[Federal Register Volume 66, Number 140 (Friday, July 20, 2001)]
[Notices]
[Pages 37980-37988]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 01-18166]


-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Health Resources and Services Administration


Children's Hospitals Graduate Medical Education (CHGME) Payment 
Program: Final Methodology for Determination of FTE Resident Count, 
Treatment of New Children's Teaching Hospitals, and Calculation of 
Indirect Medical Education Payment

AGENCY: Health Resources and Services Administration, HHS.

[[Page 37981]]


ACTION: Final notice.

-----------------------------------------------------------------------

SUMMARY: This notice sets forth final methodology for determining full 
time equivalent (FTE) resident count, treatment of new children's 
teaching hospitals, and calculation of indirect medical education (IME) 
payments for the Children's Hospitals Graduate Medical Education 
(CHGME) Payment program, authorized by section 340E of the Public 
Health Service Act (42 U.S.C. 256e), amended by Pub. L. 106-310, The 
Children's Health Act, 2000. In compliance with the Paperwork Reduction 
Act of 1995, the Department obtained Office of Management and Budget 
(OMB) approval of the data collections required and imposed on the 
public (OMB No. 0915-0247).

FOR FURTHER INFORMATION CONTACT: Ayah E. Johnson, Graduate Medical 
Education Branch, Division of Medicine and Dentistry, Bureau of Health 
Professions, Health Resources and Services Administration, Room 8A-08, 
Parklawn Building, 5600 Fishers Lane, Rockville, Maryland 20857; 
telephone (301) 443-1058 or e-mail address 
[email protected].

SUPPLEMENTARY INFORMATION: The CHGME program, as authorized by section 
340E of the Public Health Service (PHS) Act (the Act) (42 U.S.C. 256e), 
provides funds to children's hospitals to address disparity in the 
level of Federal funding for children's hospitals that results from 
Medicare funding for graduate medical education (GME). Pub. L. 106-310 
amended the CHGME statute to continue the program until Federal Fiscal 
Year (FFY) 2005.
    On March 1, 2001, the Secretary published a notice in the Federal 
Register (66 FR 12940) establishing final rules for eligibility, 
funding criteria, payment methodology and performance measures for the 
CHGME program. That notice also sought public comments on proposals for 
(1) The criteria for determining full time equivalent (FTE) resident 
count; (2) the treatment of new children's teaching hospitals with 
respect to resident count; and (3) the methodology for IME payments. 
During the comment period, the Department received comments from 
seventeen interested parties, including hospitals, hospital and 
professional associations, Medicare consulting companies, and law 
firms.
    The Secretary thanks the respondents for the quality and 
thoroughness of their comments. As a result of these comments, the 
Department has made revisions and clarifications in this final notice. 
The comments and Department's responses to the comments, and the final 
rules are set forth below.

General Comments

    Several respondents recommended that the CHGME program follow 
Medicare's rules as closely as possible: (1) Because these rules are 
well defined and are known to those children's hospitals that file 
Medicare cost reports (MCR); and (2) to conform to Congress' intent to 
provide funds to children's hospitals to address disparity in the level 
of Federal funding for children's hospitals that results from Medicare 
funding for graduate medical education. The respondents indicated that 
the Department should make exceptions to compliance with policy 
following Medicare principles only in those instances in which the 
unique characteristics of children's hospitals render the application 
of Medicare principles impossible or undesirable, and it should explain 
the specific rationale for each exception.
    In the implementation of the CHGME program, the Department has 
incorporated applicable Medicare rules and regulations. However, it is 
important to recognize that fundamental differences exist between the 
Medicare and CHGME programs that make certain Medicare rules and 
regulations inapplicable to the CHGME program. For instance:
    (1) The CHGME program includes children's hospitals that span the 
spectrum of pediatric patient care, including acute, rehabilitation, 
oncology, orthopedics, and long term care;
    (2) The CHGME program includes resident training that occurs in all 
areas of the hospital complex for both DME and IME;
    (3) The CHGME program is bound to the FFY in which appropriated 
funds must be distributed without the opportunity to reconcile funding 
across FFYs;
    (4) The Medicare GME payments are associated with treatment of 
Medicare patients;
    (5) The Medicare patient population is primarily non-pediatric; and
    (6) The Medicare program monies come from a trust fund.

Determining FTE Resident Counts Beginning in FFY 2001

    With the exception of some revisions for clarification, the 
criteria for determining FTE resident counts beginning in FFY 2001 are 
unchanged from those proposed in the March Federal Register notice. 
Beginning in FFY 2001, for hospitals, that report residents to 
Medicare, there will be an order of priority for acceptance of resident 
counts submitted to the CHGME program:
    (1) For the most recent cost report periods ending on or before 
December 31, 1996, a hospital must report the latest settled FTE 
resident count or a ``preliminary'' fiscal intermediary (FI) determined 
resident count. All preliminary FI determined counts must be determined 
according to HCFA and Medicare criteria. Hospitals may not use the 
``preliminary'' numbers that were used for the FFY 2000 CHGME program 
unless those FTE resident counts have since become finalized or are 
validated according to HCFA and Medicare standards.
    (2) For settled cost reports in other years, the CHGME program will 
accept the latest settled cost report. If a settled cost report has 
been reopened, the CHGME program will accept the latest settled count 
or, if available, the most recent ``preliminary'' FI determined FTE 
count.
    (3) For unsettled cost reports, the CHGME program will accept in 
order of priority:
    (a) The most recent preliminary FI determined FTE resident count 
prior to the application deadline; if not available, then
    (b) The amended filed FTE resident count; if not available then
    (c) The as filed FTE resident count.
    For hospitals that do not report residents to Medicare (i.e., file 
low or no utilization cost reports) but have been operating a residency 
training program and participated in the CHGME program in FFY 2000, the 
calculation of FTE resident counts remains unchanged from the FFY 2000 
application. Unlike the FFY 2000 applications, however, beginning in 
FFY 2001, the CHGME program requires hospitals to report FTE resident 
counts based on the hospital cost reporting period rather than on the 
FFY. In the June 19, 2000, Federal Register notice the Department 
provided examples of how these hospitals could determine FTE resident 
counts for the 1996 cap year and the 3-year rolling average. The CHGME 
program will accept this methodology for the 1996, 1998 and 1999 cost 
reporting periods.
    If these hospitals wish to revise their FTE resident counts for 
these cost reporting periods, they must submit a detailed explanation 
of the revision with supporting documentation. The supporting 
documentation must be in compliance with HCFA/Medicare standards used 
to determine FTE resident counts (e.g., rotation schedules).

