Home-Based Primary Care Practices in the United States: Current State and Quality Improvement Approaches Bruce Leff, MD,abc Christine M. Weston, PhD,b Sarah Garrigues, BA,def Kanan Patel, MBBS, MPH,def and Christine Ritchie, MD, MSPH,def for the National Home-Based Primary Care and Palliative Care Network
OBJECTIVES: To describe the characteristics of homebased primary care practices: staffing, administrative, population served, care practices, and quality of care challenges. DESIGN: Survey of home-based primary care practices. SETTING: Home-based primary care practices in the United States. PARTICIPANTS: Members of the American Academy of Home Care Medicine and nonmember providers identified by surveyed members. MEASUREMENTS: A 58-item questionnaire that assessed practice characteristics, care provided by the practice, and how the quality of care that the practice provided was assessed. RESULTS: Survey response rate was 47.9%, representing 272 medical house calls practices. Mean average daily census was 457 patients (median 100 patients, range 1–30,972 patients). Eighty-eight percent of practices offered around-the-clock coverage for urgent concerns, 60% held regularly scheduled team meetings, 89% used an electronic medical record, and one-third used a defined quality improvement process. The following factors were associated with practices that used a defined quality improvement process: practice holds regularly scheduled team meetings to discuss specific patients (odds ratio (OR) = 2.07, 95% confidence interval (CI) = 1.02–4.21), practice conducts surveys of patients (OR = 8.53, 95% CI = 4.07–17.88), and practice is involved in National Committee for Quality Assurance patient-centered medical home (OR = 3.27, 95% CI = 1.18–9.07). Ninety percent From the aDivision of Geriatric Medicine and Gerontology, School of Medicine; bDepartment of Health Policy and Management, Bloomberg School of Public Health; cDepartment of Community and Public Health, School of Nursing, Johns Hopkins University, Baltimore, Maryland; d Division of Geriatrics, University of California at San Francisco; eSan Francisco Veterans Affairs Medical Center; and fJewish Home San Francisco, San Francisco, California. Address correspondence to Bruce Leff, MD, Division of Geriatric Medicine, Johns Hopkins University School of Medicine, 5505 Hopkins Bayview Circle, Baltimore, MD 21224. E-mail: [email protected] DOI: 10.1111/jgs.13382
JAGS 63:963–969, 2015 © 2015, Copyright the Authors Journal compilation © 2015, The American Geriatrics Society
of practices would or might participate in quality improvement activities that would provide them timely feedback on patient and setting-appropriate quality indicators. CONCLUSIONS: There is a substantial heterogeneity of home-based primary care practice types. Most practices perform activities that lend themselves to robust quality improvement efforts, and nearly all indicated interest in a national registry to inform quality improvement. J Am Geriatr Soc 63:963–969, 2015.
Key words: house calls; home-based primary care; home-based palliative care; quality of care
M
illions of frail older adults in the United States with multiple chronic conditions and functional limitations are homebound and are unable to access office-based primary care.1,2 Such individuals often receive fragmented care and consume disproportionate amounts of healthcare services.3,4 Home-based primary care practices have been demonstrated to be an effective model of care.5 Home-based primary care in the Veterans Affairs system is associated with lower hospital and nursing home use, 30-day hospital readmissions, and costs of care,6–9 without shifting costs to Medicare.10 Primary care practices focused on providing care in the home are growing rapidly. There was a 100% increase in house calls provided to Medicare beneficiaries between 2000 and 2006.11 The Independence at Home Demonstration project (section 3024 of the Affordable Care Act) will test the effectiveness of home-based primary care in the context of an innovative shared savings payment model.12 Despite the growing use of home-based primary care, no quality of care framework for it is in use at the national level. Previously developed quality-of-care standards for home-based primary care are relatively condition-specific and are not widely used.13 The absence of an existing quality framework will limit the ability of the
0002-8614/15/$15.00
964
LEFF ET AL.
model to thrive in an era of provider accountability for patient outcomes and the need for health service delivery models to prove their value to purchasers of that care. In recognition of this situation, the National Home-Based Primary Care and Palliative Care Network was formed to develop a quality-of-care framework, quality-of-care indicators, and practice-based quality improvement processes for home-based primary care and palliative care. To inform this work, it is important to understand current practice characteristics and quality-of-care activities of the field. The aim of this study was to describe the current state of home-based primary care practices in the United States, including practice arrangements, approaches to care provision, and quality of care concerns. In addition, the goal was to understand practice characteristics associated with practice-based quality-of-care improvement efforts.
METHODS The National Home-Based Primary Care and Palliative Care Network This study was undertaken as part of the agenda of the National Home-Based Primary Care and Palliative Care Network (the Network), formed in 2013 in recognition of the dearth of appropriate quality indicators for homebased primary care and palliative care practices and the individuals they serve. The long-term goal of the Network is to develop quality indicators for the field that will be widely adopted and the infrastructure (e.g., a national registry) to use such data for quality benchmarking, practicebased quality improvement, performance reporting, and comparative effectiveness research. The Network consists of representatives from 12 exemplar home-based primary care practices, three patient advocacy and consumer-focused groups (American Association of Retired Persons Public Policy Institute, Kaiser Family Foundation, National Partnership for Women and Families), and three professional associations (American Academy of Home Care Medicine (AAHCM), American Association of Hospice and Palliative Medicine, American Geriatrics Society).
Survey Development The survey was developed iteratively with input of Network members in three teleconferences. The survey focused on practice characteristics, care that the practice provided, and how the quality of care that the practice provided was assessed. The final version of the survey had 58 questions that addressed the domains of administrative and clinical characteristics of the practices, characteristics of people that the practices served, and practicerelated quality of care. Respondents were permitted to skip survey questions as they deemed appropriate. When appropriate, space was provided for additional free-text comments.
Survey Procedures The survey was distributed to the members of the AAHCM (aahcm.org), formerly the American Academy
MAY 2015–VOL. 63, NO. 5
JAGS
of Home Care Physicians, the professional organization that unites many home-based primary care practices nationally. An Internet link to the survey was e-mailed to all AAHCM members. Additional e-mails were sent every other week for 1 month to persons who did not complete the survey. A print version was mailed to those who did not respond to the on-line version of the survey. In recognition of the fact that not all home-based primary care practices are members of the AAHCM and that practices that are members of AAHCM may not be typical practices, a “snowball” recruitment approach was used by asking respondents to identify other house calls practices of which they knew. Surveys were then sent to those practices, as well. Finally, practices that provide care through the Veterans Affairs (VA) health system were excluded. Although the VA is one of the largest home-based primary care providers in the country, information about VA practices is easily accessible; quality-of-care efforts are also more centralized. The goal for this study was to assess non-VA practices, about which little is known. Practices that were solely home-based podiatry practices were also excluded.
Analysis The unit of analysis was a home-based primary care practice. A practice organization comprising multiple sites was considered to represent one practice. In the instance in which multiple surveys were received from a single practice, the data from only one survey were used, prioritizing surveys completed by practice administrators, then physicians, then nurse practitioners, then other providers. Descriptive statistics were used to describe the characteristics of the practices. Bivariate analyses were used to describe practice characteristics associated with practices that used a defined quality improvement process in which quality of care data are collected and acted upon to improve the quality of care delivery as opposed to those that did not use such a process. Factors included in bivariate analyses were preselected: practice type (solo vs group), number of practice sites in the practice (one vs multiple sites), profit status of practice (for profit vs not for profit), provider compensation (productivity only vs salary and salary plus incentives), academic affiliation of practice (yes vs no), average daily census of practice (quartiles), practice holds regularly scheduled team meetings to discuss specific patients (yes vs no), practice uses an electronic medical record (yes vs no), practice conducts surveys of patients (yes vs no), practice conducts surveys of caregivers (yes vs no), practice involved in National Committee for Quality Assurance (NCQA) patient-centered medical home (yes vs no), practice is an Independence at Home Demonstration site (yes vs no). Finally, the odds of a practice using a defined quality improvement process were modeled using logistic regression. Statistically significant factors in bivariate analyses at P < .05 were included as covariates in the regression analysis. All analyses were conducted using SAS version 9.4 (SAS Institute, Inc., Cary, NC).
