Canadian Study Shows Promising Results on Home Care Cost Effectiveness
by Catherine Sullivan
This article first appeared in the June, 2000 CARING magazine
Researcher Marcus Hollander, PhD, a former director of continuing care in British Columbia, is leading a national effort to compare costs of providing treatment at home to alternative settings including hospitals and nursing homes. The first study compared costs incurred by home care clients to costs incurred by similar persons in skilled nursing facilities and found home care costs were significantly less for home care. The average savings ranged from 25% to 60%. The first of 15 planned research projects on the cost effectiveness of home care to be released during 2000 offers evidence of significant savings associated with providing services in the home as compared to long-term institutions.
Recent results from a Canadian study investigating the relative cost effectiveness of home care compared to other services provides further evidence that one of home care's many benefits to patients is that it is a bargain. Researcher Marcus Hollander, PhD, a former director of continuing care in British Columbia, is leading a national effort to compare costs of providing treatment at home to alternative settings including hospitals and nursing homes. The first study, "Substudy 1: Comparative Cost Analysis of Home Care and Residential Care Services," compared costs incurred by home care clients to costs incurred by similar persons in skilled nursing facilities and found home care costs were significantly less for home care (Hollander, 1999). The average savings ranged from 25% to 60%.
The first of 15 planned research projects on the cost effectiveness of home care to be released during 2000 provides evidence of significant savings associated with providing services in the home as compared to long-term institutions.
Hollander reviews and critiques the existing literature on home care cost-effectiveness, including the American experience. Studies on the cost-effectiveness of home care in the United States have produced mixed results. Those that have investigated use of home care for particular populations, for example, low-birthweight children, ventilator-dependent adults, and a secondary analysis of previous research have found home care to be significantly less costly compared to hospital and long-term care facilities (Casiro et al, 1993; Bach et al, 1992; Hughes et al, 1997). However, US researchers have also published articles based on evaluations of the national long-term care demonstrations (often called the channeling project) that have concluded or indicated that home care is not a cost-effective substitute for nursing home services.
Hollander's premise is that there are three distinct models of home care that should be evaluated separately. One model is home care as a source of preventive and maintenance model designed to reduce the rate of deterioration for persons with relatively low-level care needs. A second model is an acute care substitution model in which home care replaces hospital care. A third model is the long-term care substitution model in which home care replaces facility-based care.
Hollander's chief criticism of the American approach to cost-effectiveness research in home care is its failure to distinguish among different models of home care. Hollander concludes that channeling researchers took a preventive and maintenance program model and evaluated it based on its impact as a substitution for long-term care facility care. In addition, Hollander believes that the nationalized Canadian health care system encourages more appropriately targeted resources than does the more fragmented private-pay and government funded US system. Probably the best known US researcher on the subject of home care cost-effectiveness is William Weissert, PhD, who has published several articles on the topic. In Figure 1, Hollander compares the Canadian long-term care system to Weissert's views of the US system published in 1985 to illustrate his main points of disagreement with the US approach to providing community-based care and to evaluating its cost-effectiveness (Weissert, 1985).
| FIGURE 1 Comparing Weissert's Seven Reasons Why it is Difficult to Make Community Care Cost-Effective to the British Columbia Experience |
Weissert's Seven Reasons (1985)
|
| Source: Table 7-3, Substudy 1: Comparative Cost Analysis of Home Care and Residential Care Services, 1999 |
The British Columbia Experience
|
There are higher overhead costs for smaller agencies but many small communities can only support one agency. There is no room for increased efficiencies in such cases.
True, this is a characteristic of a care-based system such as continuing care.
In conducting the study, Hollander used a database unique to British Columbia that contained patient status measures for both residents and community-based elderly for three cohorts, new admissions during a six-month period in 1987/88, 1990/91 and 1993/94. Patient costs were tracked for one year prior to admission, and for three years after admission to either institution or community-based care. Cost and utilization data were collected for fee-for-service physician care, hospital care, pharmaceuticals, residential care (including extended-care beds in hospitals), home care nursing and therapy, homemaker services and adult day care.
Because the research model had identical measures of patient status, the frequent problem of matching patients with similar needs was easily mastered. An initial review of the data revealed that care recipients were frequently reassessed and many recipients moved from home care to residential services or vice versa during the study period. To maximize the sample size, Hollander defined a full-time equivalent measure of care, so that persons moving from one type of care to another would be included in the database based on the number of days they were in either residential or home care and differentiated by their level of need at different points in time. This approach to assignment of costs also allowed Hollander to compare persons who were stable (stayed at the same level of care in the same setting over the study period) to those who changed, and he found significant differences between the groups in terms of the costs of care. Hollander states the advantage of his approach is that "it allows for a direct comparison of costs, by level of care, for home/community services and residential services, without a significant loss of data for the analysis (Hollander, 1999)."