[[Page 37982]]

    Beginning with the cost report period ending in 2000, these 
hospitals will be required to use the methodology described in 42 CFR 
413.86(f)(2) to determine FTE resident counts; that is, to measure the 
amount of time that a resident works during the cost report period 
based on the number of days. In addition, these hospitals will continue 
to be required to apply Medicare standards for documenting the 
residents to be counted and calculating their FTE time for purposes of 
determining an FTE resident count.
    Hospitals which did not report residents to Medicare and did not 
participate in the CHGME program in FFY 2000, although they were 
training residents at that time, are required to use the methodology 
described in 42 CFR 413.86(f)(2) to determine their FTE resident count 
for their cap year and 3-year rolling average. Like all hospitals which 
do not report residents to Medicare, they will be required to apply 
Medicare standards for documenting the calculating of their FTE 
resident counts.
    Some hospitals have filed a combination of full, low utilization, 
and no utilization cost reports. For these hospitals, the Department 
requires that they file the actual FTE resident counts reported for 
those cost report periods where an E-3, Part IV worksheet has been 
filed. For those cost report periods where a low or no utilization cost 
report period was used, the hospitals should recreate their FTE 
resident count using the methodology referenced above.
    Several respondents recommended that resident counts used for 
distribution of funds after FFY 2002 for all hospitals be based on 
Medicare cost reporting data. The respondents indicated that such a 
change should include sufficient time to resolve any technical issues 
that arise for hospitals that did not report residents in 1996 for 
determination of their resident cap. They noted that, while in the 
short term, it is necessary and appropriate to accommodate those 
hospitals that did not report residents to Medicare, it is important 
over the longer term for consistency and equity in the resident 
counting methodology that all eligible hospitals file resident counts 
on their Medicare cost reports.
    The Department does not have the option of requiring resident 
counts used for distribution of funds to be based on Medicare cost 
reporting data since section 340E(e)(1) of the CHGME statute requires 
that:

* * * interim payments to each individual hospital shall be based on 
the number of residents reported in the hospital's most recently 
filed Medicare cost report prior to the application date for the 
Federal fiscal year for which the interim payment amounts are 
established. In the case of a hospital that does not report 
residents on a Medicare cost report, such interim payments shall be 
based on the number of residents trained during the hospital's most 
recently completed Medicare cost report filing period.

    Several respondents requested that HRSA clarify or define a 
``preliminary FI determined resident count'' and indicated that some 
FIs may not provide a ``preliminary FI determined resident count'' 
prior to the formal resettlement of the revised cost report.
    To clarify, a ``preliminary FI determined resident count'' with 
respect to a settled cost report that has been reopened is any resident 
count that the FI has determined during the normal course of cost 
report review (e.g., audit) prior to formal resettlement of the cost 
report. For example, if the FI and the hospital have negotiated the FTE 
resident count but not yet completed the paperwork to officially settle 
the reopened cost report, the hospital can submit the negotiated FTE 
resident count as a statement written by the FI describing the 
negotiated FTE resident count as ``preliminary'' to the completion of 
the resettlement paperwork. The CHGME program will not accept any FTE 
resident counts from amended reopened cost reports unless the FI 
submits it to the CHGME program as a valid ``preliminary'' FTE resident 
count.
    For cost reports that have never been settled, a ``preliminary'' 
FTE resident count issued by an FI would be any resident count the FI 
has generated during the normal course of cost report review (e.g., 
desk review) prior to settlement of the cost report.
    In some cases during the FFY 2000 CHGME application process, FIs 
issued ``preliminary'' numbers for FTE resident counts for some of the 
children's hospitals. Hospitals may not use these ``preliminary'' 
numbers for the FFY 2001 or future CHGME program application unless 
those FTE resident counts have since become finalized or are validated 
according to HCFA and Medicare standards through the normal course of 
business.
    Regarding the use of Medicare standards in issuing ``preliminary'' 
FTE resident counts, one respondent indicated it was unaware of 
Medicare standards and that individual intermediary standards are not 
published.
    HCFA provides numerous manuals for FIs and hospitals which outline 
the standards and definitions used in preparation and review of 
Medicare cost reports. These manuals are available electronically on 
the Internet at http://www.hcfa.gov and for purchase through the 
National Technical Information Service (NTIS) Clearinghouse. If 
hospitals have questions or concerns about their FI's interpretation/
application of these standards, they should communicate with their FI 
or HCFA Regional Offices.
    Several respondents raised the issue of applying a written 
agreement for purposes of training residents between a hospital and a 
non-hospital site retrospectively in order to count FTE residents 
rotating through those non-hospital sites.
    As stated in the March 1, 2001 Federal Register notice, all 
resident training in non-hospital sites may be included in the FTE 
resident count as long as the hospital and non-hospital site are in 
compliance with 42 CFR 413.86(f)(3) and (4).

