Supporting Statement PartA REVISED 11-19-2012

Supporting Statement – Part A
Surveys of Physicians and Home Health Agencies to Assess Access Issues for Specific
Medicare Beneficiaries as Defined in Section 3131(d) of the ACA
CMS-10429, OMB 0938-New
Background
The Centers for Medicare and Medicaid Services (CMS) has contracted with L&M
Policy Research (L&M) and its partners, Avalere Health (Avalere), Mathematica
Policy Research (MPR), and Social & Scientific Systems, Inc. (SSS) to support the
Agency in responding to provisions of the Patient Protection and Affordable Care Act
(ACA) Section 3131(d) and concerns that some subsets of the Medicare population
may have decreased access to home health services. This study is a follow-on study
to a Think Tank project on the same subject led by L&M to explore potential revisions
to the home health prospective payment system (HH PPS) that account for costs
related to severity of illness and access to care improvement. The proposed surveys
are one small facet of the large follow-on study that, taken together, may yield
recommendations for revisions to the HH PPS to ensure that home health agencies
(HHAs) are adequately reimbursed for providing services to vulnerable populations
as defined by the ACA – low income beneficiaries, those living in medically
underserved areas, and beneficiaries with high levels of severity of illness.
To supplement the larger quantitative analyses conducted as part of this follow-on
study, the research team plans to conduct two surveys. The surveys are designed to
be both explanatory and confirmatory in nature. The questions posed in the surveys
were informed by the research conducted during the Think Tank project including
multiple discussions with the Think Tank Technical Expert Panel, as well as
additional conversations with stakeholders, and are being asked in a survey because
the answers are not available in claims data or from other data sources. They will
help the research team confirm what we will learn through our analysis of claims, but
the questions will also help us identify and in some instances, confirm, characteristics
of beneficiaries who have been reported too difficult to place in home health care.
They will also help the research team explore whether there are any access issues
that we may not expect to detect through our analysis of claims data, which may then
result in recommendations for further study. To learn more about the beneficiaries
who experience access issues, respondent specific surveys have been developed to
administer to (1) physicians who refer vulnerable patients to the Medicare home
health benefit, and (2) Medicare certified home health agencies. Both target
populations offer unique perspectives on the characteristics of Medicare beneficiaries
who may have access issues and can help answer questions that cannot be
addressed as well through other research methods (primarily through analysis of
administrative claims).

1

A. Justification
1 . Need and Legal Basis

This data collection is part of a larger study called for under section 3131(d) of the
Patient Protection and Affordable Care Act (ACA). The larger study is focused on two
major issues (1) supporting CMS’ efforts to improve payment accuracy and (2)
understanding issues of access for the ACA populations under the existing home
health prospective payment system.
The larger quantitative portion of the project aims to understand payment accuracy
for the specific study populations through claims and cost data analyses. Regression
analyses on home health claims data, patient-level OASIS data on functional and
clinical status, Medicare beneficiary eligibility status and characteristics, and agency
level cost reports will inform potential HH PPS revisions. The samples for these
analyses are large—hundreds of variables across millions of episodes will inform any
potential HH PPS revision. The quantitative analyses serve as the foundation for any
potential policy recommendation.
Responses to the survey instruments will supplement the quantitative analyses by
identifying access issues for the ACA defined populations and the extent to which
further study is necessary. These surveys will help the study team better understand
the characteristics of Medicare beneficiaries who are not able to gain access to or
have experienced delays in gaining access to home health services.
2.

Information Users
As a new collection, the information collected is expected to support CMS’ efforts to
improve the HH PPS payment accuracy for vulnerable populations and thereby
ensure the payment system does not inadvertently cause avoidable access
problems. The questions are designed to provide insights into access issues for
vulnerable populations that cannot be learned through analyses of administrative
data.

3.

Use of Information Technology
The research team is proposing a multi-mode data collection, with a selfadministered paper survey mailed to participants, along with telephone prompting for
participants who fail to complete and return the mailed questionnaire within the
designated time period. During the phone prompt, interviewers will encourage
participants to return the questionnaire by mail or fax and will offer to complete the
survey over the phone.
Additionally, subsequent follow-up mailings (up to two) will be made using USPS
2

Priority mail services. Reminder postcards will be sent to those participants who have
not responded or who have misplaced or lost their packets, followed by a second
packet and, if necessary, a third packet. The survey instrument itself has been kept
brief has been formatted and will be printed as to minimize respondent burden. We
will also provide options for submitting the questionnaire via mail, fax, or over the
telephone if requested.
The research team will not make the tool available electronically in order to keep the
resources used to field the survey to a minimum. Based on past experiences, the
research team has found it more productive and cost effective to collect information
from providers in a paper survey than in an electronic format when available
resources are limited to using one primary means of data collection. This collection
method is supplemented with telephone prompting and assistance to those otherwise
unable to complete the questionnaire. The survey does not require a signature from
participants.
4.