JAGS
MAY 2015–VOL. 63, NO. 5
Approval The institutional review boards of the Johns Hopkins University School of Medicine and the University of California at San Francisco School of Medicine approved this study.
RESULTS The survey was sent by e-mail or mail to 1,010 people, including all 985 members of the AAHCM and 25 additional people that AAHCM members identified in survey responses. Of the 1,010, 53 surveys were not eligible for inclusion (24, incorrect e-mail addresses; 29, incorrect mailing address). Four hundred fifty-eight of 957 surveys were returned (response rate 47.9%). Data from the 458 respondents were reviewed to reconcile data to the practice level; 78 were from the same practice, four were not actively serving patients, three were podiatry practices, 69 were not part of a home-based primary care practice, 13 were part of a VA practice, 19 had no useable or primarily missing data. The final sample consisted of data from 272 practices. Ninety-five percent of respondents were members of the AAHCM, 59% were physicians, 24% were nurse practitioners, 13% were practice administrators, and 3% were physician assistants. Table 1 depicts the basic characteristics of the sample of practices. Twenty-four percent of practices had been established since 2011. A majority of practices were in urban or suburban settings and were group practices. Eighty-eight percent of practices consisted of a single site (one practice had 34 sites), 75% were for-profit entities, and in 69%, the primary sponsor or owner of the practice was an independent provider or provider group. The revenue model of nearly all practices included insurance reimbursement; 14% of practices were subsidized by hospitals or health systems. Practice personnel included a wide range of disciplines and skills. Physicians and nurse practitioners were the most common providers. The median number of physicians per practice was 1 (range 0–165). Provider compensation included a productivity component for a majority of providers; a minority of practices included a quality-of-care or patient satisfaction component as part of compensation incentives. One-quarter of practices had an academic affiliation, and nearly as many participated in teaching learners of various types. Table 2 describes the clinical profile of the practices. The mean average daily census was 457, with a median of 100 patients, and maximum of 30,972. The top decile of practices started at an average daily census of 800 patients. Practices owned or outsourced a range of services to care for their patients, most commonly home health agency services, hospice, palliative care, laboratory services, radiology services, and mental health services. Nearly 88% of practices offered 24-hour, 7-day-per-week coverage for urgent concerns, and a majority provided visits for urgent or emergency problems within 24 hours. Nearly half of practices scheduled follow-up visits, on average, monthly or more frequently if needed, and 80% always or usually assumed the role of primary care providers for their patients. Sixty percent of practices held regularly scheduled team meetings with a wide variety of provider types; in cases in which the practice employed a
HOME-BASED PRIMARY CARE PRACTICE SURVEY
965
Table 1. Basic Characteristics of Medical House Calls Practices (N = 272) Characteristic (% Missing Data by Item)
Value
Practice location (0.4%), n (%) Northeast 64 (24) Southeast 73 (27) Midwest 60 (22) Southwest 32 (12) West 42 (15) Practice setting (3%), n (%) Urban only 77 (29) Suburban only 74 (28) Rural only 31 (12) Urban and suburban 36 (14) Suburban and rural 13 (5) Urban, suburban, and rural 34 (13) Group practice (2%), n (%) 144 (54) One site per practice (0.4%), 238 (88) n (%) Number of sites per practice, 1.8 3.2, 1 (1–34) mean SD, median (range) For profit (2%), n (%) 199 (75) Primary sponsor or owner of the practice (2%) Independent provider 184 (69) (doctor, nurse practitioner, physician assistant) or provider group, n (%) Hospital or health system 53 (20) Independent investor group, 24 (9) government, other Home healthcare company 6 (2) Practice funding source or revenue model (more than one may apply) (2%), n (%) Insurance reimbursement 250 (94) (including Medicare, Medicaid, private insurers) Self-pay 81 (30) Subsidy from hospital or 36 (14) health system Subsidy from philanthropy 19 (7) Practice personnel, full-time equivalent positions, % of practices using the provider type, mean SD, median (range) (13%) Physician 85, 3.2 12.4, 1 (0–165) Nurse practitioner 73, 3.5 7.2, 2 (0–85) Physician assistant 32, 1.4 2.3, 0.6 (0–11) Registered nurse 37, 1.9 7.1, 1 (0–60) Licensed practical nurse 27, 1.2 2.9, 0.5 (0–21) Nursing assistant 19, 1.4 6.2, 0 (0–40) Medical assistant 44, 5.0 22.7, 2 (0–225) Aides 17, 0.9 4.8, 0 (0–30) Social worker 25, 0.9 1.9, 0 (0–10) Case manager, care 23, 1.6 4.3, 0.2 (0–30) manager, care coordinator Mental health provider 17, 0.5 1.7, 0 (0–10) Physical or occupational 16, 0.6 3.3, 0 (0–20) therapist Clinical pharmacist 17, 0.4 1.3, 0 (0–7) Administrative staff 61, 11.6 85.5, 2 (0–1,020) Provider compensation (more than one may apply) (14%), n (%) Productivity: billed services 112 (48) Salary only 84 (36) Salary plus incentive 55 (24) compensation
(Continued)
966
LEFF ET AL.
MAY 2015–VOL. 63, NO. 5
Table 1 (Contd.) Characteristic (% Missing Data by Item)
JAGS
Table 2. Clinical Profile of the Medical House Call Practices (N = 272) Value
Type of incentive (of those that provide incentive compensation), n (%) Productivity 46 (84) Quality of care 22 (40) Patient satisfaction 13 (24) Teamwork 10 (18) Academic affiliation (11%), 60 (25) n (%) Practice participates in 60 (100) teaching learners (0%), n (%) Types of learners practice works with (of those that teach learners), n (%) Medical students 41 (68) Nurse practitioners 37 (62) Residents 36 (60) Subspecialty fellows 20 (33) Physician assistants 15 (25) Social workers 13 (22) Nurses 11 (18) SD = standard deviation.