New Children's Teaching Hospitals

    The Department is making final the definition of ``new children's 
teaching hospitals'' as proposed in the March 1 Federal Register 
notice. For purpose of the CHGME program, a ``new children's teaching 
hospital'' is a hospital which:
    1. Has its own Medicare provider number as a children's hospital 
described in Sec. 1886(d)(1)(B)(iii) of the Social Security Act but did 
not train residents until it began training residents from an already 
existing program, less than three cost report periods prior to the FFY 
in which CHGME payments are being made; and
    2. Has historically participated in a residency training program 
(e.g., a pediatric department within a larger teaching hospital) and 
subsequently receives its own Medicare provider number as a children's 
hospital described in Sec. 1886(d)(1)(B)(iii) of the Social Security 
Act.
    ``New children's teaching hospitals'' are distinct from those 
teaching hospitals that are participating in a new medical residency 
training program defined under 42 CFR 413.86(g)(12). Medicare 
regulations at 42 CFR 413.86(g)(6)(i) and (7) set forth criteria for 
applying the caps and rolling averages in these teaching hospitals with 
new medical residency training programs.

Establishing the Cap for New Children's Teaching Hospitals

    Unlike children's hospitals that can receive adjustments to their 
caps for new residency training programs according to 42 CFR 
413.86(g)(6), ``new children's teaching hospitals'' are treated like 
all other hospitals that have trained residents for 3 years after the 
first program began training residents,

[[Page 37983]]

as explained in 42 CFR 413.86(g)(6)(i)(C). According to 42 CFR 
413.86(g)(4), the hospital's FTE resident cap is based on the 
unweighted FTE resident count from the most recently completed cost 
report period ending on or before December 31, 1996. Since ``new 
children's teaching hospitals'' would not have trained residents during 
the most recent Medicare cost reporting period ending on or before 
December 31, 1996, they would have a cap of zero.
    To provide an adjustment to the cap of zero, the CHGME program will 
allow these hospitals to add FTE residents to their cap based on the 
following-described Medicare regulations:
    1. The formation of a new residency program within the first 3 
years after the first program begins training residents as described in 
42 CFR 413.86(g)(6); or
    2. The execution of an affiliation agreement for an aggregate cap, 
as set forth in 42 CFR 413.86(g)(4) and 63 FR 26338, published in the 
Federal Register on May 12, 1998, with the following exceptions:
    a. A ``new children's teaching hospital'' participating in the 
CHGME program for the first year must establish an effective date of 
the agreement for the purposes of the CHGME program. For the first 
year, unless otherwise specified, the Department will use as the 
effective date of the affiliation agreement for an aggregate cap the 
date that the hospital becomes eligible for the CHGME program. This 
effective date will only apply to the CHGME program. A hospital must 
also have an effective date of July 1 for the Medicare program. 
Subsequent to the first year of the affiliation agreement, the 
effective date must comply with the above cited Federal Register final 
rule which specifies an effective date of July 1 for all affiliation 
agreements. The CHGME program allows this exception because hospitals 
must meet eligibility criteria and have their caps determined prior to 
the CHGME application deadline. If the CHGME program application 
deadline occurs before July 1, some hospitals would have a cap of zero 
and thus be excluded from receiving funds. By deviating from the 
prescribed Medicare final rule, the CHGME program will not place some 
hospitals in this position.
    b. Unlike the Medicare program, for the first year, the CHGME 
program will not prorate the cap based on the effective date of the 
cap. Instead, the full value of the cap as determined by the 
affiliation agreement will be used.

Establishing FTE Resident Counts for New Children's Teaching 
Hospitals

    In general, the FTE resident count from each hospital reflects the 
residents trained during the Medicare cost report period, limited by 
the cap (the unweighted allopathic and osteopathic FTE resident count 
from the most recent cost report period ending on or before December 
31, 1996). Payments to each hospital are based on the average of the 
FTE resident count for the most recent Medicare cost report and the 
prior two cost reports (3-year rolling average), subject to funds 
available for DME and IME, respectively.
    For establishing FTE resident counts, ``new children's teaching 
hospitals'' are divided into two categories: (1) Those training 
residents from an existing residency program that received and will 
continue to receive funds under the CHGME program; and (2) those 
training residents from an existing residency program that has never 
received funds under the CHGME program (i.e., residents that have not 
previously been claimed for payment under the CHGME program).

``New Children's Teaching Hospitals'' Training Residents Previously 
Claimed For Payment Under the CHGME Program: FTE Resident Count

    The Department requires ``new children's teaching hospitals'' 
training residents who were originally trained in a program that 
received and will continue to receive funds under the CHGME program to 
wait until they have completed a medicare cost report period before 
applying for payments from the CHGME program. The CHGME program would 
have provided payment to the hospital originally training the 
residents, prior to the completion of a Medicare cost report period by 
the new children's teaching hospital, and would not want to pay two 
hospitals for training the same residents.
    These ``new children's teaching hospitals'' must apply the 3-year 
rolling average according to Medicare regulations at 42 CFR 
413.86(g)(5). Over a 3-year period, the ``new children's teaching 
hospital'' will gradually increase its number of FTE residents that can 
be claimed on the CHGME application as the children's hospital that 
originally trained those FTE residents gradually decreases its resident 
count.

``New Children's Teaching Hospitals'' Training Residents Not 
Previously Claimed for Payment Under the CHGME Program

