Report Evaluation of the San Francisco Support at Home Program: Year 1 Report by Laura Wagner, Jacqueline Miller, Susan Chapman, Timothy Bates & Joanne Spetz Philip R. Lee Institute for Health Policy Studies University of California San Francisco September 5, 2018 Abstract The purpose of this report is to outline the preliminary findings from the first year of a two year pilot for the San Francisco Support at Home Program (S@H). This evaluation has two purposes: (1) Support continuous quality improvement of the S@H program through ongoing rapid data collection and analysis, and (2) Assess the overall efficacy of the program in maintaining residence at home, reducing hospitalizations and emergency department visits, controlling costs, and supporting a high quality of life. This report summarizes the context of the program and the importance within the current landscape within San Francisco. Next, it describes the Support at Home enrollees and their characteristics using a series of cross-tabulations to make comparisons between key variables. Third, it presents early data regarding the impact of the program on outcomes such as emergency department visits, including information gathered from focus groups held with family of program participants that illuminate how the program is affecting families. Finally, next steps for the evaluation as well as interim considerations are provided. 1 Support At Home – Year 1 Evaluation Report 2 Acknowledgements This report was funded by the San Francisco Department of Aging and Adult Services (DAAS). This report was developed by the Philip R. Lee Institute for Health Policy Studies, (IHPS) at the University of California, San Francisco, in collaboration with DAAS and the Institute on Aging (IOA). DAAS, IOA, and IHPS wish to thank all those who contributed to this pilot program, including but not limited to those listed below (in alphabetical order), and all the community members and service providers who participated in the research and planning efforts by attending the community forum, focus groups, and/or completing surveys. Your support has been invaluable in the development and implementation of the Support at Home Program Pilot in efforts to serve San Francisco’s communities of older adults and people with disabilities.  The Arc San Francisco  Bay Area Care Council  Bayview Senior Services  California Domestic Workers Coalition  Caring Across Generations  Changing One Mind At A Time (COMAAT)  Community Alliance of Disability Advocates  Community Living Campaign  Community Living Policy Center at UCSF  Creative Explored  Family Caregiver Alliance  Felton Institute, Senior Division  Gray Panthers  Hand in Hand: The Domestic Employers Network  Homebridge  IHSS Public Authority  Independent Living Resource Center-SF  Institute on Aging  Jobs with Justice  La Colectiva  Mujeres Unidas y Activas  National Domestic Workers Alliance  S@H Advisory Committee  San Francisco Long Term Care Coordinating Council  San Francisco Board of Supervisors  San Francisco Aging and Adult Services Commission  Senior and Disability Action Support At Home – Year 1 Evaluation Report 3 Contents Abstract .....................................................................................................................................................................1 Acknowledgements ..................................................................................................................................................2 Chapter 1 – Introduction ..........................................................................................................................................4 Logic model ................................................................................................................................................................4 Chapter 2 – Support at Home Enrollees .................................................................................................................6 Demographics .............................................................................................................................................................6 Care needs and financial needs .................................................................................................................................8 Pre-enrollment and post-enrollment home care services ...........................................................................................9 Quality of life among Support at Home enrollees .................................................................................................... 11 Chapter 3 – Preliminary Results on the Impact of Support at Home ............................................................... 14 Preliminary data from initial and quarterly assessments ......................................................................................... 14 Perspectives of family and friends of Support at Home enrollees ........................................................................... 16 Chapter 4 – Next Steps for Evaluation ................................................................................................................ 19 Chapter 5 – Considerations for Year 2 ................................................................................................................ 20 Appendix: Data and methods for describing the targeted population ............................................................ 22 Support At Home – Year 1 Evaluation Report 4 Chapter 1 – Introduction The purpose of this project is to conduct a two-year formative and summative evaluation of the San Francisco Support at Home (S@H) program. The Support at Home program provides financial support for the purchase of home care services by adults with disabilities and older adults living in San Francisco. The eligible population is comprised of those who have a demonstrated need of assistance with two or more activities of daily living (ADLs) or instrumental activities of daily living (IADLs), income up to 100% area median income in San Francisco, assets up to $40,000 (excluding one house and one car), a demonstrated need for financial assistance paying for home care, and who agree to pay a copayment towards the purchase of additional home care services and participate in program evaluations. Anticipated enrollment is 175 to 250 individuals over a 2-year period. The original program plan was that half of enrollees would be aged 60 years and older, and half would be under 60 years old. Enrollees are required to contribute copayments for home care services prior to availability of the voucher, with the copayment rate based on the enrollee’s financial need demonstrated by monthly income. Those with low financial need pay 50% of the voucher amount towards home care services, those with medium financial need pay 33% of the voucher amount towards home care services, and those with high financial need pay 20% of the voucher amount towards home care services. Voucher values are based on the level of functional need demonstrated by the enrollee, which is determined by an assessment of the individual’s limitations in 17 ADLs and IADLs ranging from independent through dependent/paramedical levels of need. For enrollees with low functional need, a $346 voucher per month is available to apply to scheduled home care services, medium functional need, a $693 voucher