specific provider type, attendance of that provider type at the team meeting was generally high. Most practices used an electronic medical record for a variety of care functions. Table 3 depicts patient and practice characteristics of the practices surveyed. Practices focused on providing care to elderly adults; Medicare was the primary insurer for 79% of people served, and 63% of practices provided care to individuals under capitated arrangements. More than 20% of practices had a majority of patients who were dually eligible for Medicare and Medicaid. In terms of quality of care, 13% of practices were involved in a NCQA patient-centered medical home, 12% were involved in an accountable care organization, and 7% were participants in the Centers for Medicare and Medicaid Services Independence at Home Demonstration. Thirty-eight percent of practices surveyed patients at least annually regarding their experience with care. One-third of practices used a defined quality improvement process (collected and used data to improve care delivery); nearly half of practices collected and monitored quality indicators. Ninety percent of practices indicated that they would or might participate in a quality improvement program that would provide timely feedback on home-based primary care and palliative care– specific quality indicators, such as could be provided through a practice-based registry; this proportion remained essentially unchanged if such a program helped practices qualify for performance-based incentive programs. In bivariate analyses, the following factors were associated with practices that used a defined quality improvement process: profit status (for profit 29% vs not for profit 47%, P = .01), practice holds regularly scheduled team meetings to discuss specific patients (44% hold meetings vs 18% do not, P < .001), practice conducts surveys of patients (55% yes vs 11% no, P < .001), practice conducts
Characteristic (% Missing Data by Item)
Value
Daily census, mean standard 457 2,119, 100 (1–30,972) deviation, median (range) (8%) Number of home health agencies practice frequently works with (4%), n (%) 1 27 (10) 2–5 152 (58) 6–10 56 (22) >10 25 (10) Services, owned or partially owned by practice or outsourced, provided by practice (4%), n (%) Home health agency 226 (87) Hospice 221 (84) Palliative care 220 (85) Podiatry 210 (82) Laboratory services 232 (90) Pharmacy 210 (82) Radiology 219 (85) Infusion therapy 171 (66) Mental health services 207 (81) Practice offers 24-hour, 7-day-per217 (88) week telephone coverage for urgent patient concerns (9%), n (%) After-hours coverage provided (10%), n (%) Practice physician 161 (66) Practice nurse practitioner 115 (47) Practice physician assistant 36 (15) Nonpractice personnel 29 (12) Type of after-hours coverage (more than one may apply) (10%), n (%) Telephone call 229 (94) In-person visit 129 (53) Rapidity with which practice patients can be seen at home for urgent or emergency complaints (10%), n (%) Same day 73 (30) Next day, but within 24 hours 93 (38) Within 48–72 hours 59 (24) Within a week 17 (7) Frequency of scheduled follow-up visits for patients who are clinically stable (10%), n (%) More than once a month 6 (2) Every month 104 (43) Every other month 74 (30) Every 3 months 39 (16) Every 4–6 months 14 (6) Less often then every 6 months 2 (1) No scheduled visits 5 (2) Role of practice in patient care (more than one may apply) (9%), n (%) Always or usually assume 197 (80) primary care Sometimes assumes primary 35 (14) care Patients always keep their 19 (8) “regular” physician Practice co-manages patient 32 (13) after hospital discharge as bridge to office-based primary care
(Continued)
JAGS
MAY 2015–VOL. 63, NO. 5
HOME-BASED PRIMARY CARE PRACTICE SURVEY
Table 2 (Contd.) Characteristic (% Missing Data by Item)
Value
Practice holds regularly scheduled 148 (60) team meetings to discuss specific patients (10%), n (%) Frequency of regularly scheduled team meetings (1%), n (%) Daily 9 (6) Weekly 62 (42) Every other week 23 (16) Monthly 39 (27) Less often than monthly 11 (8) Team meeting participants across all practices that hold team meetings, n (%) [% of team meetings with participation of the provider type in cases in which the practice employs that type of provider] (0%) Physician 127 (86) [93] Nurse practitioner 111 (75) [96] Administrative staff 66 (45) [61] Nurse 45 (30) [73] Medical assistant 43 (29) [65] Case manager, care manager, 39 (26) [75] care coordinator Social worker 38 (26) [96] Physician assistant 30 (20) [89] Licensed practical nurse 24 (16) [62] Staff from other collaborating 19 (13) organizations Physical or occupational 9 (6) therapists Family members of patient 6 (4) Clinical pharmacist 10 (7) Mental health provider 9 (6) Aides 7 (5) Nursing assistant 6 (4) Other 9 (6) Practice uses electronic medical 215 (89) record (11%), n (%) Practice uses electronic medical record for (0%), n (%) Documentation 209 (97) E-prescribing 188 (87) Coordinating patient care with 127 (59) other practice-based providers Patient registry functions 101 (47) Coordinating care with home 94 (44) health agency Generating patient instructions 89 (41) Generating family or caregiver 67 (31) instructions Communicating treatment 63 (29) preferences or code status across care settings (e.g., Physician Orders for Life-Sustaining Treatment, Medical Orders for Life-Sustaining Treatment) Signing home health agency 62 (29) orders or recertifying care Practice electronic medical record is interoperable with (0%), n (%) Pharmacy 54 (25) Hospital or health system 45 (21) Home health agency 20 (9) Hospice 16 (7) Skilled nursing facility 4 (2)
967
Table 3. Characteristics of Individuals Served by Medical House Calls Practices and Quality-of-Care Issues (N = 272) Characteristic (% Missing Data by Item)
Value
Percentage of practice’s patients according to age, mean (SD) (9%) 0–17 1 7.0 18–64 12 14.8 ≥65 87 16.8 Proportion of practice’s patients that 62 31, 75 (0–100) live in home or apartment vs domiciliary facility or assisted living facility, mean SD, median (range) (9%) Mean proportion of patients in the practice with primary insurance type (11%) Medicare 79 Medicaid 9 Commercial insurance 9 Private pay 4 Practice provides care for Medicare 157 (63) Advantage, Special Needs Plan, or Program of All-Inclusive Care for the Elderly insured (8%), n (%) Proportion of patients in the practice eligible for Medicare and Medicaid (dual-eligible) (16%), n (%) 75 20 (9) Practice involved in a National 33 (13) Committee for Quality Assurance patient-centered medical home (4%), n (%) Practice is Centers for Medicare and 17 (7) Medicaid Services–designated Independence at Home demonstration site (4%), n (%) Practice is involved in an Accountable 30 (12) Care Organization (4%), n (%) Practice surveys patients about their experience with care (12%), n (%) More often than annually 47 (20) Annually 43 (18) Less often than annually 30 (13) Practice does not survey patients 119 (50) Practice surveys family members of the patient about their experience with care (12%), n (%) More often than annually 40 (17) Annually 28 (12) Less often than annually 38 (16) Practice does not survey family 133 (56) members Practice uses a defined quality 80 (33) improvement process (12%), n (%) Practice collects and monitors 117 (49) quality indicators (12%), n (%) If given opportunity, practice would participate in quality improvement program that would provide timely feedback on house call–specific quality indicators (e.g., through a practice-based registry) (13%), n (%) Yes 135 (57) No 24 (10) Maybe 79 (33)
(Continued)
968
LEFF ET AL.
MAY 2015–VOL. 63, NO. 5
Table 3 (Contd.) Characteristic (% Missing Data by Item)
Value
Would participate if such a program helped practice qualify for performance-based incentive program (13%) Yes 139 (58) No 25 (11) Maybe 74 (31) SD = standard deviation.
surveys of caregivers (50% yes vs 20% no, P < .001), and practice involved in NCQA patient-centered medical home (67% yes vs 29% no, P < .001). In logistic regression analysis, the following factors were associated with practices that used a defined quality improvement process: practice holds regularly scheduled team meetings to discuss specific patients (odds ratio (OR) = 2.07, 95% confidence interval (CI) = 1.02–4.21), practice conducts surveys of patients (OR = 8.53, 95% CI = 4.07–17.88), and practice is involved in a NCQA patient-centered medical home (OR = 3.27, 95% CI = 1.18–9.07). In a sensitivity analysis from which practices in the top decile of average daily census were excluded, bivariate and regression analyses were essentially unchanged.