    Since payments under the CHGME program are based on FTE resident 
counts from a completed cost report filing period, ``new children's 
teaching hospitals'' training residents never previously claimed for 
CHGME payment that have not completed a cost report filing period at 
the time of the CHGME program application would not have an FTE 
resident count for a full Medicare cost reporting period to report to 
the program. These ``new children's teaching hospitals'' must submit a 
partial-year FTE resident count in their initial applications to the 
CHGME program according to the following methodology:
    a. Divide the number of FTE residents trained during the period 
from the day the children's hospital becomes eligible for the CHGME 
program to the CHGME application deadline by the number of days during 
this period to produce the average number of FTEs per day.
    b. Multiply the average number of FTEs per day by the number of 
days the hospital will train residents during the FFY in which payments 
are being made.
    The concept of converting a partial period into a full cost report 
period is found in the Medicare regulations at 42 CFR 413.86(g)(4) and 
(e)(5)(ii). Since the CHGME program is paying hospitals for training 
residents during the FFY for which payments are being made, the 
Department will convert a partial training period to reflect the amount 
of time the hospital will train residents during the FFY for which 
payments are being made. Although this methodology delineates the 
method by which partial-year residents are counted, it is important to 
note that all counts are subjected to the cap set by the affiliation 
agreement.
    After the initial application year, payments to ``new children's 
teaching hospitals'' training residents never previously claimed for 
CHGME payment will be based on the actual FTE resident count from the 
most recently completed Medicare cost report period. Once these 
hospitals have completed three Medicare cost report periods, the 3-year 
rolling average will apply.
    Under Medicare, hospitals training residents that are not in a new 
residency program, as defined in 42 CFR 413.86(g)(12), are subjected to 
the 3-year rolling average. For example, under Medicare, in the first 
year these hospitals would calculate the 3-year rolling average as 
follows: [FTE resident count for current year + 0 (FTE residents for 
prior cost report period) + 0 (FTE residents per penultimate cost 
report period)] divided by three (3).
    One purpose of this Medicare policy is to avoid paying two 
hospitals for the same residents. Over the course of 3 years the 
hospital which was originally training the residents ``rolls down'' its 
FTE resident count and the hospital which is assuming training ``rolls 
up'' its FTE resident count.

[[Page 37984]]

    The rationale adopted by the CHGME program in deviating from this 
Medicare policy is that, for the ``new children's teaching hospitals'' 
training residents that were never previously claimed for CHGME 
payment, the issue of double payment for residents is not relevant 
since the program is not currently paying for them. Therefore, to treat 
all hospitals participating in the CHGME program equitably, the 
Department will not impose a 3-year rolling average on the FTE 
residents counts until these ``new children's teaching hospitals'' have 
completed three cost reporting periods.

Determining Indirect Medical Education (IME) Payments to Hospitals

    The March Federal Register notice invited comments on the proposed 
methodology for calculating IME payments organized by: (1) The purpose 
and use of payments under the program, (2) case mix, (3) number of FTE 
residents, (4) teaching intensity factor, (5) patient volume, (6) 
outpatient services, and (7) determination of payments. A discussion of 
the comments received and the Department's responses follows.

Purpose and Use of IME Payments

    The CHGME statute requires the Secretary to make payments to 
children's hospitals for IME associated with operating approved 
graduate medical residency training programs for each of fiscal years 
2000 through 2005. Section 340E(b)(1)(B) describes IME payments as 
covering ``expenses associated with the treatment of more severely ill 
patients and the additional costs relating to teaching residents in 
such programs.'' Section 340E(d)(2) of the Act requires the Secretary 
to determine IME payments by considering:
    1. Variations in case mix among children's hospitals; and
    2. The hospitals' number of FTE residents in approved training 
programs.
    The Department utilized the broadest interpretation of this 
legislative mandate to determine that IME payments determined for 
purposes of the CHGME program should reflect the indirect costs of GME 
as defined by statute throughout the entire hospital complex, similar 
to the allowances for the calculation of DME payments unlike Medicare 
which limits IME payment adjustments to certain areas of the hospital.

Determination of Case Mix

    The determination of case mix is unchanged from that set forth in 
the March notice. Beginning in FFY 2001, all applicant hospitals must 
submit a case mix index (CMI), based on the discharges from the most 
recently completed cost reporting period, using HCFA-DRG Version 17 
with the appropriate HCFA Version 17 weights reported to the ten-
thousandth decimal place. All DRGs must be included in the calculation 
of this CMI. In subsequent years, the version of the HCFA-DRG, to be 
used by hospitals, will be updated annually. To determine which version 
of the HCFA-DRG grouper and weights hospitals will use in completing an 
application to the CHGME program, the following methodology will be 
used:
    1. Based on the application deadline, the year end of the most 
recently completed cost reporting period will be determined for the 
majority of applicant hospitals.
    2. The version of the HCFA-DRG grouper and weights used to 
calculate the CMI for the FFY corresponding to the year end of the most 
recently completed cost reporting period for the majority of applicant 
hospitals will be used to calculate the CMI.
    If a children's hospital eligible to participate in the CHGME 
program has not completed a Medicare cost reporting period prior to 
submission of an application to the CHGME program, it would base its 
CMI on discharges from the day it became eligible fo the CHGME program 
until the CHGME application deadline.
    Several respondents requested that DRG 391 be excluded from the 
calculation of CMI beginning in FFY 2000. These respondents argued 
that, as only a few hospitals participating in the CHGME program would 
actually use this DRG code, related to treatment of normal or healthy 
newborns, the exclusion of this DRG would assist in creating equity 
among the hospitals in the program.
    The Department will include all DRGs in the calculation of its CMI 
because the activity of all areas of the hospital complex and the 
severity of illness among the inpatient population that the hospital 
serves need to be reflected in the hospital's CMI in order to treat all 
hospitals equitably. The IME payment is meant to reflect the resources 
used to treat the more severely ill patients in children's hospitals.
    Several respondents suggested alterntive methodologies for 
calculating CMI, including the Resource-Based Relative Value Scale 
(RBRVS) or the All Patient Refined (APR)-DRGs and APR-DRG relative 
weights. In addition, several respondents supported the Department's 
exploration of developing a CMI methodology that is more reflective of 
the resource intensity of pediatric care.
    The Department continues to recognize that the current CMI may not 
be reflective of the relative resource utilization in children's 
hospitals, particularly those providing specialized services, such as 
rehabilitation and will continue to investigate the feasibility of 
developing a CMI that is more reflective of the relative resource 
utilization experienced by children's hospitals. The Department 
anticipates that this effort will be multi-year. Any analyses and 
resulting recommendations would be published in subsequent Federal 
Register publications.