per month is available to apply to scheduled home care services, and high functional need, a $1299 voucher per month is available to apply to scheduled home care services. Enrollees can elect to purchase home care services directly from an independent provider at a wage of $15 per hour, in which case enrollees are required to pay their home caregiver through an approved payroll service bi-weekly, or to purchase services monthly in advance through an approved home care agency at an hourly cost determined by the agency. The total hours of service received per week are determined by each Enrollee’s choices regarding provider and scheduling of home care services. The Support at Home program is administered by the Institute on Aging (IOA) via a contract from the San Francisco Department of Aging and Adult Services (DAAS). The University of California San Francisco is conducting an independent evaluation of the program via a contract from DAAS. This evaluation has two purposes: (1) Support continuous quality improvement of the Support at Home program through ongoing rapid data collection and analysis, and (2) Assess the overall efficacy of the program in maintaining residence at home, reducing hospitalizations and emergency department visits, controlling costs, and supporting a high quality of life. The evaluation is using a mixed-methods approach, incorporating qualitative, survey, and quantitative data. In order to assess the unique impact of the Support at Home program, the evaluation intends to compare the data from S@H enrollees with a comparison group of individuals who applied for S@H services but do not receive them because they did not meet income or asset eligibility requirements or chose to not enroll for any reason. Logic model The logic model underpinning the evaluation is presented in Figure 1. The S@H contractor, Institute on Aging, has engaged in outreach activities to identify potentially qualified applicants for the program. From that pool, Support At Home – Year 1 Evaluation Report 5 qualified applicants who choose to enroll in the program receive vouchers for home care services that support the purchase of domestic, non-medical personal and accompaniment services from home care agencies and individuals. These services provide enrollees with support to accomplish ADLs and IADLs, receive recommended health care services, and enhance social engagement. As a result, the services are expected to lead to higher self-reported quality of life,1 lower risk of nursing home admission or other residential changes, 2,3 and lower risk of hospitalization and emergency department visits.4 Greater availability of home care services should benefit family members as well, enabling them to maintain or increase their employment and reduce their personal spending on home care. Cost savings overall would be derived from reduced hospitalizations, emergency department visits, and nursing home admissions.5,6,7,8 Figure 1. Logic model for Support at Home program evaluation Outreach activities Applicants Qualified applicants receiving vouchers Domestic services (e.g., Non-medical personal grocery shopping, meal services (e.g., bowel preparation, cleaning, Accompaniment services care, bathing, dressing, laundry) transfers) Support to accomplish IADLs Social engagement Receipt of recommended Support to accomplish ADLs health care services Higher self-reported quality of life and lower reported depressive symptoms Lower risk of residential changes Lower risk of hospitalization or emergency department visit Lower family burden Higher employment & satisfaction of family members Lower net costs 1 Low L-F, Yap M, Brodaty H. (2011) A systematic review of different models of home and community care services for older persons. BMC Health Services Research 11:9. 2 Kane RL, Lum TY, Kane RA, Homyak P, Parashuram S, Wysocki A. (2013) Does home- and community-based care affect nursing home use? J Aging Soc Policy 25(2):146-60. 3 Thomas KS, Keohane L, Mor V. (2014) Local Medicaid home- and community-based services spending and nursing home admissions of younger adults. Am J Public Health 104(11):e15-7. 4 Xu H, Weiner M, et al. (2010) Volume of home- and community-based Medicaid waiver services and risk of hospital admissions. J Am Geriatr Society 58(1):109-115. 5 Segelman M, Intrator O, Li Y, Mukamel D, Veazie P, Temkin-Greener H. (2017) HCBS Spending and Nursing Home Admissions for 1915(c) Waiver Enrollees. J Aging Soc Policy. 2017 Apr 17:1-18. 6 Mitchell G 2nd, Salmon JR, Polivka L, Soberon-Ferrer H. (2006) The relative benefits and cost of Medicaid home- and community-based services in Florida. Gerontologist 46(4):483-94. 7 Harrington C, Ng T, Kitchener M. (2011) Do Medicaid home and community based service waivers save money? Home Health Care Serv Q. 30(4):198-213. 8 Newcomer RJ, Ko M, et al. (2016) Health Care Expenditures After Initiating Long-term Services and Supports in the Community Versus in a Nursing Facility. Medical Care 54(3):221-8. Support At Home – Year 1 Evaluation Report 6 Chapter 2 – Support at Home Enrollees Demographics Data from IOA enrollment records were analyzed to describe the population of S@H enrollees. Data from the American Community Survey, which is conducted by the U.S. Census Bureau, were analyzed to describe the population of individuals who might be qualified for the program in San Francisco (see the Appendix for details regarding these data). Data from a UCSF-administered survey on quality of life also were analyzed. The data received by UCSF included 105 people enrolled in Support at Home as of May 11, 2018. The American Community Survey indicates there are approximately 27,940 individuals with disabilities in San Francisco who meet the income requirements for Support at Home. Table 1 provides information about the specific disabilities of the population of people in San Francisco estimated as potentially eligible for Support at Home. The conditions are not mutually exclusive, and thus total more than the eligible population. The most common condition is ambulatory difficulty (62.5%), followed by independent living difficulty (46.8%) and cognitive difficulty (41.5%). Nearly one-quarter reported self-care difficulty (22.5%) and 6.8% reported a VA service-connected disability. Table 1. Disability conditions of the eligible population DESCRIPTION* NUMBER PERCENT Ambulatory difficulty 17,471 62.5% Independent living difficulty 13,073 46.8% Cognitive difficulty 11,609 41.5% Self-care difficulty 6,297 22.5% VA service-connected disability rating 1,901 6.8% *These conditions are not mutually exclusive. Total eligible population=27,940. As seen in Table 2, S@H enrollees are notably older than the estimated eligible population in San Francisco. There are about 10,000 people under 60 years old who are likely to be eligible for the program, accounting for 36% of the total eligible population. However, only 11.4% of enrollees are under 60 years old. Table 2. Age distribution of Support at Home program enrollees and eligible population S@H ELIGIBLE AGE CATEGORY ENROLLEES POPULATION 18-59 years 11.4% 36.1% 60-79 years 46.6% 32.9% 80 years & older 42.0% 31.0% Total 100% 100% There are a number of reasons why people under 60 years old may be under-represented among S@H enrollees. More than half of eligible people in this age group have a cognitive disability and no physical disability (as reported in