DISCUSSION Healthcare service delivery is in a phase of rapid evolution in the wake of the Affordable Care Act. Home-based primary care and palliative care can serve an important function in providing value-based care to a vulnerable and costly group of individuals. To the knowledge of the authors, this is the first study to characterize non-VA medical house calls practices at the national level. It found that there is a range of practice types in terms of size, business models, provider types, and approaches to care and quality-of-care. The majority of practices are small in terms of number of providers and people served, although there are some large practices. The top decile includes practices that are multisite and use multiple types of providers. To the knowledge of the authors, such larger practices are newer and may represent a growing phenomenon. This study was performed in the context of a broader agenda to develop quality indicators for homebased primary care and palliative care that will be applicable to and embraced by such practices, eventually allowing for practice quality-of-care benchmarking and improvement, performance reporting, and comparative effectiveness research. Understanding the nature of these practices is an important foundational step in informing that work. Eighty percent of practices always or usually assume the role of primary care provider for these individuals. Only one-third of practices use a defined quality improvement process, although a substantial portion of practices engaged in activities that could feed into quality improvement processes if provided proper incentives and support, including the use of team meetings, surveying individuals and caregivers about their care
JAGS
experience, and the use of an electronic medical record. An overwhelming majority of practices expressed interest in participating in practice-based quality improvement efforts. Although there has been much work at the national level to develop a coordinated approach to quality of care and performance measurement across sites of care in work led by the National Quality Forum, the Measure Application Partnership, and other organizations, home-based primary care has been overlooked. No standardized quality-of-care framework is used for these practices or these patients. As a result of this void, these practices must use currently available, chiefly disease-specific, quality indicators that usually do not apply to the health needs and health status of their patients and that may create incentives to provide inappropriate care. The development and testing of appropriate quality indicators for home-based primary care and palliative care is needed to provide incentives for appropriate person-centered care for such individuals. The Network is in the process of developing the first iteration of those indicators.14 There are several limitations to this study. The sampling strategy, using membership rolls of AAHCM, almost certainly did not identify all of the home-based primary care practices in the United States, and AAHCM member practices may be different from non-member-associated practices. In an attempt to overcome this, a “snowball” approach was used in which survey respondents were asked to identify other practices that they knew of, and the survey was sent to those practices. Practices selfreported data, and it was not possible to confirm the accuracy of the data. VA practices were excluded from the sample. In the context of developing data to inform the development of quality indicators and methods of practice improvement for home-based primary care practices, it was thought that it would be relevant to focus on non-VA practices, although VA representation is included in Network deliberations regarding quality indicators being developed for the field. Finally, the data are cross-sectional and limit the ability to draw clear inferences on trends in this sphere of practice. In summary, there is a range of house calls practice types in terms of size, business models, provider types, and approaches to care and quality of care. As the health service delivery system continues its rapid evolution, homebased care will need to be part of successful population health strategies to improve care and value. This study will serve as a point of comparison for future work in this area.
ACKNOWLEDGMENTS We acknowledge Ms. Deborah Statom for assistance with manuscript preparation. Portions of this work were presented at the 2014 annual meeting of the American Academy of Home Care Medicine, Orlando, Florida. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Dr. Leff is a member of the Board of Directors of the American Academy of Home Care Medicine. Bruce Leff,
JAGS
MAY 2015–VOL. 63, NO. 5
Christine M. Weston, Sarah Garrigues, and Kanan Patel were funded by the Commonwealth Fund, Retirement Research Foundation. This work was supported by the Retirement Research Foundation and the Commonwealth Fund, a national private foundation based in New York City that supports independent research on healthcare and makes grants to improve healthcare practice and policy. The views presented here are those of the authors and not necessarily those of the Commonwealth Fund, its directors, officers, or staff. We acknowledge the contributions made by members of the National Home-Based Primary Care and Palliative Care Network to the development of the survey instrument: Gresham Bayne, MD, Nant Health; Lynn Beatty, MD, Visiting Physicians Association; Peter A. Boling, MD, Virginia Commonwealth University; Tom Cornwell, MD, Home Centered Care Institute; Eric De Jonge, MD, Medstar Washington Hospital Center Medical House Call Program; Tom Edes, MD, FACP, U.S. Department of Veterans Affairs; Lynn Friss Feinberg, AARP Public Policy Institute; Alanna Goldstein, MPH, American Geriatrics Society; Jen Hayashi, MD, Johns Hopkins Elder House Call Program; Benneth Husted, DO, Housecall Providers; Julia Jung, CPA, House Call Doctors; Tricia Neuman, ScD, Kaiser Family Foundation; Patricia Tomsko Nay, MD, CMD, American Academy of Hospice and Palliative Medicine; Tom Reed, Senior Advocate Resources; Constance Row, American Academy of Home Care Medicine; Christine Broderick, National Partnership For Women & Families; Theresa Soriano, MD, MPH, Mount Sinai Visiting Doctors Program; Robert Sowislo, Visiting Physicians Association. Author Contributions: Concept and design: Leff, Ritchie. Data acquisition, analysis, interpretation: Leff, Weston, Garrigues, Patel, Ritchie. Drafting of manuscript, critical revision: Leff, Weston, Garrigues, Patel, Ritchie. Final approval of submitted manuscript: Leff, Weston, Garrigues, Patel, Ritchie.
HOME-BASED PRIMARY CARE PRACTICE SURVEY
969
Sponsor’s Role: The sponsors had no role in development of the study, data acquisition, data interpretation, writing, or editing of manuscript.
REFERENCES 1. Qiu WQ, Dean M, Liu T et al. Physical and mental health of homebound older adults: An overlooked population. J Am Geriatr Soc 2010;58:2423– 2428. 2. Simonsick EM, Kasper JD, Phillips CL. Physical disability and social interaction: Factors associated with low social contact and home confinement in disabled older women (The Women’s Health and Aging Study). J Gerontol B Psychol Sci Soc Sci 1998;53B:S209–S217. 3. Congressional Budget Office. High-Cost Medicare Beneficiaries. Washington, DC: Congressional Budget Office, 2005. 4. Komisar H, Feder J. Transforming Care for Medicare Beneficiaries with Chronic Conditions and Long-Term Care Needs: Coordinating Care Across All Services. Washington, DC: National Health Policy Forum, 2011. 5. De Jonge KE, Jamshed N, Gilden D et al. Effects of home-based primary care on Medicare costs in high-risk elders. J Am Geriatr Soc 2014;62:1825–1831. 6. Beales JL, Edes T. Veteran’s Affairs home based primary care. Clin Geriatr Med 2009;25:149–154, viii–ix. 7. Kinosian B, Edes T, Davis D et al. Financial savings of home based primary care for frail veterans with chronic disabling disease. J Am Geriatr Soc 2010;58:S1 (Paper Abstracts). 8. DeCherrie LV, Soriano T, Hayashi J. Home-based primary care: A needed primary-care model for vulnerable populations. Mt Sinai J Med 2012;79:425–432. 9. Finucane TE, Fox-Whalen S, Burton JR. The Elder Housecall program at Johns Hopkins. J Long Term Home Health Care 1994;13:29–36. 10. Edes T, Kinosian B, Vukovis N et al. Better access, quality, and cost for clinically complex veterans with home-based primary care. J Am Geriatr Soc 2014;62:1954–1961. 11. Peterson LE, Landers SH, Bazemore A. Trends in physician house calls to Medicare beneficiaries. J Am Board Fam Med 2012;25:862–868. 12. Independence at home demonstration fact sheet [on-line]. Available at https://www.cms.gov/Medicare/Demonstration-Projects/DemoProjectsEvalRpts/ downloads/IAH_FactSheet.pdf Accessed February 5, 2015. 13. Smith KL, Soriano TA, Boal J. Brief communication: National quality-ofcare standards in home-based primary care. Ann Intern Med 2007;146:188–192. 14. Leff B, Carlson CM, Saliba D, Ritchie C. The invisible homebound: Setting quality-of-care standards for home-based primary and palliative care. Health Aff (Millwood) 2015;34:21–29.