Determining the Number of FTE Residents for IME Payments

    The criteria for determining FTE residents for IME payments is 
different from those proposed. In the March 1, 2001 Federal Register 
notice, the Secretary proposed to determine FTE resident counts for IME 
payment calculation using the ``caps and rolling averages'' consistent 
with Medicare regulation 42 CFR 412.105(f) with the exception of 42 CFR 
412.105(f)(1)(ii)(A). The Department's final criteria for determining 
the FTE resident count for IME payments include all areas of the 
hospital complex as specified in 42 CFR 413.86(f)(1), the regulations 
used to determine FTE resident counts for DME. Time spent by residents 
on required research is also included if it is part of the resiency 
program and the resident carries out the research in either: (1) The 
children's hospital (clinical or bench research); or (2) in a 
nonhospital site where the research involves direct patient care and 
the salaries of both the resident and supervising faculty are paid by 
the children's hospital. Since the FTE resident count used to calculate 
both DME and IME payments will reflect residents rotating through all 
areas of the hospital complex, the unweighted FTE resident count is the 
same for the DME and IME (MCR worksheet E-3, Part IV, line 3.05).
    The criteria used by the Department for hospitals reporting FTE 
resident counts will be the same for IME as they are for DME (see 
description in previous section). ``New children's teaching hospitals'' 
that have not completed a cost report period would use a partial-year 
FTE resident count methodology similar to the methodology used to 
determine FTE resident counts for DME payments (see previous section).
    The calculation of FTE resident counts remains unchanged from the 
FFY 2000 application for hospitals that do not report residents to 
Medicare, have been operating a residency training

[[Page 37985]]

program and participated in the CHGME program in FFY 2000. Unlike the 
FFY 2000 applications, however, beginning in FFY 2001, the CHGME 
program requires hospitals to report FTE resident counts based on 
hospital cost reporting period rather than on FFY. In the June 19, 2000 
Federal Register notice the Department provided examples of how these 
hospitals could determine FTE resident counts for the 1996 cap year and 
the 3-year rolling average. The CHGME program will accept this 
methodology for the 1996, 1998 and 1999 cost reporting periods.
    If these hospitals wish to revise their FTE resident counts for 
these cost reporting periods, they must submit a detailed explanation 
of the revision with supporting documentation that is in compliance 
with HCFA/Medicare standards used to determine FTE resident counts 
(e.g., rotation schedules).
    Beginning with the cost report period ending in 2000, these 
hospitals will be required to use the methodology described in 42 CFR 
413.86(f)(2), without application of the weighting factors described in 
42 CFR 413.86(g)(1), (2), and (3), to determine total unweighted FTE 
resident counts. Medicare measures the amount of time based on the 
number of days during the cost reporting period that a resident works. 
In addition these hospitals will be required to apply Medicare 
standards for documenting the counting of residents and calculation of 
their FTE time for purposes of determining an FTE resident count.
    Hospitals which did not report residents to Medicare and did not 
participate in the CHGME program in FFY 2000 although they were 
training residents at that time are required to use the methodology 
described in 42 CFR 413.86(f)(2), without application of the weighting 
factors described in 42 CFR 413.86(g)(1), (2), and (3), to determine 
their FTE resident count for their cap and 3-year rolling average. Like 
all hospitals not reporting residents to Medicare, they will be 
required to apply Medicare standards for documenting the calculating of 
their FTE resident counts.
    Some hospitals file a combination of full, low utilization, and no 
utilization cost reports. For these hospitals, the Department requires 
that they file the actual FTE resident counts reported for those cost 
report periods where an E-3, Part IV worksheet has been filed. For 
those cost report periods where a low or no utilization cost report 
period was used, the hospitals should recreate their FTE resident count 
using the methodology described above.

Caps and Rolling Average

    Beginning with FY 2001, the Secretary will apply the ``caps and 
rolling averages'', consistent with the Medicare regulatory section 42 
CFR 412.105(f), with the exception of 42 CFR 412.105(f)(1)(ii). In 
place of this subsection, the Department will use the criteria of 42 
CFR 413.86(f)(1), which define FTE counts for DME.
    The Department received a variety of comments on application of the 
cap and rolling averages to calculating IME payments. Several 
respondents recommended that the Department postpone the application of 
the cap and rolling averages to the FTE resident count for calculating 
IME payments until after the FFY 2002 application deadline so hospitals 
which reported residents to Medicare for the cap year (most recently 
completed cost reporting period ending on or before December 31, 1996) 
would have adequate time to resolve any outstanding issues with their 
FIs related to this cost reporting period. Other respondents suggested 
that the Department not apply the caps and rolling average to the IME 
at all, as the CHGME statute does not require it.
    The Department will apply the cap and rolling average to the 
calculation of IME payments beginning with FFY 2001 in order to comply 
as closely as possible with Medicare rules and regulations. The 
Secretary maintains that hospitals which report residents on Medicare 
cost reports have been aware of an FTE cap as early as their 1998-cost 
report and assumes that these hospitals are reporting an accurate FTE 
cap number.
    In addition to the above comments, two respondents argued that if 
the Department were to implement the cap and rolling averages on the 
FTE resident count used in the IME payments, then the cap should be 
based on the unweighted FTE resident count from the most recently 
completed cost reporting period ending on or before December 31, 2000, 
to correspond with the initial year of the CHGME program, FFY 2000. The 
basis for their argument was that previously, children's hospitals did 
not receive IME payments and that, in some cases, the hospitals may 
have added residency programs after the cap year that could not be 
counted toward the cap on residents. In addition, there was a 
misunderstanding that hospitals that did not report residents on 
Medicare cost reports could base their unweighted FTE resident cap on a 
year other than the most recently completed cost reporting period 
ending on or before December 31, 1996.
    To clarify the policy regarding the year upon which the unweighted 
FTE resident count is based, all hospitals must use the most recently 
completed cost report period ending on or before December 31, 1996, to 
determine the unweighted FTE resident count that would be used as the 
cap for calculating of IME payments. This standard definition applies 
to all hospitals participating in the CHGME program regardless of 
whether or not they report residents on their Medicare cost reports. If 
a hospital certifies in its application that it has based its cap on 
the most recent cost reporting period ending on or before December 31, 
1996, and subsequent to a CHGME program review/audit, it is discovered 
that a more recent cost reporting period was used to determine the cap, 
that hospital would be subject to prosecution by the Federal Government 
as it would have committed fraud.