the American Community Survey) and may not realize that programs such as Support at Home offers services that would be of benefit to them. In addition, many of those in the eligible population under 60 years old are employed, and two-thirds live with other people. Although these individuals may benefit from enrollment in Support At Home – Year 1 Evaluation Report 7 Support at Home, they may perceive that they do not have additional care needs because they are managing their employment effectively and have other household members who support them. Table 3 presents the racial/ethnic composition of the S@H enrollee population and the eligible population; note that differences between the enrollee and eligible population in the percentages for the 18-59 year age group appear large due to the small number of enrollees in this group. Among those 18-59 years old, Latinos are under- represented among enrollees (16.7% vs. 27.1%), and Asians are slightly under-represented (16.7% vs. 20.9%). Blacks/African-Americans are slightly over-represented (16.7% vs. 11%), as are those of other race/ethnicity (16.7% vs. 6.1%). Among those 60 years and older, Blacks/African-Americans are over-represented among enrollees (32.3% vs. 9.6%), while Latinos and Asians are under-represented (7.5% vs. 11.1% for Latinos, and 11.8% vs. 36.4% for Asians). Table 3. Race/ethnicity of Support at Home program enrollees and eligible population, by age group S@H ENROLLEES ELIGIBLE POPULATION RACE/ETHNICITY 18-59 years 60 years & older 18-59 years 60 years & older 4 42 3,516 7,435 White 33.3% 45.2% 34.9% 41.6% 2 7 2,735 1,980 Latino 16.7% 7.5% 27.1% 11.1% 2 11 2,102 6,492 Asian 16.7% 11.8% 20.9% 36.4% 2 30 1,111 1,711 Black or African American 16.7% 32.3% 11.0% 9.6% 2 3 620 238 Other race/ethnicity* 16.7% 3.2% 6.1% 1.3% 12 93 10,084 17,856 Total 100% 100% 100% 100% *Other race/ethnicity includes American Indian or Alaska Native, Native Hawaiian or Pacific Islander, two or more races, and some other race. Estimate of some other race population ages 60 & older is based on fewer than 30 sample observations. About 14% of S@H enrollees identify as gay/lesbian/same-gender-loving (10.5%) or bisexual (3.8%). According to the 2005 American Community Survey, which is the most recent available, approximately 15.4% of San Francisco’s population is gay or lesbian;9 S@H enrollees represent this population well. Individuals from single-person households are over-represented among S@H enrollees, for all age groups, as seen in Table 4. There may be two reasons for this. First, individuals living in multi-person households may receive assistance from other household members and not perceive that they need additional assistance. Second, the analysis of American Community Survey data may under-state the income of multi-person households and fewer people in these household are potentially eligible than estimated. Note that enrollees of Asian/Native Hawaiian/Pacific Islander and Hispanic/Latino backgrounds are more likely to live in multi-person households. The lower enrollment rates among those living in multi-person households may be associated with the under-enrollment of individuals in some racial/ethnic groups. 9 Gates, Gary. Same-sex Couples and the Gay, Lesbian, Bisexual Population: New Estimates from the American Community Survey. The Williams Institute, UCLA School of Law, October 2006. Support At Home – Year 1 Evaluation Report 8 Table 4. Household size of enrolled and eligible population, by age group S@H ENROLLEES ELIGIBLE POPULATION HOUSEHOLD SIZE 18-59 60 years & 18-59 60 years & All ages All ages years older years older 1 person 75.0% 78.5% 78.1% 33.4% 41.2% 38.4% 2 people 16.7% 19.4% 19% 26.7% 33.2% 30.8% 3 or more people 8.3% 2.1% 2.9% 39.9% 25.6% 30.8% Total 100% 100% 100% 100% 100% 100% *This measure of household size is based on survey responses; it does not account for dependent family relationships and how those relationships would determine income eligibility. Number of sample cases in American Community Survey=1,335. Care needs and financial needs The IOA team determines care needs through a multifaceted functional assessment as part of the enrollment process. Functional needs determine eligibility as well as the voucher amount available from the program. As seen in Table 5, equal shares of enrollees have low or medium functional need (39% each), and 21.9% have high need. Table 5. Enrollee functional need level distribution LEVEL OF NUMBER PERCENT FUNCTIONAL NEED Low 41 39.0% Medium 41 39.0% High 23 21.9% Total 105 100.0% A complete review of the financial situation of individuals is part of the enrollment process as well. Potential enrollees are grouped into three categories: high financial need (0-25% of annual median income), medium financial need (26-60% of annual median income), and low financial need (61-100% of annual median income). The San Francisco Mayor’s Office of Housing and Community Development income definitions are sourced from the U.S. Department of Housing and Urban Development; these data were published on April 1 st of 2018.The level of financial need determines the share of home care for which the enrollee will need to pay. As seen in Table 6, 40% of enrollees have high financial need, 45.7% have medium need, and 14.3% have low need. Table 6. Enrollee financial need level distribution LEVEL OF NUMBER PERCENT FINANCIAL NEED High 42 40.0% Medium 48 45.7% Low 15 14.3% Total 105 100.0% Support At Home – Year 1 Evaluation Report 9 Enrollees with high functional need are more often in the medium financial need category (65.2%) than those with other functional need levels (Table 7). Those with low functional needs are more often in the high financial need category (58.5%) than those with other functional need levels. Table 7. Enrollee level of financial need by functional need LOW MEDIUM HIGH LEVEL OF FUNCTIONAL FUNCTIONAL FUNCTIONAL TOTAL FINANCIAL NEED NEED NEED NEED # % # % # % # % High 24 58.5% 15 36.6% 3 13.0% 42 40.0% Medium 14 34.1% 19 46.3% 15 65.2% 48 45.7% Low 3 7.3% 7 17.1% 5 21.7% 15 14.3% Total 41 100.0% 41 100.0% 23 100.0% 105 100.0% Enrollees’ monthly household income is summarized in Table 8. Enrollees reported monthly household incomes ranging from $751 to nearly $7000; note that this includes households of all sizes. The average monthly income was $2,700.98 and the median was $2,273.05. Note that an individual could be eligible for Medi-Cal but not qualify for fully-paid in-home support services, and thus would be eligible for Support at Home. Table 8. Monthly total household income of enrollees INCOME Mean $2,700.98 Median $2,273.05 Minimum $751 Maximum $6,924.59 Note: Data include all household sizes. The initial assessment collects detailed information about functioning levels for specific activities. The activities for which enrollees are most often fully dependent on assistance are housework (66.7%), laundry (64.8%), shopping and errands (54.3%), meal preparation and cleanup (49.5%), and transportation (48.6%). The activities for which they are most often independent are eating (81%), telephone (74.3%), toileting (61.9%), transferring (58.1%), indoor mobility (54.3%), medication management (53.3%), dressing (47.6%), and money management (44.8%). Pre-enrollment and post-enrollment home care services The initial assessment of S@H enrollees includes questions about whether the enrollee already had some