Duplication of Efforts
The research team conducted an extensive literature review to understand access
issues for the vulnerable populations defined in the ACA. During the literature review,
the team found that no surveys focused on access issues for the specific study
populations. Some research has been completed on the topic of access to home
health services for Medicare beneficiaries, but it has focused on a limited number of
stakeholder interviews, which were completed in early to mid 2000. Due to both the
age and the research methods of the previous research, the proposed new collection
is not redundant, as these surveys focus on specific target populations and issues
related to their access to home health services. Further, both surveys will be
conducted on larger samples than has been done previously.

5.

Small Businesses
The completion of the survey instruments is not likely to impose a larger burden on
small entities (HHAs or physician practices) than on larger sized entities (HHAs or
physician practices). There may be a smaller absolute burden for a smaller
organization, due to the fact that the administrator may be able to complete the
survey entirely from recall, while in a larger organization the administrator may not
have all of the information as readily at hand.
In order to estimate the number of small entities included in the sample, the research
team conducted a review of the literature. Per the Small Business Administration’s
Office of Advocacy “most hospitals and most other providers and suppliers are small
entities, either by nonprofit status or by having revenues of less than $7.0 million to

3

$34.5 million in any given year.” 1 For purposes of the small entities analysis, the
research team is assuming that 90% of the HHAs in the survey sample, which
equates to 540 HHAs, are small entities due to the following statement from the
Office of Advocacy, “90 percent or more of the health care providers [HHAs and
Hospices] meet the SBA’s size standards as measured either by their annual receipts
or nonprofit status.” 2 The research team is also assuming that 78% of the physician
practices included in the survey sample, which equates to 215 physician practices,
are small businesses based on a statement made by the American Medical
Association to the House Committee on Small Business Subcommittee on
Contracting and Workforce.3
6.

Less Frequent Collection
The focus of the instruments is to ask questions that the team cannot collect through
data analyses. By forgoing this data collection, CMS would lose the opportunity to
gather insights into the patient, caregiver and agency characteristics associated with
those beneficiaries who have been denied or experienced delayed access to home
health services. Without this information, CMS’ payment accuracy activities will only
be focused on claims and cost related analyses using administrative data. While
administrative data may improve our understanding of the characteristics of patients
who actually received home health services, it can not be used to understand the
beneficiary, caregiver and agency characteristics associated for those whose access
is denied, delayed or not fully provided consistent with the services ordered.

7.

Special Circumstances
This information collection will not involve any of the special circumstances.

8.

Federal Register/Outside Consultation
OSORA PRA staff will prepare any required Federal Register announcements for
publication.
Throughout the development period of the survey instruments, the research team has
consulted with members of the technical expert panel (TEP) established under a
recently completed project addressing the Section 3131(d) mandate. The TEP was
convened to provide expertise regarding the home health industry and input into how

1 2010. “ Medicare Program: Home Health Prospective Payment System Rate Update for Calendar Year 2011; Changes in
Certification Requirements for Home Health Agencies and Hospices (RIN: 0938-AP88)”. Small Business Administration Office of
Advocacy. Retrieved from: http://archive.sba.gov/advo/laws/comments/hhs10_0914.html#5
2 2010. “ Medicare Program: Home Health Prospective Payment System Rate Update for Calendar Year 2011; Changes in
Certification Requirements for Home Health Agencies and Hospices (RIN: 0938-AP88)”. Small Business Administration Office of
Advocacy. Retrieved from: http://archive.sba.gov/advo/laws/comments/hhs10_0914.html#5
3 2011. “Defer No More: The Need to Repeal the 3% Withholding Provision.” Retrieved from: http://www.amaassn.org/resources/doc/washington/three-percent-withhold-written-comments.pdf

4

best to identify and measure home health access issues. The TEP members
represented HHAs, national home health care associations, state and federal
agencies, consumer advocacy organizations, home health physicians, and home
health research experts. During the TEP meetings, TEP members recommended the
research team pursue surveys in order to better understand issues of access for the
ACA populations, since analyses of administrative data would not be able to provide
insights into some of the more nuanced issues. Since these meetings, the research
team has consulted with a few of the individual members and a few of their
associates to obtain more input on the type of collection, availability of data, and the
types of data elements that could be collected using a survey.
9.

