P4P_Eval - Part A_final_052909

Supporting Statement For Paperwork Reduction Act Submission – Part A

“Evaluation of Home Health Pay for Performance Demonstration”

A. Background

The Home Health Pay for Performance Demonstration is part of a movement by the Centers for Medicare & Medicaid Services toward performance-based purchasing for a variety of provider types. By providing financial incentives for achieving high levels of performance on standardized quality measures, CMS hopes to encourage health care providers to improve the quality of care provided to Medicare beneficiaries. Similar programs are planned or already underway for nursing homes, hospitals, physicians, integrated health systems, ESRD services, and chronic disease management programs. Other initiatives undertaken by CMS to improve quality of care under the home health benefit include public reporting of quality measures on the Home Health Compare web site and efforts by Quality Improvement Organizations (QIOs) to promote and provide technical assistance with outcome-based quality improvement activities of home health agencies. This multi-faceted program, known as the Home Health Quality Initiative, reflects an increased emphasis on the part of CMS to encourage providers to continually work to improve quality of care and enhance patient outcomes.

The current phase of the Home Health Pay for Performance Demonstration (HHP4PD) relies on the voluntary participation by home health agencies within several states, with random assignment of participating agencies to treatment or control groups within each state, where the control group will not be eligible for incentive payments. These two groups form the primary comparison for determining if the HHP4PD was effective in creating improved, targeted outcomes for patients served by home health agencies.

Identification of Volunteer (Control and Treatment) Home Health Agencies

Abt Associates, Inc., the demonstration implementation contractor, solicited volunteer home health agencies from seven states (Connecticut, Massachusetts, Alabama, Georgia, Tennessee, Illinois, and California) during the Fall 2007. Five hundred and seventy home health agencies volunteered to be part of the demonstration implementation. Based on the following stratification criteria (geographic region, urban/rural, agency size, agency control (nonprofit, proprietary, government), and agency type (freestanding and hospital-based)), these 570 home health agencies were randomly assigned by Abt Associates, Inc. to either the treatment (eligible for performance-based awards) or control (not eligible for performance-based awards) groups. All volunteer home health agencies were notified of their designation as either a treatment or control agency and agreed to participate in implementation- and evaluation-related activities related to the Demonstration.

Survey of Home Health Agency Practices

As part of the multifaceted evaluation protocol for the demonstration involving both primary (survey and focus group) and secondary (OBQI and Home Health Compare) data analyses, all treatment and control group providers will be asked to complete a survey questionnaire twice during the demonstration--once during year one and once during year two of the demonstration implementation. This document supports the year one request for the survey questionnaire. The year one survey questionnaire will be self-administered online as soon as feasible after demonstration implementation (January 1, 2008), to gather information about pre-demonstration organizational measures and practices as well as organizational and practice changes implemented as a consequence of demonstration participation that cannot be gathered using other existent CMS or publicly available data sets. The year two survey instrument will have a slightly different focus that the year one instrument and will necessitate a second approval cycle.

A hard copy of each of these surveys is contained in the Appendix to this document. Data elements collected include information about agency organization, clinical practices, and quality improvement activities, limited financial information, perceptions of the demonstration, and perceptions of the community environment. No personal health information will be collected as part of this survey. All data collected on the survey questions are focused on the home health agency as an entity.

Topic/Section	# Items (Treatment)	# Items (Control)
Demographic	3	3
Facility: staffing, care processes and quality improvement activities	9	9
Contextual/community factors	1	1
Impact of participation	3	0
Commitment to quality improvement and feedback on demonstration	3 (including 1 open-ended item)	3 (including 1 open-ended item)
Total number of items	19	16

Security of data collection

The self-administered online questionnaire (19 items for the treatment home health agencies and 16 items for the control home health agencies) will be hosted on a secure Web site at the University of Colorado. Information to access the secure Web site will be provided as part of the instruction sheet provided via a mailing to each of the 570 home health agencies (See Part A—Attachment 1). Additionally, access to the online instrument itself will be controlled by individual provider identification codes and passwords that will be provided to the volunteer home health agencies. The questionnaire information cannot be saved online by the provider. Once the survey questions are answered, the provider clicks on a "Submit" button and two actions occur. First, the data from the survey questions are automatically sent to a different secure server at the Division of Health Care Policy and Research at the University of Colorado (our organization) for data analysis. Second, the information on the secure University of Colorado Web site is re-set to its initial (neutral) settings.

B. Justification

1 . Need and Legal Basis

The need for this survey questionnaire is based on the Evaluation Plan that is part of the "Evaluation of Home Health Pay for Performance Demonstration", Health and Human Services, Centers for Medicare and Medicaid Services contract #HHSM-500-2005-00022I, Task Order #1, August 2007.

2. Information Users

The information collected will be used as part of the evaluation of the Home Health Pay for Performance Demonstration sponsored by the Centers for Medicare and Medicaid Services. This information will be at an aggregate level; no individually identifiable HHA data will be released to the public.

3. Use of Information Technology

The self-administered survey will be available on a secure University of Colorado Denver Web site. The beta version of the Web page is: https://hschealth.uchsc.edu/hcpr/loginP4P.asp. A hard copy of all survey items is included as the appendix to Part A: Attachment 2. This server is used for similar applications to exchange information in a secure environment with health care providers, state surveyors, and others. Additionally, as described in the preceding section on data security, the data from the home health agencies will be gathered using this secure Web site. Once the data are submitted via this secure Web site, the data will be transferred immediate (and the original information on the secure Web site deleted) to a separate, non-Web accessible, secure server controlled by the Division for Health Care Policy and Research at the University of Colorado Denver (Anschutz Medical Center).