home care services at the time of the assessment. As seen in Table 9, more than 90% of enrollees were receiving some home care at the time of their assessment. The majority were relying at least in part on temporary home care solutions. Table 9. Receipt of home care by enrollees at initial assessment HAS HOME CARE? NUMBER PERCENT No home care 10 9.5% Currently has permanent home care (including paid and unpaid caregivers) 34 32.5% Currently has fully temporary home care 8 7.6% Currently has care that is partly temporary and partly permanent 46 43.8% Currently has home care with unspecified duration 7 6.7% Total 105 100.0% Support At Home – Year 1 Evaluation Report 10 Individuals who were not receiving home care more often had high financial need than those receiving home care (70% vs. 36.8%) (Table 10). Those without existing home care at time of S@H enrollment had medium to high financial need. Enrollees who reported they had some home care services were asked during their assessment whether their care needs were being met by their current services. Most (76.8%) reported that their care needs were not being met. Table 10. Enrollee financial need level by home care status RECEIVING HOME CARE AT TIME OF ASSESSMENT LEVEL OF Yes No FINANCIAL NEED # % # % High 35 36.8% 7 70.0% Medium 45 47.4% 3 30.0% Low 15 15.8% 0 0.0% Total 95 100.0% 10 100.0% Although some enrollees indicated that their care needs were met at the time of their initial S@H assessment, many enrollees reported that they experienced significant financial stress associated with paying for home care. Enrollees were asked “how much of a financial strain would you say paying for home care is/would be for you?” They were asked to respond on a five-point scale, for which 5 would be the biggest burden. As seen in Table 11, 59% of enrollees reported a 5, the highest level of burden. The financial strain is greatest among those with medium and high functional needs. Table 11. Enrollee financial strain of paying for home care by functional need level FUNCTIONAL NEED LEVEL FINANCIAL ALL ENROLLEES LOW MEDIUM HIGH STRAIN # % # % # % # % 1 (low) 6 5.7% 4 9.8% 2 4.9% 0 0.0% 2 5 4.8% 3 7.3% 2 4.9% 0 0.0% 3 17 16.2% 10 24.4% 5 12.2% 2 8.7% 4 15 14.3% 8 19.5% 4 9.8% 3 13.0% 5 (high) 62 59.0% 16 39.0% 28 68.3% 18 78.3% Total 105 100.0% 41 100.0% 41 100.0% 23 100.0% Enrollees with high and medium financial need levels more often reported that paying for home care would produce the greatest possible level of financial strain (Table 12). The open-ended comments related to the assessment were analyzed, and in these notes more than 40 enrollees reported their stress level was high due to their financial situation. Such comments included a heightened level of anxiety that enrollees cannot pay for additional services on their own given their expenses already exceed their current budget on a lower income. One enrollee stated they “needed services and help but figuring out how to afford it is stressful.” Other enrollees noted that they were currently paying for care, yet it was draining their funds and increasing their debt. In a few cases, funds were limited as a family member stopped working to care for the enrollee. Other enrollees said that the financial stress “may be affecting my health” and that “living on a limited income is hard.” Enrollees also reported significant assistance from family members and neighbors/friends. Limitations to this type of caregiving included that the family was busy and working as well as experiencing caregiver burnout. Some Support At Home – Year 1 Evaluation Report 11 informal caregivers have exhausted their FMLA benefits. Overall, this level of care was noted as inconsistent and not sustainable. Table 12. Enrollee financial strain of paying for home care by financial need level FINANCIAL NEED LEVEL FINANCIAL ALL ENROLLEES LOW MEDIUM HIGH STRAIN # % # % # % # % 1 (low) 6 5.7% 0 0.0% 2 4.2% 4 9.5% 2 5 4.8% 2 13.3% 1 2.1% 2 4.8% 3 17 16.2% 2 13.3% 8 16.7% 7 16.7% 4 15 14.3% 3 20.0% 4 8.3% 8 19.0% 5 (high) 62 59.0% 8 53.3% 33 68.8% 21 50.0% Total 105 100.0% 15 100.0% 48 100.0% 42 100.0% During the assessment, enrollees are asked about their stress level “based on your current financial responsibilities to pay for home care.” They respond on a scale of 1 to 5, with 5 indicating the highest stress level. As seen in Table 13, nearly one-third of respondents indicated the highest level of stress associated with paying for home care, and another 20% reported the second-highest level. The intake assessment also asks enrollees to rate on a 1 to 5 scale the “harm to your health and well-being today based on your current financial responsibilities to pay for your caregiver expenses?” The two highest levels of harm (4-5) were reported by a total of 41% of enrollees. Nearly one-third of enrollees reported the lowest level of harm to their health and well-being from their current financial responsibilities to pay for caregiver expenses. Table 13. Enrollee stress level and harm level based on financial responsibilities STRESS LEVEL HARM LEVEL STRESS OR # % # % HARM LEVEL 1 (low) 20 19.0% 33 31.4% 2 7 6.7% 6 5.7% 3 23 21.9% 23 21.9% 4 21 20.0% 17 16.2% 5 (high) 34 32.4% 26 24.8% Total 105 100.0% 105 100.0% Quality of life among Support at Home enrollees The Support at Home program is intended to improve the quality of life of those enrolled. During the initial assessment, enrollment coordinators ask several questions about the quality of life of enrollees. The evaluation also asks that enrollees complete a separate survey with more detailed questions about quality of life. During the initial assessment, enrollees are asked whether, in the last 12 months, their overall health has suffered because they could not afford needed home care. Note that this is a distinct question from the one described above regarding the harm to health and well-being based on financial responsibilities to pay for home care. As seen in Table 14, 47.6% of enrollees responded that their health had suffered due to the unaffordability of home care. Support At Home – Year 1 Evaluation Report 12 Table 14. Enrollee overall health suffering due to home care affordability OVERALL HEALTH # % SUFFERING Yes 50 47.6% No 55 52.4% Total 105 100.0% During the initial assessment, IOA staff give those who enroll an evaluation survey that includes the 13 questions of the Older Person’s Quality of Life scale (OPQOL),10 the 2 questions of the Patient Health Questionnaire that are used to screen for depression (PHQ-2),11 questions about their current home care, and a question regarding what they would do if they were not enrolled in Support at Home. As of June 30, 2018, surveys had been returned by 34 enrollees. Not all enrollees responded to every question; the number of respondents to each question are indicated in the tables below. The first question of the OPQOL asks the respondent to rate their “quality of life as a whole,” with response choices of very bad, bad, alright, good, and very good. As seen in Table 15, half of respondents indicated that their quality of life was “alright,” one respondent responded “bad,” and none responded “very bad.” Nearly 40% reported their quality of life as a whole was “good” and 7.1% reported it was “very good.” Table 15. Enrollee quality of life as a whole QUALITY # % OF LIFE Bad 1 3.6% Alright 14 50.0% Good 11 39.3% Very Good 2 7.1% Total 28 100.0% Table 16 presents enrollees’ responses for the remaining 12 questions of the OPQOL scale, to which enrollees can respond with strongly disagree, disagree, neither agree nor disagree, agree, or strongly agree. The 34 program enrollees who responded most often agreed or strongly agreed that they feel safe where they live, they get pleasure from their home, they take life as it comes and make the best of things, they feel lucky compared to most people, they look forward to things, they enjoy their life overall, and their family, friends, or neighbors would help them if needed. Enrollees most often disagreed or strongly disagreed that they are healthy enough to get out and about, they are healthy enough to have their independence, they have social or leisure activities/hobbies that they enjoy, they try to stay involved with things, and they have enough money to pay their bills. 