Do not-for-profit nursing homes provide better quality?

In the linked systematic review (doi:10.1136/bmj.b2732), Comondore and colleagues assess the relation between profit status and the quality of care in nursing homes and conduct a meta-analysis of four quality measures.1 The association between profit status and the quality of health care has been controversial for decades. The controversy stems partly from theoretical ambiguity and partly from lack of definitive empirical evidence. In theory, not-for-profit healthcare providers may provide a higher quality of care because their mission might include quality and because they do not need to divert resources to shareholders and taxes. On the other hand, for-profit providers may feel greater pressure to compete on price and quality, and this may result in higher quality care that is also more efficient. Unfortunately, rigorous testing of these competing theories is limited because it is impractical to conduct randomised controlled trials of profit status. Methods to mimic randomisation in observational studies (for example, instrumental variables2) are not always viable. Thus, empirical evidence cannot determine with certainty which theory is closer to the truth.

The nursing home sector is no exception to this controversy. Comondore and colleagues report that existing studies predominantly favour not-for-profit nursing homes, in that 40 of 82 studies showed significantly better quality in non-profit making homes. The meta-analysis favoured such homes for two of four outcomes; differences between the other outcomes were not significant. An important benefit of this review is that the authors were able to compare these results with similar reviews they conducted on hospitals and dialysis centres, both of which also showed higher quality in not-for-profit organisations.3 4 What is most remarkable about the results, however, is the inconsistency of the findings. Three of the studies reviewed showed significantly better quality in for-profit homes, and the remaining 39—almost half—were equivocal.

The inconsistency of the findings probably reflects the challenges of using observational data. No review or meta-analysis can overcome the empirical limitations common to all the studies reviewed—we still do not know whether not-for-profit status is the reason for higher quality of care. In other words, if a for-profit nursing home became not-for-profit, would its quality improve? The authors note this caveat, but it is worthy of greater consideration in terms of practical implications.

The lack of causal evidence is particularly problematic in that most of the studies were conducted in the United States, where the relation between nursing home profit status and payer mix is unique among healthcare sectors. Whereas not-for-profit status is normally associated with a community oriented mission, including care for the indigent, which would justify the tax exemption status, not-for-profit nursing homes in the US tend to focus on the clinically more severe and financially more lucrative end of the payer spectrum.5 For-profit facilities usually have a less lucrative payer mix and take on a larger proportion of (indigent) Medicaid beneficiaries. Thus, it could be argued that differences in quality stem from differential revenues rather than mission or diversion of resources to shareholders. Indeed, not-for-profit nursing homes with large Medicaid populations often provide a similar level of quality to that of for-profit homes.6 7 It could be that inadequate risk adjustment confounds the association, because for-profit and not-for-profit nursing homes tend to have different populations in terms of clinical needs.

The degree to which this uncertainty over causality matters depends largely on perspective. To prospective nursing home residents, their families, and care providers concerned about placement in a high quality nursing home, not-for-profit status may act as an indicator of high quality care. The reasons for higher quality in not-for-profit homes are largely irrelevant. Consumers already seem to be guided by this indicator,5 8 although it should be supplemented with personal experience in visiting the home and monitoring care.

From a policy perspective, the uncertainty over causality matters a great deal. It is not at all clear, for example, that banning for-profit providers from the nursing home sector would raise quality. It might, but many factors other than profit status have been strongly linked to the quality of nursing home care, such as the proportion of residents on Medicaid and the extent of poverty in the surrounding neighbourhood.7 Facilities that change profit status will probably maintain these other characteristics. Thus, if differences in quality between for-profit and not-for-profit nursing homes stem at least in part from differences in revenues rather than mission, eliminating for-profit homes may do little to eliminate differences in quality.

Experimental data—data from a situation in which nursing homes are forced to change profit status—are needed to increase our understanding of the causal association between profit status and the quality of nursing homes. Comondore and colleagues note that many European countries with historically public, not-for-profit, healthcare systems are now considering privatisation. Although current evidence is too limited to inform the potential effect of such a policy change on quality of care, the policy change itself could provide useful experimental data.