Teaching Intensity Factor

    In the March notice, the Department invited comments on:
    1. The proposed continuation of the use of the Medicare residents-
to-bed ratio (IRB)-based teaching intensity factor in the calculation 
of IME payments. The CHGME program would use the most current PPS IRB 
in its calculation of IME payments;
    2. Application of a cap on the IRB ratio, similar to the cap 
applied by the Medicare program, 42 CFR 412.105(a)(1), whereby the 
ratio may not exceed the ratio for the hospital's most recent prior 
cost reporting period. Application of this cap will not be initiated 
until FFY 2002 due to the proposed change in the definition of bed 
count;
    3. Suggestions on alternative teaching intensity factors, such as 
the Medicare resident-to-average daily census (RADC)-based teaching 
intensity factor (2.8 percent per 0.1 percent increase in RADC ratio) 
or any other analytically justified teaching intensity factor; and
    4. The proposed definition of ``bed count'' to be used in 
calculating the Medicare IRB teaching intensity factor--the sum of all 
available beds per day in the most recently completed cost report 
filing period, including beds and bassinets in the healthy newborn 
nursery, divided by the number of days in that period. If a children's 
hospital eligible to participate in the CHGME program has not completed 
a Medicare cost report period prior to submission of an application to 
CHGME program, it would base its ``bed count'' on the sum of all 
available beds per day, including beds and bassinets in the healthy 
newborn nursery, in the period from the day it became eligible for the 
CHGME program until the CHGME application

[[Page 37986]]

deadline, divided by the number of days in that period.

Teaching Intensity Factor

    Beginning in FFY 2001, the Department will use the IRB ratio to 
determine IME payments. The Department will use the same teaching 
intensity factor that is used by the Medicare Inpatient PPS in 
calculating its operating IME adjustment for the FFY in which payments 
are being made.
    One respondent encouraged the use of the resident-to-average daily 
census (RADC) ratio in factoring in teaching intensity, because the 
RADC ratio measures actual utilization that occurs in the inpatient 
unit and thus provides a more realistic measure of intensity. Three 
respondents supported using the Medicare methodology of computing the 
number of residents per available bed, as consistency with Medicare is 
desirable without a compelling reason to depart from the Medicare 
formula.
    The Department intends to continue to assess various teaching 
intensity factors and formulas designed to capture the IME costs 
associated with caring for more severely ill patients in a children's 
hospital.

A Cap on the IRB Ratio

    To comply as closely as possible with Medicare rules and 
regulations, beginning in FFY 2002, the Department will apply a cap on 
the IRB ratio, similar to the cap applied by the Medicare program 
pursuant to regulations at 42 CFR 412.105(a)(1), whereby the ratio may 
not exceed the ratio for the hospital's most recent prior cost 
reporting period. For those hospitals whose IRB ratio changes, there 
will be a one-year delay in the implementation of the revised IRB.

Beds To Be Included in Calculation of Bed Count

    Beginning in FFY 2001, a bed is defined, for the purposes of the 
CHGME program, as an adult or pediatric bed, including beds or 
bassinets assigned to healthy newborns, available for lodging 
inpatients, including beds in intensive care units, coronary care 
units, neonatal intensive care units, short stay units, and other 
special care inpatient hospital units. Beds in the following locations 
are excluded from the definition: Labor rooms, post-anesthesia or post-
operative recovery rooms, outpatient areas, emergency rooms, ancillary 
departments, nurses' and other staff residences, and other such areas 
as are regularly maintained and utilized for purposes other than 
inpatient lodging.
    Beginning in FFY 2001, children's hospitals will calculate bed 
count to be used in calculation of the teaching intensity factor used 
to determine IME payments using the following methodology: The sum of 
all available inpatient beds per day within the hospital complex in the 
most recently completed cost report filing period divided by the number 
of days in that period. If a children's hospital, eligible to 
participate in the CHGME program, has not completed a Medicare cost 
reporting period prior to submission of an application to the CHGME 
program, it calculates its ``bed count'' using a prorated number. The 
prorated number is based on the sum of all available inpatient beds per 
day within the hospital complex in the period from the day it became 
eligible for the CHGME program until the CHGME application deadline, 
divided by the number of days during that period.
    To be considered an available bed, a bed must be permanently 
maintained for lodging inpatients. It must be available for use and 
housed in patient rooms or wards (i.e., not in corridors or temporary 
beds). Thus, beds in a completely or partially closed wing of the 
facility are considered available only if the hospital put the beds 
into use when they are needed. The term ``available beds'' as used for 
the purpose of counting beds is not intended to capture the day-to-day 
fluctuations in patient rooms and wards being used. Rather, the count 
is intended to capture changes in the size of a facility as beds are 
added to or taken out of service.
    Several respondents recommended that the count of available beds 
used in the intensity factor exclude beds/bassinets used in the ``well-
baby'' nursery as this would be consistent with the Medicare policy. In 
addition, other respondents indicated that the exclusion or inclusion 
of short stay or observation beds should not be each individual 
hospital's determination--it should be program-wide policy consistent 
with Medicare policy.
    The Medicare definition and regulations on counting beds are 
inapplicable to the CHGME program due to the fundamental differences 
between the two programs. Therefore, the Department has defined ``bed'' 
to best carry out the purpose of the CHGME program.
    Although, traditionally, Medicare has excluded beds and bassinets 
used in the ``well-baby'' nursery, it is the understanding of the CHGME 
program that this is primarily due to the fact that beds and discharges 
from the ``well-baby'' nursery have not been factored into the 
calculation of Medicare payments because there is no Medicare 
utilization attributable to this part of the hospital. As all areas of 
the hospital complex are included in the determination of IME payments 
for the CHGME program, the Department feels that this includes all 
relevant available inpatient beds that are utilized within the hospital 
as defined above.
    In addition, if the Department were to follow Medicare policy, as 
stated in Medicare program manual HCFA Pub. 15-1 S. 2405.3.G, on the 
definition of beds to be included in the bed count, beds in hospital-
based skilled nursing facilities or in any inpatient area(s) of the 
facility not certified as an acute care hospital (e.g., long term care 
beds) or beds in excluded units (e.g., rehabilitation, psychiatric) 
would need to be excluded from the definition of beds used by the CHGME 
program in addition to the exclusion of beds/bassinets in the ``well-
baby'' nursery. Because the hospitals participating in the CHGME 
program are not limited to acute care hospitals and the Medicare 
definition of bed count refers only to acute care beds, the Department 
believes that the inclusion of all of these beds would be an equitable 
treatment of all hospitals participating in the CHGME program.
    The Department has followed the Medicare policy as closely as 
possible (see definition above) regarding the inclusion or exclusion of 
short stay or observation beds. Hospitals participating in the CHGME 
program must certify the accuracy of the numbers reported on their 
applications. Hospitals reporting bed counts that include other than 
inpatient beds are subject to prosecution for fraud by the Federal 
Government.

