HOS Highlights: Functional Impairment Levels in PACE Enrollees

Introduction

A variant of the Health Outcomes Survey, the Medicare Health Outcomes Survey-Modified (HOS-M) is a brief mixed mode annual survey (mail with telephone followup) specifically designed to collect functional status information from enrollees in the Program of All Inclusive Care for the Elderly (PACE) organizations or in the dually eligible beneficiaries demonstrations. CMS uses the HOS-M results to adjust Medicare capitation rates paid to these plans, taking the frailty of the enrollees into account. The frailty adjuster is applied in conjunction with the CMS-Hierarchical Coexisting Condition model and is based on activities of daily living (ADLs) limitations. After a brief description of the HOS-M and the PACE program, we present the results from the 2006 HOS-M to describe the health status of the PACE enrollees that participated in the survey.

Description

HOS-M

Previously called the PACE Health Survey, the HOS-M was fielded nationally for the first time in 2003. The protocol differs from that of the HOS in several ways. These differences are designed to lead to a high response rate despite the frailty of the target population, so that payment adjustments are as accurate as possible. The instrument has just 17 items, with ADLs as the core items used for calculating the frailty factor. Proxy respondents are encouraged and represent a substantial proportion of the respondents. Smaller plans provide detailed contact information for their enrollees and for family members to use in case a proxy is needed. The instrument is available in English, Spanish, and Chinese (Walsh et al., 2003). Each year, RTI International draws the sample, works with smaller plans to develop their contact data files, and analyzes the resulting data. DataStat, the National Committee for Quality Assurance-certified sole HOS-M vendor, fields the survey each year under contract to each of the participating plans.

To participate in the HOS-M, PACE plans have to be operational for 1 year. Because the HOS-M is used to adjust Medicare payments for specific categories of enrollees, PACE individual enrollees are included in the HOS-M sample if they are age 55 or over, reside in the community, and do not have end stage renal disease; they also have to be enrolled in their respective plans for at least 1 full month. Presented here are the results for PACE enrollees only, although the HOS-M is also fielded to enrollees in the dually eligible demonstrations.

Pace

PACE integrates long-term and chronic care with the standard Medicare acute care benefit package. Eligibility is based on State-by-State determinations of NHC status, i.e., whether the applicant meets his or her State's criteria for nursing home admission under Medicaid. Much of the care is provided through day centers operated by PACE staff comprised of physicians, nurses, social workers, rehabilitation therapists, and other ancillary providers, as well as inhome services. A small proportion of PACE enrollees become residents of nursing facilities when they can no longer be supported adequately in other settings (these enrollees are not included in the HOS-M). Most PACE enrollees are dually eligible Medicare/Medicaid beneficiaries for whom PACE organizations receive monthly capitation payments from both Medicare and Medicaid.

On January 1, 2006, there were 32 PACE organizations nationwide serving 10,867 Medicare beneficiaries.¹

Results

In 2006, 6,247 of the 8,649 PACE eligibles responded to the HOS-M, yielding an overall response rate of 72.2 percent for PACE plan enrollees. This response rate is similar to those for previous years. Thirty-two PACE plans participated in 2006 HOS-M. The entire eligible enrollment was included in the HOS-M sample for plans enrolling less than 1,400 beneficiaries (31 PACE plan) while we drew a sample of 1,000 eligible beneficiaries from the Medicare files for 1 PACE plan with larger enrollment.

Figure 1. Demographic Characteristics of the Medicare Health Outcomes Survey-Modified (HOS-M) Respondents: Mean Age by Sex.

• Program of All Inclusive Care for the Elderly (PACE) program enrolls Medicare beneficiaries age 55 or over.
• HOS-M respondents' demographic information was obtained from the Medicare Enrollment Database as it is not collected by the HOS-M to preserve the instrument's brevity.
• PACE enrollees represent an older segment of the Medicare population, with an average age of 80 for females and 76.6 for males.

Figure 2. Mean Unadjusted VR-12 Physical Component and Mental Component Summary Scores (PCS and MCS).

• The average PCS score of the Program of All Inclusive Care for the Elderly (PACE) enrollees, all of whom have nursing home level of need, is 28.9, about 20 points lower than for the general population, signifying a subpopulation in relatively poor physical health.
• The MCS scores for PACE enrollees average around 44, signifying that PACE enrollees report better mental health than physical health; the scores are only 6 points lower than those of the general population.

Figure 3. Functional Status: Levels of Activities of Daily Living (ADL) Impairment.

• ADL impairment is defined as being unable to do or having difficulty in the activity. Six ADLs are included: bathing, dressing, walking, transferring, toileting, and eating.
• Program of All Inclusive Care for the Elderly enrollees reported high levels of frailty: only 12.2 percent reported no ADL limitations, and over 40 percent reported maximum levels of frailty (5 to 6 ADL impairments).

Figure 4. Individual Activity of Daily Living (ADL) Impairments.

• Program of All Inclusive Care for the Elderly (PACE) enrollees have high levels of impairment on each individual ADL. Eighty percent reported having difficulty with walking or being unable to walk.
• Medicare Health Outcomes Survey-Modified (HOS-M) respondents have the most difficulty walking, followed by bathing and transferring.
• Although a smaller proportion of the HOS-M respondents have trouble with eating and toileting, the rates of impairment in these tasks are substantial.

Figure 5. Proxy Status of the Medicare Health Outcomes Survey-Modified (HOS-M) Respondents.

• Due to high levels of frailty, cognitive impairment, and other factors that may preclude HOS-M participants from responding to the survey, proxy respondents are allowed. The HOS-M protocol permits family members, friends, or paid caregivers, including program staff, to serve as proxies.
• Most of the HOS-M participants requested help from a family member or friend, and a small proportion used a health professional as proxy.
• Even though the HOS-M survey takes only a few minutes to complete, only 26.6 percent of Program of All Inclusive Care for the Elderly (PACE) respondents were able to do so independently. Over 45 percent received help from a family member or friend, and 16.5 percent relied on the help of a health professional or other caregiver.

Figure 6. Rates of Memory Impairment and Incontinence for the Medicare Health Outcomes Survey-Modified (HOS-M) Respondents.

• HOS-M respondents or their proxies were asked about memory loss that interferes with daily activities and about incontinence on a daily basis.
• The rates of memory loss are substantial, with 52.2 percent reporting memory loss that interferes with daily activities.
• Program of All Inclusive Care for the Elderly respondents also reported high levels of incontinence with 36.6 percent having daily urination difficulties or a catheter.