not have access to cash claims.
    Those members who are getting the paper copy prescription, 
walking into the pharmacy under the radar, obtaining those 
medications, we have no visibility and no ability to intervene 
subsequently.
    All of the things that we are doing right now with sharing 
information with doctors, sending a member a disposal bag, 
educating them in their home, we do not have the ability to do 
it if it is going under the radar in the cash processing 
system.
    Giving them the ability to e-prescribe would automatically 
give all of us the visibility to these claims, and we would 
have the ability to appropriately intervene on these members 
when it is most needed in real time.
    Senator Warren. In a large sense, it lets us see how many 
opioids are being prescribed.
    Dr. Mahon. Yes.
    Senator Warren. Who is prescribing them, where there are 
pockets of use in the country, and that alerts us to problems 
as we go through.
    I want to ask another part of this, because I think you 
have worked on this question.
    Does e-prescribing actually have the potential to change 
how many opioids doctors prescribe?
    Dr. Mahon. Absolutely.
    In terms of these limits that we have set from an over-
utilization perspective, the physician would be able to see if 
I am limited to, let us say, two 7-day fills within a certain 
time period, that would be visible in the system.
    Ideally, we would be in a world where anytime a physician 
would write a prescription, they would see how are they 
adhering to the CDC guidelines, to the state prescribing 
requirements of opioids that would ultimately help us limit.
    Senator Warren. The advantages to collecting this data in 
e-prescribing are not only how we use them from a public health 
point of view. From an enforcement point of view, we actually 
have some evidence, it makes a difference in the decisions that 
doctors make patient by patient in using the best practices in 
prescribing opioids.
    I am about to run out of time, but I just want to push on 
one more part and that is, Ms. Green, you have spoken today 
about the need to make sure that the PDMP's in different states 
can talk to each other.
    Do we also need to make sure that the PDMP's can talk to 
Electronic Health Records and e-prescribing systems?
    Ms. Green. Yes, all three of the systems need to 
efficiently and seamlessly talk to one another so that we can 
timely transfer the data and proactively monitor patient and 
prescriber behavior, which will allow us to intervene at an 
early point to address any problems.
    Senator Warren. Good, just powerfully important. Thank you 
very much. I see I am out of time, but count me in for more 
data.
    The Chairman. I already was, Senator Warren.
    Thank you.
    [Laughter.]
    The Chairman. Thank you for your good questioning.
    Senator Casey.
    Senator Casey. Thank you, Mr. Chairman.
    I want to thank you and the ranking member for these 
hearings. I guess this is the fifth and we are grateful for 
that. This is an epidemic, a challenge that warrants that kind 
of attention. I wish every serious subject in the Senate was 
the subject of this kind of a review, all these hearings. So we 
are grateful for that.
    I wanted to start with the challenge. There are lots of 
ways to describe this crisis and we are grateful to have the 
expertise in the room to help us understand one part of the 
challenge.
    This is a trauma which is affecting every member of the 
American family in one way or the other. It does not matter 
where you live. It does not matter what age you are. It does 
not matter who you are. It is affecting all of us in some way 
or another; everyone from infants born addicted, so-called 
Neonatal Abstinence Syndrome, all the way to other members of 
the American family.
    One of the real telling indicators in my state is rural and 
urban. We have data that tells you that in urban America, it 
could not be worse. Here is an example.
    Overdose deaths in Philadelphia estimated to have reached 
1,200 last year with fatal overdose death rate in 2015 of 
almost 47 deaths per 100,000. In Chicago, it is a fraction of 
that, 15. In New York, it is 11. So Philadelphia at 47, 
unfortunately, is a high number.
    You can actually make the case that it is worse in rural 
Pennsylvania. We have 48 rural counties out of 67. We are told 
that the pace of rural deaths is actually faster; a 42 percent 
increase year over year of 15 to 16 versus 34 in urban areas.