Patient Volume

    As set forth in the March notice, the Department will use inpatient 
discharges for the hospital's most recently completed Medicare cost 
report filing period as the measure of patient volume for IME payments. 
The hospital should include all inpatient discharges from all parts of 
the hospital complex.
    If a children's hospital eligible to participate in the CHGME 
program has not completed a Medicare cost report period prior to 
submission of an application to the CHGME program, its patient volume 
will be calculated by the following methodology:
    a. Divide the number of inpatient discharges from the date the 
hospital became eligible to the CHGME application deadline by the 
number of days during this period to produce the average number of 
discharges per day.
    b. Multiply the average number of discharges per day by the number 
of days the hospital will provide inpatient

[[Page 37987]]

care as a hospital eligible to participate in the CHGME program during 
the FFY in which payments are being made.
    One respondent comment that accounting for discharges in the IME 
payment formula is unnecessary, since it is not a factor for Medicare, 
and that volume would be reflected by the number of residents in the 
interns and residents to bed (IRB) ratio.
    The Department disagrees with this comment. Since the Medicare IME 
adjustment is an increase in the PPS payment based on a single 
discharge, the number of discharges is a critical factor in determining 
how much IME adjustment a hospital receives from HCFA upon settlement 
of the cost report by Medicare. For the CHGME program, volume, as 
determined by the number of discharges, is one of the measures of 
resource utilization in the children's hospitals.
    The FTE resident count in the IRB ratio reflects teaching 
intensity, not patient volume. The Department assumes that the 
respondent believes that a hospital with more residents would see a 
larger volume of inpatients; however, since residents rotating through 
the outpatient parts of the hospital are included in the FTE resident 
count, a hospital could have few discharges and a large number of 
residents.

Outpatient Services

    Several respondents were in support of the Department's proposed 
development of a factor to indicate the resources associated with 
training in outpatient settings. They suggested that this factor 
include the development of a case mix index that is more reflective of 
the relative resource utilization experienced by children's hospitals 
in both an inpatient and outpatient setting. Other respondents were not 
in favor of the Department pursuing this avenue of investigation and 
encouraged the Department to rely on the work being done by HCFA in 
this area.
    Currently HCFA does not have an IME adjustment factor for the 
outpatient PPS; however, it is collecting data to determine if there is 
a need for such an adjustment. The CHGME program will consider HCFA's 
research in addition to pursuing the issue independently.

Determination of IME Payments

    Beginning in FFY 2001, the Department will use the following 
formula for calculating IME payments:

[GRAPHIC] [TIFF OMITTED] TN20JY01.005

    The following variables will be used in the formula to determine 
IME payments:
NoD = number of discharges for hospital
CMI = average case mix index for hospital
WI = area wage index for hospital
IME = IME adjustment/teaching intensity factor for hospital. 
Currently, the teaching intensity factor is: 
1.6((1+residentsi-to-bedsi 
ratio).405-1)
Zime = total dollars available for CHGME program IME 
payments
IME Pay = total IME payments to hospital
i = individual hospital
m = total number of hospitals participating in the CHGME program
residents = average number of unweighted FTE residents in the most 
recently completed cost reporting period and the prior two cost 
reporting periods with application of the cap.
beds = sum of all available beds, including beds and bassinets in 
the healthy newborn nursery, in the most recently completed cost 
report filing period, divided by the number of days in that period.
    This formula differs from that published in the March notice in 
that it omits the adjustment factor for hospitals with average lengths 
of stay greater than 30 days.

Hospitals With Average Length of Stay Greater Than 30 Days

    In the March notice, the Department proposed to apply an adjustment 
factor in the calculation of IME payments for children's hospitals with 
average lengths of stay greater than or equal to 30 days. These 
hospitals provide a variety of services, including rehabilitative 
services, that requires their patients to remain as inpatients for a 
prolonged period of time. The Department found that the FFY 2000 
formula for determining CHGME IME payments may have disadvantaged these 
hospitals.
    Since the length of stay is a major factor in determining the 
relative costliness of an inpatient stay, the Department proposed an 
adjustment factor based on the average length of stay (ALOS) to more 
adequately reflect the relative costliness of patients treated by the 
children's hospitals with significantly long lengths of stay. For 
hospitals with ALOS greater than or equal to 30 days, the adjustment 
factor proposed was the ALOS for the individual hospital divided by the 
average ALOS for all hospitals with ALOS less than 30 days.
    Several respondents supported the principle of adjusting the IME 
payments for those children's hospitals with average lengths of stay 
greater than or equal to 30 days as these hospitals are demonstrably 
different from all other children's hospitals. They noted that it is 
important that hospitals providing the types of services that require 
prolonged inpatient lengths of stay (e.g., rehabilitation) not be 
penalized for providing such services, as length of stay is a major 
factor in the relative costliness of an inpatient stay. However, the 
respondents indicated that the aggregate impact of an adjustment would 
be minimal, since it would involve only a very few small hospitals, and 
among them, they collectively train only a very few residents. These 
respondents recommended that HRSA make available the analysis 
underlying this particular adjustment and seek further comment before 
making the adjustment final and implementing it.
    The Department will postpone the implementation of an adjustment 
factor based on ALOS to the IME payment formula until it conducts 
additional analyses. These analyses and subsequent proposed 
recommendations related to the IME payment formula will be published in 
a future Federal Register notice.