10 Bowling, G, Stenner, P. (2011) Which measure of quality of life performs best in older age? A comparison of the OPQOL, CASP-19 and WHOQOL-OLD. J Epidemiol Community Health 65: 273-280. 11 Patient Health Questionnaire – 9. http://www.phqscreeners.com/ Support At Home – Year 1 Evaluation Report 13 Table 16. Enrollee responses for specific quality of life components NEITHER STRONGLY STRONGLY DISAGREE AGREE NOR AGREE DISAGREE AGREE DISAGREE 1 3 6 17 7 I enjoy my life overall 2.9% 8.8% 17.6% 50.0% 20.6% 1 3 4 17 9 I look forward to things 2.9% 8.8% 11.8% 50.0% 26.5% I am healthy enough to get out and 3 12 8 9 2 about 8.8% 35.3% 23.5% 26.5% 5.9% My family, friends, or neighbors 1 6 4 12 11 would help me if needed 2.9% 17.6% 11.8% 35.3% 32.4% I have social or leisure 4 8 6 13 3 activities/hobbies that I enjoy doing 11.8% 23.5% 17.6% 38.2% 8.8% 2 8 5 14 5 I try to stay involved with things 5.9% 23.5% 14.7% 41.2% 14.7% I am healthy enough to have my 4 9 6 11 4 independence 11.8% 26.5% 17.6% 32.4% 11.8% 0 1 4 14 15 I feel safe where I live 0.0% 2.9% 11.8% 41.2% 44.1% 0 2 3 18 11 I get pleasure from my home 0.0% 5.9% 8.8% 52.9% 32.4% I take life as it comes and make the 0 2 5 19 8 best of things 0.0% 5.9% 14.7% 55.9% 23.5% I feel lucky compared to most 1 1 5 17 10 people 2.9% 2.9% 14.7% 50.0% 29.4% I have enough money to pay for 1 8 11 12 2 household bills 2.9% 23.5% 32.4% 35.3% 5.9% Number of respondents = 34. Responses to the PHQ-2 indicate that depression is a concern for many S@H enrollees (Table 17). Of the 32 enrollees who answered these questions, 9 indicated that they have little interest or pleasure doing things more than half the days or nearly every day, and 7 felt down, depressed, or hopeless more than half the days or nearly every day. Table 17. Enrollee PHQ-2 NOT AT SEVERAL MORE THAN NEARLY ALL DAYS HALF THE DAYS EVERY DAY 6 17 3 6 Little interest or pleasure in doing things 18.8% 53.1% 9.4% 18.8% 8 17 4 3 Feeling down, depressed, or hopeless 25.0% 53.1% 12.5% 9.4% Support At Home – Year 1 Evaluation Report 14 Chapter 3 – Preliminary Results on the Impact of Support at Home Preliminary data from initial and quarterly assessments IOA staff have quarterly phone calls with S@H enrollees, during which enrollees are asked some of the same questions about their quality of life that were asked during the initial assessment. Additional questions also are asked to ascertain the impact of the program on the quality of enrollees’ lives. As of May 11, 2018, 40 enrollees had completed a quarterly phone call; this section focuses on these individuals. Enrollees are asked about the number of falls they have had during the prior three months at the initial assessment (pre-S@H) and quarterly review (post-S@H). The numbers of falls reported by enrollees declined on average, with the share reporting no falls in the prior three months increasing from 65% to 70% (Table 18). Table 18. Number of falls in prior three months reported by enrollees during initial assessment and at quarterly review NUMBER OF PRE-S@H POST-S@H FALLS # % # % 0 26 65.0% 28 70.0% 1 6 15.0% 8 20.0% 2 4 10.0% 3 7.5% 3 1 2.5% 0 0.0% 6 1 2.5% 0 0.0% 8 1 2.5% 0 0.0% 10 1 2.5% 0 0.0% 30 0 0.0% 1 2.5% Total 40 100.0% 40 100.0% As seen in Table 19, enrollees were less likely to have been hospitalized after enrollment in Support at Home than before. The share with no hospitalizations rose from 67.5% pre-S@H to 87.5% post-S@H. Table 19. Number of hospitalizations in prior three months reported by enrollees during initial assessment and at quarterly review NUMBER OF PRE-S@H POST-S@H HOSPITALIZATIONS # % # % 0 27 67.5% 35 87.5% 1 9 22.5% 4 10.0% 2 2 5.0% 1 2.5% 3 1 2.5% 0 0.0% 4 1 2.5% 0 0.0% Total 40 100.0% 40 100.0% S@H enrollees were less likely to have emergency department (ED) visits after enrollment than before (Table 20). Note that not all enrollees were asked about their ED use; this was added to the initial assessment after enrollment had started. Of the 33 people for whom pre-S@H data were available, 69.7% had reported no ED visits during the three months prior to their initial assessment. Of the 40 people for whom post-S@H data were available, 82.5% reported no ED usage. Support At Home – Year 1 Evaluation Report 15 Table 20. Number of emergency department visits in prior three months reported by enrollees during initial assessment and at quarterly review NUMBER OF ED PRE-S@H POST-S@H VISITS # % # % 0 23 69.7% 33 82.5% 1 6 18.2% 7 17.5% 2 1 3.0% 0 0.0% 3 2 6.1% 0 0.0% 6 1 3.0% 0 0.0% TOTAL 33 100.0% 40 100.0% Table 21 reports the number of missed medical appointments in the prior three months report pre-S@H and post- S@H enrollment for the 40 enrollees. There was a small decrease in the share reporting they did not miss any appointments, from 90% to 85%. Table 21. Number of missed medical appointments in prior three months reported by enrollees during initial assessment and at quarterly review NUMBER OF MISSED PRE-S@H POST-S@H APPOINTMENTS # % # % 0 36 90.0% 34 85.0% 1 3 7.5% 4 10.0% 2 1 2.5% 1 2.5% 3 0 0.0% 0 0.0% 4 0 0.0% 0 0.0% 5 0 0.0% 0 0.0% 6 0 0.0% 0 0.0% 7 0 0.0% 0 0.0% 8 0 0.0% 1* 2.5% Total 40 100.0% 40 100.0% *This client had increased medical needs, which led to a reassessment Enrollees were asked about their stress level based on their financial responsibilities, with a rating of 1 indicating the lowest level of stress and 5 indicating the greatest stress. There was a notable increase in the share of enrollees reporting the lowest level of stress before (17.5%) versus after (40.0%) enrollment, and coincident decline in the share reporting the highest level of stress (35% vs. 17.5%) (Table 22). Table 22. Enrollee stress level based on financial responsibilities reported by enrollees during initial assessment and at quarterly review STRESS PRE-S@H POST-S@H LEVEL # % # % 1 (low) 7 17.5% 16 40.0% 2 1 2.5% 4 10.0% 3 9 22.5% 8 20.0% 4 9 22.5% 5 12.5% 5 (high) 14 35.0% 7 17.5% Total 40 100.0% 40 100.0% Support At Home – Year 1 Evaluation Report 16 Enrollees were asked about the degree of harm to their quality of life they experienced due to their financial responsibilities to pay for caregiver expenses during the initial assessment and quarterly review. Enrollees reported a lower level of harm post-S@H enrollment as compared to pre-S@H enrollment (Table 23). Prior to S@H enrollment, 22.5% reported the highest level of harm and 35% reported the lowest level; after S@H enrollment, 15% reported the highest level of harm and 45% reported the lowest level. Table 