The key findings of the study are summarized as follows:
Costs for home care clients by level of care range from 40% to 75% of the costs of facility care, with lower need patients in the community costing only 40% of lower need institutionalized patients and higher need patients costing 75% of higher need institutionalized patients. Thus, savings range from 25% to 60% depending on the level of care required.
For stable patients, those whose needs remain relatively unchanged for six months or more, the overall costs are about one half of the overall costs for facility patients. For home care clients who changed their type or level of care, but did not die, costs are about 70% of the costs for lower need institutional patients and 90% of higher need institutional patients. The costs for home care clients who died were higher for all levels of care, than for facility clients who died.
There was a wider differential in comparative costs for hospital care between the residents and home care recipients. Hospital costs accounted for 50% to 60% of total overall costs for home care clients with medical services accounting for 5%-10% for a total of 70% of costs for community-based clients. In contrast, hospital and medical costs accounted for approximately 15% or less of costs for residents.
Hollander discusses the limitations of the study's approach to the study of the relative cost-effectiveness of home care as a substitute for residential care by constructing full-time equivalent patients. To address the issue of whether the matched pairing of FTE constructs can be duplicated by comparing "real world patients," Hollander conducted several sensitivity analyses. He found that costs were similar to the FTE findings when he examined patient costs using the client's status at first admission to home care or residential facility. Likewise, he attempted to examine the potential bias that may occur in any study that uses matched pairing without randomization. Overall, he found the home care users and facility residents with similar assessment scores had similar activities of daily living limitations, but residents were more likely to score higher on measures of instrumental activities of daily living and mental health status.
To further investigate differences between FTE clients and "real world patients," Hollander examined costs for individual clients who received both home care and residential care during one of the study periods. He found similarities between the use of residential services by the FTE cohort and the individual patients who used both types of care. However, home care utilization was greater for the individual patient cohort, a difference that he attributes to the costs associated with borderline cases of persons whose home care costs increase as efforts are made to keep them at home instead of admitting them to a residential facility.
Hollander examined the sensitivity of his pricing of the unit costs of various types of care and found that fluctuations in the pricing of drugs, home care services and adult day care did not have a noticeable effect on the relative cost-effectiveness of home care services compared to residential services. Not surprisingly, fluctuations in the per-diem rates of facility care, which account for most of the cost for facility clients, had a more direct impact on the relative cost-effectiveness of home care services compared to residential services. As in the US, Hollander notes that the per-diem rates of residential care are variable across Canada and within British Columbia. The available data did not allow for precise case costing, which led researchers to estimate unit costs for home care and facility care. Thus, he concludes "there is considerably more room to make cost-effective substitutions in areas with comparably higher facility per diem rates, assuming home care costs are not also proportionately higher (Hollander, 1999)."
As it has happened in the United States, the rising costs of long-term care in Canada have led to public debate over the government's role in funding services for an aging population. Before adopting reforms or reducing funding for services, Canada has opted to study the situation to help determine the best approach to delivering long-term care services in the 21st century. This study and the 14 others to follow will provide information to help shape Canadian policy (see sidebar for description of the overall project).
Hollander suggests that at least in British Columbia, home care is a cost-effective substitute to facility care in most cases. He is fairly critical of the influential channeling evaluations conducted in the United States in the 1980s. However, he sees promise in the more recent evaluations of more managed-care types of models of home care service delivery, including Onlok's program and Arizona's Medicaid long-term care waiver program. These newer models, he states, "lend support to a hypothesis that the way service delivery systems are structured may have an impact on the cost-effectiveness of home care compared to residential care (Hollander, 1999)."
For the exceptions, patients who die, whose costs are higher when they are in home care than when they are in residential facilities, Hollander suggests there is considerable potential for new and innovative programs for home based palliative care and hospice care. Likewise, for unstable patients who also tend to cost more when they are in the community than when they are residents, Hollander calls for pilot projects and further evaluation of enhanced care at transition points to re-stabilize clients as quickly as possible.
Hollander calls for continued research on identifying areas for substituting home care for facility care. He is a proponent of a single entry system with coordinated needs assessment, client classification, placement and case management with a single, system-level administrative and funding structure. The Canadian system at this point is closer to such a model than is the US system with no single-entry system or uniform client classification and multiple payor sources and administrative layers.