4. Duplication of Efforts

None.

5. Small Businesses

The completion of either the treatment or control survey questionnaire is not likely to impose a larger burden on small HHAs than on larger sized HHAs. 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.

6. Less Frequent Collection

This is a one-time data collection during year one of the Home Health Pay of Performance Demonstration project. A request for the year two data collection--using a slightly different data collection form--may be made at a later date.

7. Special Circumstances

There are no special circumstances in the data collection.

8. Federal Register/Outside Consultation

A 60-day Federal Register notice was published on June 26, 2008. A 30-day Federal Register notice was published on September 12, 2008. No comments were received.

9. Payments/Gifts to Respondents

Data collection is part of the volunteer home health agency's commitment to participate in all aspects of the Home Health Pay for Performance Demonstration project. As is described in Part B, there are repeated and multi-modality contacts if the HHAs are not responsive to the initial request to complete the survey. A formal "Thank You" letter and Certificate of Completion will be sent to each home health agency that completes a survey questionnaire.

10. Confidentiality

There is no patient-level personal health information gathered as part of the survey questionnaire. Data will be treated in a confidential manner, unless otherwise compelled by law.

11. Sensitive Questions

There are no sensitive questions.

12. Burden Estimates (Hours & Wages)

Data collection for this project will be confined to one 2-month period (June and July 2009). Testing of "paper and pencil" versions of both the initial version of the survey submitted for OMB review in May 2008 and the current version of the survey (See Part A: Attachment 2), as well as the Web-based beta version were used to estimate the amount of time needed to complete the survey questions. The testing was conducted with three experienced home health agency nurses, all with some home health agency management experience. These individuals were asked to play the role of a "Director of Nursing" completing the survey for their agency. Average completion time for the longer (19 item) was 20 minutes, while average completion time for the shorter (16 item) survey questionnaire was 14 minutes. All participants stated that the survey questions could be completed more quickly in an electronic, online format than a paper and pencil format. Factoring in “overhead” time related to reading the original communication from the evaluation contractor, logging on to the Web site, and rare technical difficulties (e.g., logon errors), the estimated time to complete the form identified on PAPERWORK REDUCTION ACT SUBMISSION WORKSHEET (Form 83, Item #11) is 30 minutes per form (an average of 16 minutes for completion of the survey, plus 14 minutes of overhead time).

Cost per response based on Bureau of Labor Statistics mean hourly wage rate of $41.36 for a General and Operations Managers (assuming the respondent is either the administrator of the HHA or a director of professional services, plus 37.17% for benefits, for a total estimated employer cost per hour worked of $56.73. Given that the estimated time to complete a survey questionnaire for a participating home health agency is 30 minutes, the one time burden per response is $28.37 per home health agency. Annual hour burden was calculated based on 100% of the eligible (289 Control and 281 Treatment) home health agencies voluntarily completing the electronic survey. If 100% of the 570 eligible home health agencies voluntarily complete the online survey questionnaire, the total burden would be $8,084.03. This is an upper bound estimate of the total burden.

13. Capital Costs

This is a one-time data collection for research purposes that will not be repeated. No capital costs will accrue to respondents related to the collection of information for this survey.

14. Cost to Federal Government

The total cost of this data collection activity has been funded by the Centers for Medicare and Medicaid Services through the contract “Evaluation of the Home Health Pay for Performance Demonstration” with the Division of Health Care Policy and Research at the University of Colorado Denver (Anschutz Medical Center) (Contract Number HHSM 500-2005-00022I T.O.#1). The cost to design, develop, test, revise, complete cognitive burden testing, and edit the 19 items in the survey was estimated at $8,320. The costs to design, develop, and create the secure Web-based interface for the survey questionnaires (Tasks 2 & 3 of the contract) have been estimated to total $5,750.

15. Changes to Burden

Because this is a one-time only data collection for research purposes, the testing will not result in any annual reporting or recordkeeping costs or time burden.

16. Publication/Tabulation Dates

The Division of Health Care Policy and Research at the University of Colorado Denver (Anschutz Medical Center) was awarded the contract in August 2007. Work on the contract began at that time with the draft version of the survey questionnaire presented to the CMS Project Officer in February 2008. Analysis of the year one data from the survey questionnaire instrument will be integrated with other concurrent year one analyses in August 2009. Results of this data collection and analyses will be tabulated and included in a written report to CMS due October 2009.

The data analysis of the survey responses will consist primarily of descriptive statistics. The percentage of each response option selected for items 4 – 18 on the survey used by treatment home health agencies (HHAs) and 4 – 13, “17”, and “18” for the survey used by the control HHAs will be computed for each group. Simple tabular and/or graphical comparisons between treatment and control HHAs on equivalent items will be constructed. Written responses to item 19 on the treatment HHA survey and the equivalent of “19” on the control HHA survey will be compiled, grouped or categorized by similarity of response and reported by frequency (Pareto) in tabular form. The computation of inferential statistics using these data would not be appropriate given the non-random (self-selected) nature of the data (see response to “Part B #b” that follows).

Summaries of these data analyses will be included in the Final Project Report that will be available On-line at the CMS Web page. These data will not be used to construct a public use file for future analyses.

17. Expiration Date

The expiration date, along with the assigned OMB#, will be placed at the top of the Surveys. See Part A: Attachment 2, Appendices.

18. Certification Statement

There are no exceptions to the certification statement.

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Supporting Statement For Paperwork Reduction Act Submission – Part A Page 6

“Evaluation of the Home Health Pay for Performance Demonstration” 05/29/09