Copyright of Journal of the American Geriatrics Society is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
OBJECTIVES: To describe the characteristics of homebased primary care practices: staffing, administrative, population served, care practices, and quality of care challenges. DESIGN: Survey of home-based primary care practices. SETTING: Home-based primary care practices in the United States. PARTICIPANTS: Members of the American Academy of Home Care Medicine and nonmember providers identified by surveyed members. MEASUREMENTS: A 58-item questionnaire that assessed practice characteristics, care provided by the practice, and how the quality of care that the practice provided was assessed. RESULTS: Survey response rate was 47.9%, representing 272 medical house calls practices. Mean average daily census was 457 patients (median 100 patients, range 1–30,972 patients). Eighty-eight percent of practices offered around-the-clock coverage for urgent concerns, 60% held regularly scheduled team meetings, 89% used an electronic medical record, and one-third used a defined quality improvement process. The following factors were associated with practices that used a defined quality improvement process: practice holds regularly scheduled team meetings to discuss specific patients (odds ratio (OR) = 2.07, 95% confidence interval (CI) = 1.02–4.21), practice conducts surveys of patients (OR = 8.53, 95% CI = 4.07–17.88), and practice is involved in National Committee for Quality Assurance patient-centered medical home (OR = 3.27, 95% CI = 1.18–9.07). Ninety percent From the aDivision of Geriatric Medicine and Gerontology, School of Medicine; bDepartment of Health Policy and Management, Bloomberg School of Public Health; cDepartment of Community and Public Health, School of Nursing, Johns Hopkins University, Baltimore, Maryland; d Division of Geriatrics, University of California at San Francisco; eSan Francisco Veterans Affairs Medical Center; and fJewish Home San Francisco, San Francisco, California. Address correspondence to Bruce Leff, MD, Division of Geriatric Medicine, Johns Hopkins University School of Medicine, 5505 Hopkins Bayview Circle, Baltimore, MD 21224. E-mail: [email protected] DOI: 10.1111/jgs.13382
JAGS 63:963–969, 2015 © 2015, Copyright the Authors Journal compilation © 2015, The American Geriatrics Society
of practices would or might participate in quality improvement activities that would provide them timely feedback on patient and setting-appropriate quality indicators. CONCLUSIONS: There is a substantial heterogeneity of home-based primary care practice types. Most practices perform activities that lend themselves to robust quality improvement efforts, and nearly all indicated interest in a national registry to inform quality improvement. J Am Geriatr Soc 63:963–969, 2015.
Key words: house calls; home-based primary care; home-based palliative care; quality of care
M
illions of frail older adults in the United States with multiple chronic conditions and functional limitations are homebound and are unable to access office-based primary care.1,2 Such individuals often receive fragmented care and consume disproportionate amounts of healthcare services.3,4 Home-based primary care practices have been demonstrated to be an effective model of care.5 Home-based primary care in the Veterans Affairs system is associated with lower hospital and nursing home use, 30-day hospital readmissions, and costs of care,6–9 without shifting costs to Medicare.10 Primary care practices focused on providing care in the home are growing rapidly. There was a 100% increase in house calls provided to Medicare beneficiaries between 2000 and 2006.11 The Independence at Home Demonstration project (section 3024 of the Affordable Care Act) will test the effectiveness of home-based primary care in the context of an innovative shared savings payment model.12 Despite the growing use of home-based primary care, no quality of care framework for it is in use at the national level. Previously developed quality-of-care standards for home-based primary care are relatively condition-specific and are not widely used.13 The absence of an existing quality framework will limit the ability of the
0002-8614/15/$15.00
964
LEFF ET AL.
model to thrive in an era of provider accountability for patient outcomes and the need for health service delivery models to prove their value to purchasers of that care. In recognition of this situation, the National Home-Based Primary Care and Palliative Care Network was formed to develop a quality-of-care framework, quality-of-care indicators, and practice-based quality improvement processes for home-based primary care and palliative care. To inform this work, it is important to understand current practice characteristics and quality-of-care activities of the field. The aim of this study was to describe the current state of home-based primary care practices in the United States, including practice arrangements, approaches to care provision, and quality of care concerns. In addition, the goal was to understand practice characteristics associated with practice-based quality-of-care improvement efforts.
METHODS The National Home-Based Primary Care and Palliative Care Network This study was undertaken as part of the agenda of the National Home-Based Primary Care and Palliative Care Network (the Network), formed in 2013 in recognition of the dearth of appropriate quality indicators for homebased primary care and palliative care practices and the individuals they serve. The long-term goal of the Network is to develop quality indicators for the field that will be widely adopted and the infrastructure (e.g., a national registry) to use such data for quality benchmarking, practicebased quality improvement, performance reporting, and comparative effectiveness research. The Network consists of representatives from 12 exemplar home-based primary care practices, three patient advocacy and consumer-focused groups (American Association of Retired Persons Public Policy Institute, Kaiser Family Foundation, National Partnership for Women and Families), and three professional associations (American Academy of Home Care Medicine (AAHCM), American Association of Hospice and Palliative Medicine, American Geriatrics Society).
Survey Development The survey was developed iteratively with input of Network members in three teleconferences. The survey focused on practice characteristics, care that the practice provided, and how the quality of care that the practice provided was assessed. The final version of the survey had 58 questions that addressed the domains of administrative and clinical characteristics of the practices, characteristics of people that the practices served, and practicerelated quality of care. Respondents were permitted to skip survey questions as they deemed appropriate. When appropriate, space was provided for additional free-text comments.
Survey Procedures The survey was distributed to the members of the AAHCM (aahcm.org), formerly the American Academy
MAY 2015–VOL. 63, NO. 5
JAGS
of Home Care Physicians, the professional organization that unites many home-based primary care practices nationally. An Internet link to the survey was e-mailed to all AAHCM members. Additional e-mails were sent every other week for 1 month to persons who did not complete the survey. A print version was mailed to those who did not respond to the on-line version of the survey. In recognition of the fact that not all home-based primary care practices are members of the AAHCM and that practices that are members of AAHCM may not be typical practices, a “snowball” recruitment approach was used by asking respondents to identify other house calls practices of which they knew. Surveys were then sent to those practices, as well. Finally, practices that provide care through the Veterans Affairs (VA) health system were excluded. Although the VA is one of the largest home-based primary care providers in the country, information about VA practices is easily accessible; quality-of-care efforts are also more centralized. The goal for this study was to assess non-VA practices, about which little is known. Practices that were solely home-based podiatry practices were also excluded.
Analysis The unit of analysis was a home-based primary care practice. A practice organization comprising multiple sites was considered to represent one practice. In the instance in which multiple surveys were received from a single practice, the data from only one survey were used, prioritizing surveys completed by practice administrators, then physicians, then nurse practitioners, then other providers. Descriptive statistics were used to describe the characteristics of the practices. Bivariate analyses were used to describe practice characteristics associated with practices that used a defined quality improvement process in which quality of care data are collected and acted upon to improve the quality of care delivery as opposed to those that did not use such a process. Factors included in bivariate analyses were preselected: practice type (solo vs group), number of practice sites in the practice (one vs multiple sites), profit status of practice (for profit vs not for profit), provider compensation (productivity only vs salary and salary plus incentives), academic affiliation of practice (yes vs no), average daily census of practice (quartiles), practice holds regularly scheduled team meetings to discuss specific patients (yes vs no), practice uses an electronic medical record (yes vs no), practice conducts surveys of patients (yes vs no), practice conducts surveys of caregivers (yes vs no), practice involved in National Committee for Quality Assurance (NCQA) patient-centered medical home (yes vs no), practice is an Independence at Home Demonstration site (yes vs no). Finally, the odds of a practice using a defined quality improvement process were modeled using logistic regression. Statistically significant factors in bivariate analyses at P < .05 were included as covariates in the regression analysis. All analyses were conducted using SAS version 9.4 (SAS Institute, Inc., Cary, NC).