Despite differences between the Canadian and US health care systems, agencies may take cues from these findings. Those agencies that participate in managed care contracts and Medicare-certified agencies moving to prospective payment have a competitive interest in encouraging cost-effective substitutions. Hollander's recommendations about careful assessment and case management and attention to the costs of patients in transition and those at the end-of-life have particular relevance for an agency looking for ways to reduce costs to manage under newer payment systems.
The study is available to download or purchase from the research project's website at www.homecarestudy.com. It can also be purchased through the National Evaluation of the Cost-Effectiveness of Home Care, 308 - 895 Fort Street, Victoria, British Columbia, V8W 1H7. n
About the Author: Catherine Sullivan is director of policy research at the National Association for Home Care in Washington, DC.
Contact Information: If you wish to comment on this article, e-mail caring_comments@ nahc.org.
| The National Evaluation of the Cost Effectiveness of Home Care | ||
|
Each year in Canada, it is estimated that governments spend $2 billion on home care, $7 billion on intermediate-care institutions and $26 billion on hospitals. But is that money being spent wisely and in the most cost-effective way possible? Are clients getting the most appropriate care for the best price? Are governments putting their dollars to work in places where they get the best return? Are families shouldering the burden of costs for home care services? These are some of the questions the National Evaluation of Cost-Effectiveness of Home Care hopes to answer in its extensive program of research. Through 15 inter-related studies, the evaluation will assess the differences in costs and quality of care between home care and the various forms of institutional care. All studies will be completed by the fall of 2000. Under the national Evaluation of the Cost-Effectiveness of Home Care, six studies will evaluate the cost effectiveness of home care compared to residential long-term care. These studies will compare the costs (formal and informal) of clients treated in their homes to those treated in residential facilities and will examine the factors contributing to decisions on what home care services are to be provided. Nine studies will evaluate the cost effectiveness of home care as an alternative to acute care in institutions. Researchers will examine the cost-effectiveness of specific services, such as home-based intravenous therapy and care for low-birthweight infants. Study Titles |
||
| Substudy | 2 | Care Trajectories of Home Care Clients |
| Substudy | 3 | Cost Implications of Informal Supports |
| Substudy | 4 | Pilot Study of the Costs and Outcomes of Home Care and Residential Long-Term Care |
| Substudy | 5 | Large-Scale Study of the Costs and Outcomes of Home Care and Residential Long-Term Care |
| Substudy | 6 | Decision Making Processes of Assessors/Case Managers |
| Substudy | 7 | An Analysis of Home Care Clients, Service Utilization and Costs |
| Substudy | 8 | An Exploration of Client Classification for Short-term Home Care |
| Substudy | 9 | Costs of Acute Care Services Compared to Home Care Services |
| Substudy | 10 | The Cost-Benefit Impact of a Day Hospital Program |
| Substudy | 11 | An Economic Evaluation of Hospital-Based and Home-Based Antibiotic Intravenous Therapy |
| Substudy | 12 | The Cost-Effectiveness of Home Versus Hospital Management of Feeding Difficulties in Preterm Infants |
| Substudy | 13 | Victoria Geriatric Outcomes Evaluation Study |
| Substudy | 14 | Evaluation of the Cost-Effectiveness of the Quick Response Team Program of Saskatoon District Health |
| Substudy | 15 | An Analysis of Blockages to the Effective Transfer of Clients from Acute Care to Home Care |
| Source: National Evaluation of the Cost-Effectiveness of Home Care website, http://www.homecarestudy.com. | ||
Bach, J., P. Intinola, A. Alba, and I. Holland. "The Ventilator-assisted Individual: Cost Analysis of Institutionalization vs. Rehabilitation and In-home Management." Chest, vol. 101 no.1 (1992): 26-30.
Casiro, O., M.E. McKenzie, L. McFayden, et al. "Earlier Discharge with Community-based Intervention for Low Birth Weight Infants: A Randomized Trial." Pediatrics, vol. 92 no.1 (1993): 128-134.
Hollander, M. "Substudy 1; Comparative Cost Analysis of Home Care and Residential Care Services: Preliminary Findings." Victoria, British Columbia: Report for the Health Transition Fund, Health Canada, (November 1999).
Hughes,S., A. Ulasevich, F.M. Weaver, et al. "Impact of Home Care on Hospital Days: A Meta Analysis." Health Services Research, 4 (1997): 415-432.
Weissert, W.G. "Seven Reasons Why it is So Difficult to Make Community-based Long-term Care Cost Effective." Health Services Research, vol 20 no. 4 (1985): 423-433.