    Fulton County, that is a county of about 14,000 people on 
the Maryland border, a small county, a rural county, in 2016, 
the opioid death rate was 74 per 100,000. Compare that to what 
I thought was a high number of 47 in Philadelphia and much 
higher in those other urban areas around the country.
    The President declared a so-called public health emergency, 
but we have not heard nearly enough from the Administration on 
this issue. We need leadership there. It is a big vacuum. Even 
as we are here having five hearings here, work being done in 
the Congress, we need the President to lead on this because of 
the nature of the crisis.
    The last point I will make before questions is the 
importance of Medicaid, the Medicaid program as well as the 
Medicaid expansion.
    Pennsylvania, 127,000 people receive treatment via Medicaid 
for one of two general conditions: mental health challenges, as 
well as the broader category of substance use disorder, not 
just the opioid crisis.
    We need Medicaid badly and we need Medicaid expansion. So 
those who talk about cutting Medicaid had better square that 
with what they are saying about the opioid crisis.
    I wanted to start with Professor Shah. I worked on a number 
of bills that speak to these concerns. We have a grandparents 
bill that will be marked up in the Aging Committee tomorrow. I 
am working with Senator Collins to create a Federal taskforce 
charged with supporting grandparents who are raising 
grandchildren and other manifestation of the family trauma.
    I worked with Senator McConnell on a bill to deal with this 
Neonatal Abstinence Syndrome that I mentioned before. We got 
that legislation passed to do two things, really. One is to 
have a new strategy to address research and program gaps in the 
Federal Government mostly across HHS. The second part of the 
bill requires recommendations for preventing and treating so-
called NAS, Neonatal Abstinence Syndrome.
    We are told by the ``Philadelphia Inquirer and Daily News'' 
that the number of drug exposed babies born in Philadelphia and 
the four surrounding counties doubled between 2009 and 2016.
    Tell us, if you can, in light of the data that hospitals in 
Pennsylvania are collecting, tell us how we can use that data 
to positively impact what you mentioned in your testimony, the 
so-called predictive analytics to reduce either the numbers or 
the mortality when it comes to infants.
    Mr. Shah. In addition to just actual data collected at 
facilities of claims out of EMR data, we have to couple all of 
that information together to really identify meaningful 
insights.
    But then, there is additional data sources that we need to 
bring in. Some of the conversations revolved around social 
determinants of health data. Where are these individuals 
living? What is their income, their education level? These are 
all factors that contribute to meaningful and targeted 
analytics.
    To answer your question, it is a collaboration of various 
different institutions to share that data.
    Secondly, it is to bring in additional data too and then 
break down those data silos. And by having that historical data 
and in that context from an analytic perspective, we will be 
able to better identify certain instances where there may be an 
individual, perhaps a neonatal, or whatever the case may be 
that may be headed in that direction. Perhaps identify 
opportunities for intervention early to help curb that.
    To answer your question, it is, again, revolving around 
data and integration.
    Senator Casey. Collection of the data is obviously the 
foundation of it.
    Mr. Shah. That is correct.
    Senator Casey. I know I am out of time. I will maybe submit 
a question for a couple of the members of the panel.
    Ms. Green, I wanted to ask you a question. I will submit in 
writing, about the challenge with regard to different states, 
especially a state that borders, like ours does where we have 
states that are nearby having both different standards and 
requirements for the Prescription Drug Monitoring Programs.
    But in the interest of time, I will submit that for the 
record.
    Senator Casey. Thank you, Mr. Chairman.
    The Chairman. Thank you, Senator Casey.
    Senator Murray, do you have additional comments?
    Senator Murray. I just have one comment for the record I 
want to make clear on that.
    That actually the data hub on HealthCare.Gov was the part 
that actually worked. And it was responsible for identity 
proofing and eligibility verification.
    The part that did not work, that failed, was the Web site's 
ability to support high traffic and allow enrollees to actually 
browse through the options. That was actually a private 
contractor, CGI Federal; so just for the record.
    But I did want to thank all of the witnesses today. This 
was an excellent hearing and very helpful as we work to develop 
and pass legislation on this critical topic.