Payments/Gifts to Respondents
Incentives have been shown to increase response rates in mail surveys, and prepaid
incentives tend to yield higher response rates than incentives that are promised
(Singer et al., 1999). Based on our experience conducting provider surveys, we
expect that the $50 incentive will be cost-effective by saving resources that would
have been needed for additional fieldwork. The purpose of the payment is to signal
respect for the physician’s time and to establish trust, which has been shown to result
in higher response rates. We propose to include a prepaid incentive of $50 in the
initial mail packets to physicians. Because of the potential problems in providing an
incentive payment to an employee, we will not offer incentives to HHAs.

10. Confidentiality
Information will be kept secure to the extent permitted by law.
11. Sensitive Questions
This collection does not contain any sensitive questions.
12. Burden Estimates (Hours & Wages)
Table A provides an estimate of time burden for the data collection activities for
which approval is being sought. The total average burden hours for which we are
seeking approval in this package is 218.75 hours. Both the Physician Survey and
the Home Health Survey will be primarily self-administered through a mail-distributed
paper survey. Interviewers will conduct a follow up with non-respondents by
telephone to prompt completion of the survey and will provide an opportunity to
complete the paper survey over the phone. We estimate that no more than 5 percent
of total completed cases will complete the survey in this manner. The surveys are
estimated to take each respondent no more than 15 minutes to complete. A total of
275 physicians and 600 home health administrators are expected to complete the
survey. According to the Employment and Wages May 2010 national estimates from
the Occupational Employment Statistics (OES) survey, the mean hourly wage of
5

general internists and family and general practitioners is $86 and the mean hourly
wage for medical and health services managers working in the home health care
services industry is $41.
TABLE A. AVERAGE BURDEN TO RESPONDENTS IN HOURS

Data Collection Activities

Number of
Respondents

Physician Survey
Mail Surveys
Phone Surveys
Home Health Survey
Mail Surveys
Phone Surveys
Estimated Total for Both
Surveys
ANNUAL ESTIMATES

Average
Burden
Hours/
Respondent

275
261

.25
.25

14
600

Total
Average
Burden
Hours

Average
Hourly
Rate

Estimated
Monetary Cost
Burden To
Respondents

68.75
65.25

$86
$86

$5,912.50
$5,611.50

.25

3.5

$86

$301.00

.25

150

$41

$6,150.00

570

.25

142.5

$41

$5,842.50

30

.25

7.5

$41

$307.50

875

.25

218.75

$55

$12,062.50

291.7

.25

72.9

$55

$4,020.83

13. Capital Costs
No capital costs will accrue to respondents.
14. Cost to Federal Government
$445,527 is the total estimated cost to the Federal Government, allocated across two
years to include design, field testing and analysis of the finding. Annually, the costs
are estimated to be for Year 1: $214,844; for Year 2: $230,683, with no additional
costs thereafter.
15. Changes to Burden
This is a new collection.
16. Publication/Tabulation Dates
The purpose of the survey instruments is to allow the research team to analyze
potential problems of home health referral, placement, and access issues confronting
6

referring physicians as well as beneficiaries. In order to address this central research
question, the team will perform several simple univariate and bivariate analyses
including descriptive statistics to summarize pertinent variables regarding the use of
home health services, as well as referrals, placement, and access issues. Frequency
counts and cross-tabulations will also be used to show distributions of physicians’
and HHAs’ responses regarding problems facing home health referrals, placement,
and access issues. From this process we will fill several data shells showing the
distributions of these characteristics.
We will utilize questions on providers’ and HHAs’ perceptions of access to home
health services to perform subgroup analyses. We anticipate being able to make
comparisons between two subgroups of interest, though the ability to detect
differences will depend on a number of factors including the sample sizes for each of
the two subgroups and where the estimate is in the distribution. The ability to make
these comparisons will also depend on actual sample yield and will not be made for
more than two subgroups at a time. Table B below shows the main comparisons
likely to be made for the HHAs. If we are comparing two subgroups—for example,
with 200 HHAs serving ACA populations and 400 other HHAs—we will be able to
report that a difference of 9 to 12 percentage points is statistically different.
TABLE B. POSSIBLE COMPARISON GROUPS FOR SURVEY OF HOME HEALTH AGENCIES
Comparison