23. Degree of harm from financial responsibilities reported by enrollees during initial assessment and at quarterly review HARM PRE-S@H POST-S@H RATE # % # % 1 (low) 14 35.0% 18 45.0% 2 3 7.5% 6 15.0% 3 10 25.0% 4 10.0% 4 4 10.0% 6 15.0% 5 (high) 9 22.5% 6 15.0% Total 40 100.0% 40 100.0% The survey conducted specifically for the evaluation asks enrollees what they would do if they were not in Support at Home. The most common response to this question is that the enrollee would do without care (52.9%) (Table 24). Nearly one-quarter say they would spend down their assets to qualify for Medi-Cal (23.5%), 11.8% would move to an assisted living or nursing facility, and 5.9% would move out of San Francisco. Table 24. Enrollee home care actions without Support at Home # % Make do without care 18 52.9% Spend down assets to qualify for Medi-Cal 8 23.5% Move to an assisted living or nursing facility 4 11.8% Move out of San Francisco 2 5.9% Other 9 26.5% Number of respondents = 34. Perspectives of family and friends of Support at Home enrollees In order to learn how a family member or friend is affected by Support at Home, focus groups were conducted. A total of 6 family members or friends completed two separate focus groups held on May 8, 2018 (in person; n=3) and May 14, 2018 (via conference call; n=3). Each focus group lasted 45 minutes in duration. Focus group participants were asked a number of questions, including their greatest strengths and challenges to being a caregiver, a description of their role as caregiver, other informal caregivers in the home, and the impact that caregiving has on their life. Finally, participants were asked about the impact that Support at Home has had on the care of their loved one. A few major themes were identified in the focus groups. These themes centered in the areas of: (1) the impact the enrollee and/or family member or friend had/has on the San Francisco community, (2) the depth of informal caregiving occurring in addition to the support received from Support at Home, (3) family members or friends living with their own health issues and challenges on top of caring for their loved one, and (4) the impact and benefit that Support at Home has had on family members, friends, and enrollees’ lives thus far. A theme that was Support At Home – Year 1 Evaluation Report 17 threaded throughout both focus groups was the impact of culture and the caregiving role. Focus group participants noted that it was part of their culture to care for their loved one as well as prepare them culturally appropriate food in order to maximize and maintain the S@H enrollee’s quality of life. The impact the enrollee and/or family member or friend has had on the San Francisco community Several focus group participants spoke eloquently regarding the fact that either the family member or friend or the S@H enrollee has had an impact on the San Francisco community in some way that the participant was very proud to share. One participant said prior to needing to care for her husband full-time, she was considerably involved with the San Francisco Unified School District (SFUSD) as a PTA volunteer. She attends bible study once a week, which S@H funding helps allow her to do, but she still does not do as many additional activities as what she would like. Another family member or friend is a sixth-grade teacher in the SFUSD and is not only working in this capacity full-time, but takes care of the S@H enrollee before leaving for work and when she returns home in the afternoon. Two other participants shared the role their S@H enrollee played on various boards and organizations in the city. One daughter of a S@H enrollee said: “It takes a village. We all need to care about each other. The hardest thing is to see our mother being sick. She’s done a whole lot for this city. The fact that she can’t do the things she used to do, it affects her. My mother ran community centers in [our] neighborhood (e.g., infant care, food pantry, resource center). I run into people right now and it makes me feel good how my mother helped them and their children. How much they miss her. All those politician people know her. She’s been on so many boards.” The depth of informal caregiving occurring in addition to support from Support at Home Most family members or friends spoke at great length about the extensive amount of caregiving that is needed in addition to the S@H assistance. All focus group participants noted many other caregivers for their family member – in general, they have additional family members helping (of multiple generations) in addition to the hired S@H caregiver. These family members included: nieces, nephews, aunts, children, grandchildren, and less frequently, out of town relatives. Participants spoke at great length about the communication set up among the family to ensure that the loved one was being cared for. One daughter of a S@H enrollee said: “I do what I can for my mom. My mother has her son and two grandsons who live with her. I am a 6 minute drive away. My brother lives down the street. All of us see about my mom every day.” Family members or friends living with their own health issues and challenges on top of caring for their loved one Half of the focus group participants spoke of the challenges with their own health issues, in addition to caring for the S@H enrollee. Two of the participants were disabled and not working due to chronic medical issues. One participant was elderly and retired, yet was also struggling with her own health issues. One family member or friend said: Support At Home – Year 1 Evaluation Report 18 “I’ve been disabled and out of work for a lot of years. I spend my day taking care of me and then I take care of my mother. We are a very close-knit family. We are all to see about each other if we need help…. All my sisters and brothers work. The impact and benefit that Support at Home has had thus far on the lives of family members and friends of S@H enrollees When asked about the impact of Support at Home on their lives, focus group participants repeatedly emphasized wanting more financial support so they could afford more hours of service. One participant even asked the researchers who they could contact at Support at Home to ask for more assistance. One participant almost lost their house – the cognitively impaired husband was hiding the mail and not paying the bills. “My retirement was gone - I had to take that to save the house.” Support at Home has allowed the wife to keep the house, as they were almost to the end of their financial rope. In addition, Support at Home refers enrollees to other service providers to address risk issues or for additional support, if needed. Participants overwhelmingly expressed that they feel “more relaxed” due to the program. One participant lives at least two hours away from the city and is the only family member caring for her mother; she talked about how the program has allowed her to feel like she can enjoy her retirement without having to worry about her mother constantly. “I’m hoping I will be able to find someone to be there more often. I have my own health issues. If he slides out of the bed, I can’t get him back into bed. I need someone to be there during the day to help because the boys (my grandsons) are working.” Wife of a S@H enrollee “It’s that peace of mind. That many hours that I’m at work or my nephew is at work, that he is being