    Mr. Chairman, I want to thank you for holding this hearing 
today and for remaining very focused on building a bipartisan 
effort to address this issue.
    Thank you.
    The Chairman. Thank you, Senator Murray.
    These have been bipartisan hearings, which mean we agree on 
them. The witnesses have been very helpful, very good.
    I am going to pursue my point a little bit. I have been 
around long enough to know that everybody assumes, ``Here is a 
great idea. Let us get the Federal Government do it.'' But the 
Federal Government lacks the capacity, in many cases, to do 
things well.
    Do any of you have any suggestions?
    A couple of you commented about what to say to doctor. Let 
us say 20 percent are e-prescribing now. That means 800,000 are 
not. They might say back to you, ``Well, look. You have me all 
tied up in a knot on Meaningful Use Three. You have new 
regulations on merit-based payment and I am mad about that. And 
now, here you come with a whole big set of regulations about 
how I am going to do e-prescribing.''
    What can you say to that doctor or what can you say to us 
about how we can avoid creating an oppressive burden?
    Mr. Shah, do you have any suggestions?
    Mr. Shah. Going back to your example, the McDonald's 
example that you had shared with us, it is about a phased 
approach.
    I think what we need to do is identify communities and 
organizations that are a little bit ahead of the curve from a 
technological perspective, and really institute and showcase 
the value of the e-prescribing as it relates to curbing this 
opioid epidemic.
    We talked a little bit about New York and Maine, as an 
example, as a success story. I think we need to model that 
instance there to a broader community. But, again, that is 
doing it in a phased type of approach. That is what I would 
recommend to those providers, those that are willing to 
participate and then working your way out.
    The Chairman. Would there be any merit to saying that if 
you do not do it within a certain period of time, the Federal 
Government will do it? E-prescribing, obviously, could be 
required by states, seven are. Right?
    Dr. Clark.
    Dr. Clark. I think providing incentives to states to adopt 
these strategies would be another approach. That way, you can 
phase it in to those jurisdictions that saw the utility of it 
and would be early adopters, and then you could move from 
there.
    You would have states and then perhaps regional endorsers 
or adopters of the strategy. That way, you can avoid the 
encumbrance of, say, federally mandated strategies. So 
incentives versus mandates would be an approach that could 
address your concern.
    I think practitioners, clinicians, and it is not just 
physicians who are prescribers. There are practitioners and 
physician assistants who also prescribe, and it is not just 
opioids. It is benzodiazepine and other medications.
    People would be interested in facilitating better care. As 
you are suggesting, people just do not want to have so many 
obstacles that it interferes with the ability to provide that 
care.
    The Chairman. Any other comments?
    Dr. Clark, do you have any concerns about a Federal hub of 
data in the Justice Department?
    Dr. Clark. Yes, sir. I have large concerns about a Federal 
hub of data in the Justice Department.
    I love our Justice Department. They do a great job, but I 
do not think a repository of clinical information belongs in 
the Justice Department. As I pointed out, there is a difference 
between monitoring and a difference between surveillance, and 
the Justice Department is very much interested in surveillance.
    What we are trying to accomplish here is enhancing the care 
of people and not create an ``I got you'' kind of environment 
where practitioners and patients feel that they cannot rely on 
the healthcare delivery system.
    The Chairman. Thanks to all four of you. This is very 
helpful. I agree with Senator Murray. Very useful to us.
    You can see that we are about to write a bill, so if you 
have very specific suggestions beyond what you have already 
discussed or if you want to reiterate something that you said 
toady or had in your testimony you would like to make sure we 
pay attention to, we would welcome that. That will help us 
write a better bill.
    The hearing record will remain open for 10 days. Members 
may submit additional information for the record within the 
time, if they would like.
    The Chairman. Our Committee will meet again tomorrow, 
Wednesday, February 28 at 9:45 a.m. for an executive session.
    Thank you for being here.
    The Committee will stand adjourned.
    [Whereupon, at 11:45 a.m., the hearing was adjourned.]

                           [all]