Economic and Regulatory Impact

    Executive Order 12866 directs agencies to assess all costs and 
benefits of available regulatory alternatives and, when rulemaking is 
necessary, to select regulatory approaches that provide the greatest 
net benefits (including potential economic, environmental, public 
health, safety distributive and equity effects). In addition, under the 
Regulatory Flexibility Act (RFA of 1980), if a rule has a significant 
economic effect on a substantial number of small entities, the 
Secretary must specifically consider the economic effect of a rule on 
small entities and analyze regulatory options that could lessen the 
impact of the rule.
    Executive Order 12866 requires that all regulations reflect 
consideration of alternatives of costs, of benefits, of

[[Page 37988]]

incentives, of equity, and of available information. Regulations must 
meet certain standards, such as avoiding an unnecessary burden. 
Regulations which are ``significant'' because of cost, adverse effects 
on the economy, inconsistency with other agency actions, effects on the 
budget, or novel legal or policy issues, require special analysis.
    The Department has determined that the only burden this action will 
impose on children's hospitals is the resources required to submit an 
application to the CHGME program. Therefore, in accordance with the RFA 
and the Small Business Regulatory Enforcement Act of 1996, which 
amended the RFA, the Secretary certifies that this action will have a 
significant impact on a substantial number of small entities in that 
this action will provide significant funding to eligible children's 
hospitals. However, since this action will not impose a significant 
burden on a substantial number of small entities, we have not examined 
any alternatives for reducing the burden on children's hospitals. The 
Secretary has also determined that this action does not meet criteria 
for a major rule as defined by Executive Order 12866 and would have no 
major effect on the economy of Federal expenditures.
    We have determined that the proposed rule is not a ``major rule'' 
within the meaning of the statute providing for Congressional Review of 
Agency Rulemaking, 5 U.S.C. 801. Similarly, the proposed rule will not 
have effects on State, local and tribal governments and on the private 
sector such as to require consultation under the Unfunded Mandates 
Reform Act of 1995.
    Further, Executive Order 13132 establishes certain requirements 
that an agency must meet when it promulgates a rule that imposes 
substantial direct compliance costs on State and local governments, 
preempts State law, or otherwise has Federalism implications. We have 
reviewed this action under the threshold criteria of Executive Order 
13132, Federalism, and, therefore, have determined that this action 
would not have substantial direct effects on the rights, roles, and 
responsibilities of States.

Paperwork Reduction Act of 1995

    In accordance with section 3507(a) of the Paperwork Reduction Act 
(PRA) of 1995, the Department is required to solicit public comments, 
and receive final Office of Management and Budget (OMB) approval, on 
collections of information. As indicated, in order to implement the 
Children's Hospital Graduate Medical Education Payment Program (CHGME), 
certain information is required as set forth in this notice in order to 
determine eligibility for payment and amount of payment. In accordance 
with the PRA, we have received final OMB approval on our proposed 
collection of information (OMB No. 0915-0247).
    Collection of information: The Children's Hospitals Graduate 
Medical Education Payment Program.
    Description: Data is collected on the number of full-time 
equivalent residents in applicant children's hospital training programs 
to determine the amount of direct and indirect medical education 
payments to participating children's hospitals. Indirect medical 
education payments will also be derived from a formula that requires 
the reporting of case mix index information from participating 
children's hospitals. Hospitals will be requested to submit such 
information in an annual application.
    Description of Respondents: Children's hospitals operating approved 
graduate medical residency training programs.
    Estimated Annual Reporting: The estimated average annual reporting 
for this data collection is approximately 150 hours per hospital. The 
estimated annual burden is as follows:

----------------------------------------------------------------------------------------------------------------
                                                               Responses
                   Form name                      Number of       per         Total      Hours per    Total hour
                                                 respondents   respondent   responses     response      burden
----------------------------------------------------------------------------------------------------------------
HRSA-99-1
  (Annual).....................................           54            1           54         99.9        5,395
  (Reconciliation).............................           54            1           54            8          432
HRSA-99-2 (IME)................................           54            1           54           14          756
HRSA-99-4
  (Required GPRA tables).......................           54            1           54           28        1,512
                                                ----------------------------------------------------------------
    Total......................................           54            1           54  ...........        8,095
----------------------------------------------------------------------------------------------------------------

National Health Objectives for the Year 2000

    The Public Health Service is committed to achieving the health 
promotion and disease prevention objectives of Healthy People 2000, and 
its successor, Healthy People 2010. These are Department-led efforts to 
set priorities for national attention. The CHGME program is related to 
the priority area 1 (Access to Quality Health Services) in Healthy 
People 2010, which is available online at http://www.health.gov/healthypeople.

Education and Service Linkage

    As part of its long-range planning, HRSA will be targeting its 
efforts to strengthening linkages between Department education programs 
and programs which provide comprehensive primary care services to the 
underserved.

Smoke-Free Workplace

    The Depaertment strongly encourages all award recipients to provide 
a smoke-free workplace and promote abstinence from all tobacco 
products, and Public Law 103-227, the Pro-Children Act of 1994, 
prohibits smoking in certain facilities that receive Federal funds in 
which education, library, day care, health care, and early childhood 
development services are provided to children.
    This program is not subject to the Public Health Systems Reporting 
Requirements.

    Dated: June 7, 2001.
Elizabeth M. Duke,
Acting Administrator.
    Dated: July 17, 2001.
Tommy G. Thompson,
Secretary.
[FR Doc. 01-18166 Filed 7-19-01; 8:45 am]
BILLING CODE 4160-15-M