cared for and he’s safe…..I’m really glad I came upon that article. The next day I called. It was a Godsend. It was so frustrating to not have him qualify for MediCal. Then it was like you are scrambling to find help…some resource… There needs to be some kind of help. This is the biggest group of elders in the history of San Francisco. Okay, we’ve all worked… We put in our time in…” Daughter of a S@H enrollee Support At Home – Year 1 Evaluation Report 19 Chapter 4 – Next Steps for Evaluation The evaluation will continue to collect and analyze qualitative, survey, and quantitative data. Sources of data are summarized in Table 25. Table 25. Data elements to be used in Support at Home evaluation METHOD OF FREQUENCY OF DATA ELEMENT INSTRUMENT COLLECTION COLLECTION Demographics, functional needs, S@H intake, assessment, & At intake, assessment, & financial needs of applicants & IOA records reassessment documents reassessment enrollees S@H service plan, monthly & At service plan, monthly & Source of service (agency vs. quarterly review, & purchase of IOA records quarterly contact, purchase direct-hire) service documents of service Enrollees training completion S@H service plan documents IOA records At enrollment Direct-care provider training Reports from training provider IOA records At each training completion Amounts of copayments and At enrollment, new service S@H service plan documents IOA records vouchers approved plan Hours of service purchased by S@H purchase of service IOA records At purchase of service vouchers documents S@H intake, assessment, At intake, assessment, Attendance at medical quarterly review, & IOA records quarterly contact, appointments reassessment documents reassessment S@H intake, assessment, At intake, assessment, Falls, Hospitalizations, ED visits quarterly review, & IOA records quarterly contact, reassessment documents reassessment S@H intake, assessment, At intake, assessment, Enrollee satisfaction with services monthly & quarterly review, IOA records, UCSF monthly & quarterly contact, received reassessment & UCSF survey records reassessment, UCSF documents surveys every 6 months S@H assessment, quarterly At assessment, quarterly Quality of life, depression, current review, & reassessment, UCSF IOA records, UCSF contact, reassessment, caregiving received OPQOL12, PHQ-2,13 & UCSF- records UCSF surveys at enrollment developed items & every 6 months Disposition of enrollees leaving S@H documents IOA records At discharge the program Burden Scale for Family Family/Friend burden UCSF records Every 6 months Caregivers – short version14 Cost of hospitalizations, ED visits, Secondary sources Published data At end of program period & nursing home admissions gathered by UCSF Family/Friend perspectives & Focus group & interview UCSF records Twice during the program experiences protocol Perspectives of leaders & staff on Twice during by program, by Interview protocol UCSF records S@H UCSF S@H monthly and quarterly IOA records & UCSF Monthly and quarterly by Enrollee experiences review, & UCSF Interview records IOA, Annually by UCSF protocol 12 Bowling, G, Stenner, P. (2011) Which measure of quality of life performs best in older age? A comparison of the OPQOL, CASP-19 and WHOQOL-OLD. J Epidemiol Community Health 65: 273-280. 13 Patient Health Questionnaire – 9. http://www.phqscreeners.com/ 14 Graessel E, Berth H, Lichte T, Grau H. (2014) Subjective caregiver burden: validity of the 10-item short version of the Burden Scale for Family Caregivers BSFC-s. BMC Geriatrics 14:23. Support At Home – Year 1 Evaluation Report 20 Chapter 5 – Considerations for Year 2 The Support at Home program has achieved most of its first-year goals, and IOA has adapted the program as their team has gained knowledge about the targeted population and their needs. Support at Home expects reach its targeted enrollment for the pilot project around the first quarter of the second pilot year. Preliminary data suggest that the program is having a positive impact on reducing falls, hospitalizations, and emergency department visits. It is also reducing the stress felt by enrollees and their families. Ongoing reviews and assessments conducted by IOA and surveys conducted by UCSF will track the continuing impact of Support at Home on the well-being of San Franciscans. As Support at Home embarks on its second year, several subjects should be considered:  Preliminary data. Preliminary data from a limited subset of enrollees are promising and suggest positive impacts on families and individuals. The data and knowledge gained from the interviews with stakeholders and focus groups with family caregivers thus far indicate that there is a need for the services provided by Support at Home. While there will be a subset of enrollees who will not have received services for the full term due to program ramp up and enrollment challenges, the final evaluation report will provide a comprehensive picture of the program impact during the pilot, including economic impacts.  Outreach. This program has had an enormous outreach effort, particularly to recruit younger people who might be eligible for Support at Home. When the pilot was launched, there were expectations based on preliminary data and experience regarding the available populations, and enrollment spaces were held for younger adults with disabilities. IOA’s expansive recruiting efforts and subsequent data analysis suggest that, although there are younger individuals who will benefit from Support at Home, many younger individuals’ needs might be met through family and friend networks, and thus they may not perceive a need for Support at Home services. The relationships that IOA has built to educate the community about Support at Home will require ongoing sustenance throughout the second year of Support at Home. Additionally, there is likely a need to continue building relationships with new networks, particularly to identify and recruit racial/ethnic minorities and younger populations who would benefit from the program. This is an ongoing effort and IOA should continue to experiment and assess what elements will improve outreach efforts. The knowledge gained from this ongoing work will provide valuable information to DAAS and other agencies as they strive to improve services to their population.  Resources required for the program. IOA has made substantial investments in outreach, and unanticipated challenges to recruit younger and non-white populations have required a great deal of time for program staff. In addition, the rate of enrollees leaving the program (due to death, an increase in care needs that require institutional care, or other reason) has been greater than anticipated. IOA staff report that some enrollees are using Support at Home to receive services when they have gaps in care, temporary needs due to short-term disability, are waiting for eligibility for IHSS, or are waiting for a permanent supportive living arrangement. During the April 6 to May 3 period, Support at Home was managing 57 potential new enrollees and 14 departures, in addition to managing current enrollees. IOA has hired a short-term staff member to expand the team’s capacity to screen and assess potential enrollees, which will expedite the program’s reaching full capacity; IOA will need to determine whether this should be a permanent increase to maintain the program’s enrollment level.  