JAGS
MAY 2015–VOL. 63, NO. 5
Approval The institutional review boards of the Johns Hopkins University School of Medicine and the University of California at San Francisco School of Medicine approved this study.
RESULTS The survey was sent by e-mail or mail to 1,010 people, including all 985 members of the AAHCM and 25 additional people that AAHCM members identified in survey responses. Of the 1,010, 53 surveys were not eligible for inclusion (24, incorrect e-mail addresses; 29, incorrect mailing address). Four hundred fifty-eight of 957 surveys were returned (response rate 47.9%). Data from the 458 respondents were reviewed to reconcile data to the practice level; 78 were from the same practice, four were not actively serving patients, three were podiatry practices, 69 were not part of a home-based primary care practice, 13 were part of a VA practice, 19 had no useable or primarily missing data. The final sample consisted of data from 272 practices. Ninety-five percent of respondents were members of the AAHCM, 59% were physicians, 24% were nurse practitioners, 13% were practice administrators, and 3% were physician assistants. Table 1 depicts the basic characteristics of the sample of practices. Twenty-four percent of practices had been established since 2011. A majority of practices were in urban or suburban settings and were group practices. Eighty-eight percent of practices consisted of a single site (one practice had 34 sites), 75% were for-profit entities, and in 69%, the primary sponsor or owner of the practice was an independent provider or provider group. The revenue model of nearly all practices included insurance reimbursement; 14% of practices were subsidized by hospitals or health systems. Practice personnel included a wide range of disciplines and skills. Physicians and nurse practitioners were the most common providers. The median number of physicians per practice was 1 (range 0–165). Provider compensation included a productivity component for a majority of providers; a minority of practices included a quality-of-care or patient satisfaction component as part of compensation incentives. One-quarter of practices had an academic affiliation, and nearly as many participated in teaching learners of various types. Table 2 describes the clinical profile of the practices. The mean average daily census was 457, with a median of 100 patients, and maximum of 30,972. The top decile of practices started at an average daily census of 800 patients. Practices owned or outsourced a range of services to care for their patients, most commonly home health agency services, hospice, palliative care, laboratory services, radiology services, and mental health services. Nearly 88% of practices offered 24-hour, 7-day-per-week coverage for urgent concerns, and a majority provided visits for urgent or emergency problems within 24 hours. Nearly half of practices scheduled follow-up visits, on average, monthly or more frequently if needed, and 80% always or usually assumed the role of primary care providers for their patients. Sixty percent of practices held regularly scheduled team meetings with a wide variety of provider types; in cases in which the practice employed a
HOME-BASED PRIMARY CARE PRACTICE SURVEY
965
Table 1. Basic Characteristics of Medical House Calls Practices (N = 272) Characteristic (% Missing Data by Item)
Value
Practice location (0.4%), n (%) Northeast 64 (24) Southeast 73 (27) Midwest 60 (22) Southwest 32 (12) West 42 (15) Practice setting (3%), n (%) Urban only 77 (29) Suburban only 74 (28) Rural only 31 (12) Urban and suburban 36 (14) Suburban and rural 13 (5) Urban, suburban, and rural 34 (13) Group practice (2%), n (%) 144 (54) One site per practice (0.4%), 238 (88) n (%) Number of sites per practice, 1.8 3.2, 1 (1–34) mean SD, median (range) For profit (2%), n (%) 199 (75) Primary sponsor or owner of the practice (2%) Independent provider 184 (69) (doctor, nurse practitioner, physician assistant) or provider group, n (%) Hospital or health system 53 (20) Independent investor group, 24 (9) government, other Home healthcare company 6 (2) Practice funding source or revenue model (more than one may apply) (2%), n (%) Insurance reimbursement 250 (94) (including Medicare, Medicaid, private insurers) Self-pay 81 (30) Subsidy from hospital or 36 (14) health system Subsidy from philanthropy 19 (7) Practice personnel, full-time equivalent positions, % of practices using the provider type, mean SD, median (range) (13%) Physician 85, 3.2 12.4, 1 (0–165) Nurse practitioner 73, 3.5 7.2, 2 (0–85) Physician assistant 32, 1.4 2.3, 0.6 (0–11) Registered nurse 37, 1.9 7.1, 1 (0–60) Licensed practical nurse 27, 1.2 2.9, 0.5 (0–21) Nursing assistant 19, 1.4 6.2, 0 (0–40) Medical assistant 44, 5.0 22.7, 2 (0–225) Aides 17, 0.9 4.8, 0 (0–30) Social worker 25, 0.9 1.9, 0 (0–10) Case manager, care 23, 1.6 4.3, 0.2 (0–30) manager, care coordinator Mental health provider 17, 0.5 1.7, 0 (0–10) Physical or occupational 16, 0.6 3.3, 0 (0–20) therapist Clinical pharmacist 17, 0.4 1.3, 0 (0–7) Administrative staff 61, 11.6 85.5, 2 (0–1,020) Provider compensation (more than one may apply) (14%), n (%) Productivity: billed services 112 (48) Salary only 84 (36) Salary plus incentive 55 (24) compensation
(Continued)
966
LEFF ET AL.
MAY 2015–VOL. 63, NO. 5
Table 1 (Contd.) Characteristic (% Missing Data by Item)
JAGS
Table 2. Clinical Profile of the Medical House Call Practices (N = 272) Value
Type of incentive (of those that provide incentive compensation), n (%) Productivity 46 (84) Quality of care 22 (40) Patient satisfaction 13 (24) Teamwork 10 (18) Academic affiliation (11%), 60 (25) n (%) Practice participates in 60 (100) teaching learners (0%), n (%) Types of learners practice works with (of those that teach learners), n (%) Medical students 41 (68) Nurse practitioners 37 (62) Residents 36 (60) Subspecialty fellows 20 (33) Physician assistants 15 (25) Social workers 13 (22) Nurses 11 (18) SD = standard deviation.