Location of HHA—Rural vs.
Urban
Ownership—Proprietary vs.
Voluntary/Non-profit/Gov’t
Population served: Primarily ACA
populations vs. Others
Size, no. episodes or revenue—
greater than or less than median

Anticipated
sample size

Detectable difference at 80% power, in
percentage points
True proportions less than
True proportions
20% or greater than 80%
approximately 50%

125 vs. 475

10

14

435 vs. 165

9

13

200 vs. 400

9

12

300 vs. 300

8

12

The physician survey data will be used to make univariate estimates only. The
sample of physicians will yield an estimate that is plus or minus 4 to 6 percentage
points at the .05 level of significance. Due to the relatively small sample sizes, no
subgroup comparisons, or itemetric analyses, are planned.
Results of summary statistics of pertinent variables will be presented in tables. For
example, tabulations of results may include:
 Distribution of providers (HHA/physicians) in their overall assessment of the
current availability of home health care services to Medicare beneficiaries in
their locality;
7

 Average ratings of issues related to access delays in the different domains,
that is, “Medical”, “Non-medical”, “factors related to the provider” issues.
 Average ratings of issues related to home health placement in the different
domains, that is, “Medical”, “Non-medical”, “factors related to the provider”
issues.
Table C below provides an example of such a table.
TABLE C. EXAMPLE DATA SHELL

HHA/PHYSICIAN SURVEY

a.

Respondents
Reporting 1-3
(Relatively
Unimportant)

Issue related to home health agency
Nursing staff with needed skill set not
available

b.

Therapy staff not available (e.g., PT, OT,
ST)

c.

Staff not experienced with medical
conditions(s)

d.

Required equipment/supplies not available

e.

Reimbursement not sufficient
Medical issue related to patient

f.

Severity/complexity of patient's medical
condition

g.

More than two 60 day periods (episodes) of
care expected

h.

Two or more visits per day expected

i.

Routine evening or weekend care expected

j.

Patient does not qualify for Medicare home
health benefit (e.g., not homebound)
Non-medical issue related to patient

k.

Patient living conditions or local area unsafe

l.

Patient located in hard-to-reach area or
travel distance/time too great

m.

Patient/family/caregiver cannot be or is
unwilling to be trained

n.

Family/caregiver is unable to provide
necessary support

o.

Language barrier/communication problems

p.

Patient or family refused services

8

Respondents
Reporting 4
(Neutral)

Respondents
Reporting 5 - 7
(Relatively Important)

Average
Respondent
Rating

Other Analysis:
The research team will use chi-squared analysis to test whether or not differences
exist between subgroups of HHAs’ responses. For example, we can use chi-squared
tests for pertinent variables measuring access to home health. A statistically
significant difference in response categories, that is, a p-value less than 0.05, will be
interpreted to indicate that the differences in HHAs’ responses are systematic rather
than just being due to chance.
Finally, the team may estimate one or more basic regression models to identify
correlates of outcome measures that are pertinent to home health access issues. For
example, we may want to explain variations in the mean number of how many times
a physician experiences delays in finding a home health agency willing and able to
admit Medicare fee-for-service patients as a function of number of physicians in the
admitting facility, payer mix, and other relevant explanatory variables. We may also
estimate another model to explain the probability of finding a placement for potential
home health patients as a function of the explanatory variables in the model just
mentioned.
The regression specification below gives a general sense of what these analyses
may entail.
ε

Y

This represents a set of dependent variables measuring access issues. Examples include mean
number of physician delays (OLS or count regression model) or whether or not a beneficiary finds a
placement for home health (logistic regression model).

X

This is a vector of issues related to the home health agency.

M

This is a vector of medical issues related to the patient.

P

This is a vector of non-medical issues related to the patient.

The coefficients of determination (r2) from such models give an indication of the
proportion of variations in the outcome variables being explained by explanatory
variables included in the models. It should be noted that the regression analyses, if
performed, will be only exploratory in nature.
Timelines:
Depending on OMB approval, the collection is expected to start no later than January
2013. The surveys will be in the field for approximately four months. The research
team will share the analysis and findings with CMS as part of the analysis report due
no later than July 15, 2013. The findings will also be included in the report to
Congress called for under section 3131(d) of the ACA. This report must be submitted
to Congress no later than March 1, 2014, requiring the research team to submit the
9

materials to CMS no later than August 31, 2013.
17. Expiration Date
CMS would like to display the expiration date.
18. Certification Statement
CMS does not request any exemptions from the certification statement.

10