Identification and recruitment of a comparison group. The S@H evaluation includes a comparison of S@H enrollees to another similar population. A comparison group that matches the enrolled population as closely as possible is needed to identify the unique impact of Support at Home. Numerous potential comparison populations were considered. The original proposal was to have applicants who had applied for Support at Home but were placed on a waiting list due to Support at Home reaching full capacity serve as the comparison Support At Home – Year 1 Evaluation Report 21 group. However, a waiting list of sufficient size has not developed. The next-best comparison group is individuals who were referred to Support at Home but either “just barely” didn’t meet eligibility requirements or chose to not enroll. Other potential comparison groups, such as enrollees in the Medi-Cal in-home support services program, would likely have substantially greater care and financial needs than S@H enrollees, making them a suboptimal comparison group. The evaluation team and IOA staff will need to continue their coordination to recruit individuals to the comparison group over the first months of the second S@H year.  Economic analysis. DAAS is interested in both the financial impacts on a household and cost-savings to the city. The UCSF team will conduct a cost-benefit analysis focusing on the net costs of the program at the end of the evaluation period. Data on the costs of the Support at Home program will be obtained from program data maintained by IOA. These costs will be compared with the benefit of preventing a worsening of enrollees’ health, including lower long-term costs of supportive care, prevention of nursing home admission, and lower rates of hospitalization and emergency department use. In addition, family and friend caregivers will be surveyed about how their caregiving impacts their employment and personal finances. The financial value of all Support at Home impacts on both enrollees and their families will be measured using data from health care providers, published literature, IOA data, and reports provided by the California Office of Statewide Health Planning and Development. The cost-benefit analysis will be conducted using Excel. The second year of Support at Home will provide important services to San Francisco residents with disabilities and will provide DAAS with valuable information to guide future investments in serving this population. IOA has already been successful in providing services to dozens of enrollees. It has also adapted the pilot program as new information has been unearthed in an effort to continually improve the program for current and prospective enrollees. Preliminary data suggest that the positive impact of Support at Home on the well-being of enrollees could be substantial; the ongoing evaluation will measure the ongoing results, including economic impact. Support At Home – Year 1 Evaluation Report 22 Appendix: Data and methods for describing the targeted population Data from American Community Survey (ACS), which is conducted by the U.S. Census Bureau, were used to measure and describe the targeted population in San Francisco. The merged 2012-2016 5-year Public-Use Microdata Sample housing and population for the public-use microdata areas (PUMAs) defining San Francisco County were analyzed. This was the same regional definition used in reports that guided development of the Support at Home program. The population of individuals potentially eligible for Support at Home was constructed by excluding all vacant housing units and the institutionalized population, and including only sample cases where age of person was reported as 18 or older with at least one of the specified disability conditions:  VA service-connected disability rating  Cognitive difficulty  Ambulatory difficulty  Independent living difficulty  Self-care difficulty From this population, cases where health insurance coverage was reported as “Medicaid, Medical Assistance, or any kind of government-assistance plan for those with low incomes or a disability” were excluded. The approach used to evaluate the income status of this potentially eligible population approximated Medicaid eligibility rules, which are based on an applicant’s tax filing status (e.g. individual, married couple, head of household, dependent). Because the American Community Survey does not include this information, assumptions were made regarding household size and dependent relationships. For most households the relationships are relatively straightforward, such as for a single-person household, a two-person husband-and-wife household, or single adult head of household with two of her own children present. Other households are more challenging to assess in terms of size and dependent relationships, such as a multigenerational household with married and unmarried adult children who may or may not report income and who may or may not have children themselves (who also may be married and who may or may not report income), in addition to the presence of married or unmarried in-laws or other relatives, who may or may not report income. Individuals who were identified as potentially eligible, and who were living in multi-person non-family households, were treated as a single person household. The exception to this was households with unmarried partners, who were treated as a married couple (with dependents if own children were present). Biological children under the age of 18 were assumed to be dependents, as were unmarried adult children who were either not in the labor force or who reported income of less than $10,000 per year, grandchildren under the age of 18 living in the care of a grandparent, and unmarried adult grandchildren living in the household of a grandparent who were not in the labor force or who reported income of less than $10,000 per year. When the relationship could be clearly determined, non-head of household married couples living in the household of a related adult were treated as any other married couple (e.g. a mother and father-in-law, one of whom has a qualifying disability and meets the income requirements, and the couple resides in the house of an adult child). A single parent or in-law living in the household of an adult child was treated as an individual (i.e., a single person household), regardless of reported income. Adult siblings were treated as a single person household, regardless of reported income. Based on this set of assumptions, an individual’s income eligibility was evaluated in terms of household size using the 2016 Unadjusted Area Median Income (AMI) for HUD Metro Fair Market Rent Area (HMFA) that contains San Support At Home – Year 1 Evaluation Report 23 Francisco. Individuals with household income greater than 100% of the area median for household size were considered not eligible. Note that personal income as reported in the ACS may not be the same as the income used by Support at Home to determine eligibility, and information about assets (e.g., savings accounts, investments) is not available in the ACS. In most cases, cell sizes presented in tables that include American Community Survey data are based on sample count of at least 30 observations. Sample sizes were too small to explore some demographic combinations, such as the cross-tabulation of age group with disability type. Exceptions to this criterion are noted.