specific provider type, attendance of that provider type at the team meeting was generally high. Most practices used an electronic medical record for a variety of care functions. Table 3 depicts patient and practice characteristics of the practices surveyed. Practices focused on providing care to elderly adults; Medicare was the primary insurer for 79% of people served, and 63% of practices provided care to individuals under capitated arrangements. More than 20% of practices had a majority of patients who were dually eligible for Medicare and Medicaid. In terms of quality of care, 13% of practices were involved in a NCQA patient-centered medical home, 12% were involved in an accountable care organization, and 7% were participants in the Centers for Medicare and Medicaid Services Independence at Home Demonstration. Thirty-eight percent of practices surveyed patients at least annually regarding their experience with care. One-third of practices used a defined quality improvement process (collected and used data to improve care delivery); nearly half of practices collected and monitored quality indicators. Ninety percent of practices indicated that they would or might participate in a quality improvement program that would provide timely feedback on home-based primary care and palliative care– specific quality indicators, such as could be provided through a practice-based registry; this proportion remained essentially unchanged if such a program helped practices qualify for performance-based incentive programs. In bivariate analyses, the following factors were associated with practices that used a defined quality improvement process: profit status (for profit 29% vs not for profit 47%, P = .01), practice holds regularly scheduled team meetings to discuss specific patients (44% hold meetings vs 18% do not, P < .001), practice conducts surveys of patients (55% yes vs 11% no, P < .001), practice conducts
Characteristic (% Missing Data by Item)
Value
Daily census, mean standard 457 2,119, 100 (1–30,972) deviation, median (range) (8%) Number of home health agencies practice frequently works with (4%), n (%) 1 27 (10) 2–5 152 (58) 6–10 56 (22) >10 25 (10) Services, owned or partially owned by practice or outsourced, provided by practice (4%), n (%) Home health agency 226 (87) Hospice 221 (84) Palliative care 220 (85) Podiatry 210 (82) Laboratory services 232 (90) Pharmacy 210 (82) Radiology 219 (85) Infusion therapy 171 (66) Mental health services 207 (81) Practice offers 24-hour, 7-day-per217 (88) week telephone coverage for urgent patient concerns (9%), n (%) After-hours coverage provided (10%), n (%) Practice physician 161 (66) Practice nurse practitioner 115 (47) Practice physician assistant 36 (15) Nonpractice personnel 29 (12) Type of after-hours coverage (more than one may apply) (10%), n (%) Telephone call 229 (94) In-person visit 129 (53) Rapidity with which practice patients can be seen at home for urgent or emergency complaints (10%), n (%) Same day 73 (30) Next day, but within 24 hours 93 (38) Within 48–72 hours 59 (24) Within a week 17 (7) Frequency of scheduled follow-up visits for patients who are clinically stable (10%), n (%) More than once a month 6 (2) Every month 104 (43) Every other month 74 (30) Every 3 months 39 (16) Every 4–6 months 14 (6) Less often then every 6 months 2 (1) No scheduled visits 5 (2) Role of practice in patient care (more than one may apply) (9%), n (%) Always or usually assume 197 (80) primary care Sometimes assumes primary 35 (14) care Patients always keep their 19 (8) “regular” physician Practice co-manages patient 32 (13) after hospital discharge as bridge to office-based primary care
(Continued)
JAGS
MAY 2015–VOL. 63, NO. 5
HOME-BASED PRIMARY CARE PRACTICE SURVEY
Table 2 (Contd.) Characteristic (% Missing Data by Item)
Value
Practice holds regularly scheduled 148 (60) team meetings to discuss specific patients (10%), n (%) Frequency of regularly scheduled team meetings (1%), n (%) Daily 9 (6) Weekly 62 (42) Every other week 23 (16) Monthly 39 (27) Less often than monthly 11 (8) Team meeting participants across all practices that hold team meetings, n (%) [% of team meetings with participation of the provider type in cases in which the practice employs that type of provider] (0%) Physician 127 (86) [93] Nurse practitioner 111 (75) [96] Administrative staff 66 (45) [61] Nurse 45 (30) [73] Medical assistant 43 (29) [65] Case manager, care manager, 39 (26) [75] care coordinator Social worker 38 (26) [96] Physician assistant 30 (20) [89] Licensed practical nurse 24 (16) [62] Staff from other collaborating 19 (13) organizations Physical or occupational 9 (6) therapists Family members of patient 6 (4) Clinical pharmacist 10 (7) Mental health provider 9 (6) Aides 7 (5) Nursing assistant 6 (4) Other 9 (6) Practice uses electronic medical 215 (89) record (11%), n (%) Practice uses electronic medical record for (0%), n (%) Documentation 209 (97) E-prescribing 188 (87) Coordinating patient care with 127 (59) other practice-based providers Patient registry functions 101 (47) Coordinating care with home 94 (44) health agency Generating patient instructions 89 (41) Generating family or caregiver 67 (31) instructions Communicating treatment 63 (29) preferences or code status across care settings (e.g., Physician Orders for Life-Sustaining Treatment, Medical Orders for Life-Sustaining Treatment) Signing home health agency 62 (29) orders or recertifying care Practice electronic medical record is interoperable with (0%), n (%) Pharmacy 54 (25) Hospital or health system 45 (21) Home health agency 20 (9) Hospice 16 (7) Skilled nursing facility 4 (2)
967
Table 3. Characteristics of Individuals Served by Medical House Calls Practices and Quality-of-Care Issues (N = 272) Characteristic (% Missing Data by Item)
Value
Percentage of practice’s patients according to age, mean (SD) (9%) 0–17 1 7.0 18–64 12 14.8 ≥65 87 16.8 Proportion of practice’s patients that 62 31, 75 (0–100) live in home or apartment vs domiciliary facility or assisted living facility, mean SD, median (range) (9%) Mean proportion of patients in the practice with primary insurance type (11%) Medicare 79 Medicaid 9 Commercial insurance 9 Private pay 4 Practice provides care for Medicare 157 (63) Advantage, Special Needs Plan, or Program of All-Inclusive Care for the Elderly insured (8%), n (%) Proportion of patients in the practice eligible for Medicare and Medicaid (dual-eligible) (16%), n (%) 75 20 (9) Practice involved in a National 33 (13) Committee for Quality Assurance patient-centered medical home (4%), n (%) Practice is Centers for Medicare and 17 (7) Medicaid Services–designated Independence at Home demonstration site (4%), n (%) Practice is involved in an Accountable 30 (12) Care Organization (4%), n (%) Practice surveys patients about their experience with care (12%), n (%) More often than annually 47 (20) Annually 43 (18) Less often than annually 30 (13) Practice does not survey patients 119 (50) Practice surveys family members of the patient about their experience with care (12%), n (%) More often than annually 40 (17) Annually 28 (12) Less often than annually 38 (16) Practice does not survey family 133 (56) members Practice uses a defined quality 80 (33) improvement process (12%), n (%) Practice collects and monitors 117 (49) quality indicators (12%), n (%) If given opportunity, practice would participate in quality improvement program that would provide timely feedback on house call–specific quality indicators (e.g., through a practice-based registry) (13%), n (%) Yes 135 (57) No 24 (10) Maybe 79 (33)
(Continued)
968
LEFF ET AL.
MAY 2015–VOL. 63, NO. 5
Table 3 (Contd.) Characteristic (% Missing Data by Item)
Value
Would participate if such a program helped practice qualify for performance-based incentive program (13%) Yes 139 (58) No 25 (11) Maybe 74 (31) SD = standard deviation.
surveys of caregivers (50% yes vs 20% no, P < .001), and practice involved in NCQA patient-centered medical home (67% yes vs 29% no, P < .001). In logistic regression analysis, the following factors were associated with practices that used a defined quality improvement process: practice holds regularly scheduled team meetings to discuss specific patients (odds ratio (OR) = 2.07, 95% confidence interval (CI) = 1.02–4.21), practice conducts surveys of patients (OR = 8.53, 95% CI = 4.07–17.88), and practice is involved in a NCQA patient-centered medical home (OR = 3.27, 95% CI = 1.18–9.07). In a sensitivity analysis from which practices in the top decile of average daily census were excluded, bivariate and regression analyses were essentially unchanged.
