BMJ, ONLINE FIRST, bmj.com. 2009; 339: 1-15. Vikram R Comondore, PJ Devereaux, Qi Zhou, Samuel B Stone, Jason W Busse, Nikila C Ravindran, Karen E Burns, Ted Haines, Bernadette Stringer, Deborah J Cook, Stephen D Walter, Terrence Sullivan, Otavio Berwanger, Mohit Bhandari, Sarfaraz Banglawala, John N Lavis, Brad Petrisor, Holger Schϋnemann, Katie Walsh, Neera Bhatnager, Gordon H Guyatt.

Backgound

Medicare’s commitment to nursing home care in Canada is ill-defined and variable, resulting in a patchwork of for-profit and not-for-profit (NFP) ownership, with the latter being charitable, government, or privately owned facilities. Payment to nursing homes is similarly patchwork, with a mixture of both public and private funding. In a previous edition of e-Rounds (#17), we examined long-term care; this month’s edition re-examines how facility ownership influences outcomes.

Methodology

This elegant systematic review and meta-analysis of studies examines the quality of care in for-profit versus not-for-profit (some privately and others publicly-owned) nursing homes. A multi-modal strategy was focused on 18 bibliographic data bases and used terms specific to nursing homes (e.g. long-term care, homes for the aged) and related to ownership (e.g. investor-owned, proprietary). Eligible for inclusion in the analysis were for-profit and not-for-profit nursing home where patients reside and for which quality of care outcomes measures were available. The quality of care assessment was based on commonly used measures to capture quality and appropriateness of care, including number of staff per resident or level of training (this measure has been shown to be highly correlated with process and outcome quality); use of physical restraints (this measure has been employed as an important process measure); pressure ulcers (this measure has been stressed because pressure ulcers are predictable and are associated with pain, infection and outcomes); and, regulatory deficiencies (this measure is an overall measure of quality).

Two reviewers independently assessed titles and abstracts, and research personnel who were not involved in the screening of abstracts masked the study results from texts and tables. Eligibility was determined independently by two reviewers. Outcomes described in the studies were pooled using random effects models. Studies were grouped as favouring for-profit or not-for-profit status or mixed results. 

Results

Of the 8,827 abstracts screened, 956 were selected for full-text review, from which 82 papers spanning 1965-2003 were deemed eligible and subjected to data abstraction and analysis. Supplementary information was requested from 36, and received from 25, authors. In 40 studies, all statistically significant analyses favoured not-for-profit facilities, and in three studies all statistically significant analyses favoured for-profit status. In 34 studies, specifically comparing for-profit with privately-owned not-for-profit nursing homes, 16 found all statistically significant comparisons favouring a higher quality of care in not-for-profit homes. None had all statistically significant analyses favouring higher quality in for-profit homes.

In a meta-analysis of all studies, focusing on the four quality indicators, two analyses showed statistically significant results favouring higher quality in not-for-profit nursing homes – namely, higher quality staffing in not-for-profit homes and lower prevalence of pressure ulcers. The two remaining meta-analyses produced differences failing to reach statistical significance.

Limitations

While this study’s findings are in alignment with two previous systematic reviews (1991 and 2002), research and comparisons of this type are always limited by the nature and quality of the data or studies from which they are drawn. Additionally, assessment of quality using the aggregate measures indicated may not provide a robust or valid assessment of the actual quality of individual patient care delivered, particularly when more specific processes or outcome measures are included. Descriptions of nursing home ownership status may not always be accurate in this type of study and the specificity of the classification may not reflect actual practice and business models, which may be mixed models or otherwise misclassified. Finally, of course, we cannot attribute causality, only association, using observational studies.

Discussion

Although a large proportion of studies showed no significant difference per se in quality of care by ownership status, in the majority of studies the quality of care was found to be lower in most, though not all, for-profit nursing homes. Few studies showed that quality of care was better in for-profit than in not-for-profit facilities. Mixed results in many studies clearly show that a variety of factors besides ownership status influence the nature and the quality of outcomes of care in nursing homes.

Relevance

Nursing homes provide long-term housing, support, and nursing care for elderly and/or disabled patients. Given the increasing age of our population, it is in the best interest of governments, patients, and physicians to understand how quality and appropriateness of care tends to vary depending on whether facilities are operating on a for-profit or not-for-profit basis. Since all investor-owned facilities are, by definition, for-profit, the proliferation of investor-owned facilities, as is occurring in BC, is likely to result in inferior quality of care. The trend toward higher quality of care typical of not-for profit nursing homes ought to promote greater commitment of public funding to these types of facilities or, at the very least, to identify the need for better government regulation in investor-owned, for-profit health care facilities.