DISCUSSION Healthcare service delivery is in a phase of rapid evolution in the wake of the Affordable Care Act. Home-based primary care and palliative care can serve an important function in providing value-based care to a vulnerable and costly group of individuals. To the knowledge of the authors, this is the first study to characterize non-VA medical house calls practices at the national level. It found that there is a range of practice types in terms of size, business models, provider types, and approaches to care and quality-of-care. The majority of practices are small in terms of number of providers and people served, although there are some large practices. The top decile includes practices that are multisite and use multiple types of providers. To the knowledge of the authors, such larger practices are newer and may represent a growing phenomenon. This study was performed in the context of a broader agenda to develop quality indicators for homebased primary care and palliative care that will be applicable to and embraced by such practices, eventually allowing for practice quality-of-care benchmarking and improvement, performance reporting, and comparative effectiveness research. Understanding the nature of these practices is an important foundational step in informing that work. Eighty percent of practices always or usually assume the role of primary care provider for these individuals. Only one-third of practices use a defined quality improvement process, although a substantial portion of practices engaged in activities that could feed into quality improvement processes if provided proper incentives and support, including the use of team meetings, surveying individuals and caregivers about their care
JAGS
experience, and the use of an electronic medical record. An overwhelming majority of practices expressed interest in participating in practice-based quality improvement efforts. Although there has been much work at the national level to develop a coordinated approach to quality of care and performance measurement across sites of care in work led by the National Quality Forum, the Measure Application Partnership, and other organizations, home-based primary care has been overlooked. No standardized quality-of-care framework is used for these practices or these patients. As a result of this void, these practices must use currently available, chiefly disease-specific, quality indicators that usually do not apply to the health needs and health status of their patients and that may create incentives to provide inappropriate care. The development and testing of appropriate quality indicators for home-based primary care and palliative care is needed to provide incentives for appropriate person-centered care for such individuals. The Network is in the process of developing the first iteration of those indicators.14 There are several limitations to this study. The sampling strategy, using membership rolls of AAHCM, almost certainly did not identify all of the home-based primary care practices in the United States, and AAHCM member practices may be different from non-member-associated practices. In an attempt to overcome this, a “snowball” approach was used in which survey respondents were asked to identify other practices that they knew of, and the survey was sent to those practices. Practices selfreported data, and it was not possible to confirm the accuracy of the data. VA practices were excluded from the sample. In the context of developing data to inform the development of quality indicators and methods of practice improvement for home-based primary care practices, it was thought that it would be relevant to focus on non-VA practices, although VA representation is included in Network deliberations regarding quality indicators being developed for the field. Finally, the data are cross-sectional and limit the ability to draw clear inferences on trends in this sphere of practice. In summary, there is a range of house calls practice types in terms of size, business models, provider types, and approaches to care and quality of care. As the health service delivery system continues its rapid evolution, homebased care will need to be part of successful population health strategies to improve care and value. This study will serve as a point of comparison for future work in this area.
ACKNOWLEDGMENTS We acknowledge Ms. Deborah Statom for assistance with manuscript preparation. Portions of this work were presented at the 2014 annual meeting of the American Academy of Home Care Medicine, Orlando, Florida. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Dr. Leff is a member of the Board of Directors of the American Academy of Home Care Medicine. Bruce Leff,
JAGS
MAY 2015–VOL. 63, NO. 5
Christine M. Weston, Sarah Garrigues, and Kanan Patel were funded by the Commonwealth Fund, Retirement Research Foundation. This work was supported by the Retirement Research Foundation and the Commonwealth Fund, a national private foundation based in New York City that supports independent research on healthcare and makes grants to improve healthcare practice and policy. The views presented here are those of the authors and not necessarily those of the Commonwealth Fund, its directors, officers, or staff. We acknowledge the contributions made by members of the National Home-Based Primary Care and Palliative Care Network to the development of the survey instrument: Gresham Bayne, MD, Nant Health; Lynn Beatty, MD, Visiting Physicians Association; Peter A. Boling, MD, Virginia Commonwealth University; Tom Cornwell, MD, Home Centered Care Institute; Eric De Jonge, MD, Medstar Washington Hospital Center Medical House Call Program; Tom Edes, MD, FACP, U.S. Department of Veterans Affairs; Lynn Friss Feinberg, AARP Public Policy Institute; Alanna Goldstein, MPH, American Geriatrics Society; Jen Hayashi, MD, Johns Hopkins Elder House Call Program; Benneth Husted, DO, Housecall Providers; Julia Jung, CPA, House Call Doctors; Tricia Neuman, ScD, Kaiser Family Foundation; Patricia Tomsko Nay, MD, CMD, American Academy of Hospice and Palliative Medicine; Tom Reed, Senior Advocate Resources; Constance Row, American Academy of Home Care Medicine; Christine Broderick, National Partnership For Women & Families; Theresa Soriano, MD, MPH, Mount Sinai Visiting Doctors Program; Robert Sowislo, Visiting Physicians Association. Author Contributions: Concept and design: Leff, Ritchie. Data acquisition, analysis, interpretation: Leff, Weston, Garrigues, Patel, Ritchie. Drafting of manuscript, critical revision: Leff, Weston, Garrigues, Patel, Ritchie. Final approval of submitted manuscript: Leff, Weston, Garrigues, Patel, Ritchie.
HOME-BASED PRIMARY CARE PRACTICE SURVEY
969
Sponsor’s Role: The sponsors had no role in development of the study, data acquisition, data interpretation, writing, or editing of manuscript.
REFERENCES 1. Qiu WQ, Dean M, Liu T et al. Physical and mental health of homebound older adults: An overlooked population. J Am Geriatr Soc 2010;58:2423– 2428. 2. Simonsick EM, Kasper JD, Phillips CL. Physical disability and social interaction: Factors associated with low social contact and home confinement in disabled older women (The Women’s Health and Aging Study). J Gerontol B Psychol Sci Soc Sci 1998;53B:S209–S217. 3. Congressional Budget Office. High-Cost Medicare Beneficiaries. Washington, DC: Congressional Budget Office, 2005. 4. Komisar H, Feder J. Transforming Care for Medicare Beneficiaries with Chronic Conditions and Long-Term Care Needs: Coordinating Care Across All Services. Washington, DC: National Health Policy Forum, 2011. 5. De Jonge KE, Jamshed N, Gilden D et al. Effects of home-based primary care on Medicare costs in high-risk elders. J Am Geriatr Soc 2014;62:1825–1831. 6. Beales JL, Edes T. Veteran’s Affairs home based primary care. Clin Geriatr Med 2009;25:149–154, viii–ix. 7. Kinosian B, Edes T, Davis D et al. Financial savings of home based primary care for frail veterans with chronic disabling disease. J Am Geriatr Soc 2010;58:S1 (Paper Abstracts). 8. DeCherrie LV, Soriano T, Hayashi J. Home-based primary care: A needed primary-care model for vulnerable populations. Mt Sinai J Med 2012;79:425–432. 9. Finucane TE, Fox-Whalen S, Burton JR. The Elder Housecall program at Johns Hopkins. J Long Term Home Health Care 1994;13:29–36. 10. Edes T, Kinosian B, Vukovis N et al. Better access, quality, and cost for clinically complex veterans with home-based primary care. J Am Geriatr Soc 2014;62:1954–1961. 11. Peterson LE, Landers SH, Bazemore A. Trends in physician house calls to Medicare beneficiaries. J Am Board Fam Med 2012;25:862–868. 12. Independence at home demonstration fact sheet [on-line]. Available at https://www.cms.gov/Medicare/Demonstration-Projects/DemoProjectsEvalRpts/ downloads/IAH_FactSheet.pdf Accessed February 5, 2015. 13. Smith KL, Soriano TA, Boal J. Brief communication: National quality-ofcare standards in home-based primary care. Ann Intern Med 2007;146:188–192. 14. Leff B, Carlson CM, Saliba D, Ritchie C. The invisible homebound: Setting quality-of-care standards for home-based primary and palliative care. Health Aff (Millwood) 2015;34:21–29.
Copyright of Journal of the American Geriatrics Society is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
