Answers are grounded in this paper and cite their sources.
In this issue of the Journal of the American Geriatrics Society, Vass et al.1 report the results of a randomized, controlled trial that evaluated whether an educational intervention for primary care health professionals, added to a preexisting home visitor program, was effective in reducing functional decline in community-living older persons. The authors report a modest reduction in self-reported mobility disability after 3 years, with no significant effect on nursing home admissions or mortality. The benefit of the educational intervention was observed in 80-year-olds but not in 75-year-olds. In addition, no benefit was found in persons who had preexisting mobility disability. These findings do not help to resolve the conflicting results of prior studies that have evaluated the effectiveness of preventive home visits,2–5 because participants in both the intervention and control groups had access to such a program through the Danish health system. Because preventive home visits are not generally available in the United States, many readers of the Journal may question the clinical relevance of the current study. In this commentary, I will highlight three important issues that are at least implicitly raised: the value of geriatric education, the role of preventive health care, and translating research into practice. The current study adds to the relatively scant literature on the effectiveness of strategies to enhance the knowledge, attitudes, and skills of clinicians caring for older persons, particularly those who are frail or at an advanced age. Although the educational intervention was not fully described, key features appear to include training home visitors in the principles of geriatric assessment, with special attention to fall prevention and to tiredness in daily activities as an indicator of incipient disability; enhancing communication between the home visitors and general practitioners; and heightening awareness of general practitioners about the clinical relevance of geriatric syndromes. Although the educational intervention does not appear to be particularly potent, it was implemented in a single-payer system that has embraced continuity of care and preventive home visits as a covered benefit. The authors speculate that attention to early triggers of functional decline, manifested as tiredness in daily activities,6 might explain the benefit of the educational intervention. We have found that restricted activity, defined as staying in bed for at least half a day or cutting down on one's usual activities because of an illness or injury, is strongly associated with functional decline7 and the development of disability8 and that fatigue is the most common reason for restricted activity.9 Similarly, exhaustion has been identified as one of five criteria for frailty, a geriatric syndrome that confers high risk for an array of adverse outcomes.10 It is likely, although not proven, that tiredness, fatigue, and exhaustion are manifestations of the same phenomenon. Regardless, there is now sufficient evidence to suggest that these and other early warning signs of functional decline and disability11,12 should no longer be accepted as benign features of old age.13 Given the inadequate attention paid to geriatrics and interdisciplinary care in professional health education and clinical practice,14 there is ample opportunity to improve the care of older persons through educational efforts such as those evaluated in the current study, along with others that have shown promise, such as academic detailing.15 It is hoped that major initiatives sponsored by the Donald W. Reynolds Foundation and John A. Hartford Foundation14,16–18 will lead to a series of innovative educational programs that can be widely disseminated to strengthen the training of health professionals in the principles and practices of geriatric medicine. Despite the burgeoning evidence base supporting a diverse array of home- and center-based programs for the prevention of functional decline, disability, and falls,19–27 significant financial and organizational barriers have precluded implementation of these programs in most settings. Foremost among these has been the absence of a Medicare benefit for preventive health services other than immunizations and cancer screening, but with passage of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003, these rather meager preventive benefits have been modestly expanded, effective January 2005, to include an initial physical examination, cardiovascular blood screening tests, education, and counseling.28 Also, in lieu of a new benefit for geriatric assessment, a new 3-year demonstration project has been initiated to improve the quality of care and health outcomes for Medicare recipients living with multiple chronic conditions.28 These innovations, coupled with recent policy changes regarding payment for new and expensive treatments and diagnostic tests,29 suggest that a revitalized Centers for Medicare and Medicaid Services may be more receptive to considering the addition of new, cost-effective interventions, including those designed to prevent adverse functional outcomes, as covered benefits. It is well known that the benefit of an intervention that is developed and tested under optimal conditions is usually diminished when the intervention is applied under real-world conditions. This phenomenon underscores the important distinction between efficacy and effectiveness.30 Studies of efficacy typically include a homogeneous and highly motivated sample, which is assembled using stringent inclusion and exclusion criteria. The intervention is highly standardized, is often intensive, and is implemented by well-trained research staff, usually in a single setting. In contrast, studies of effectiveness typically include a broad, heterogeneous sample that is intended to be representative of a defined target population. The intervention is designed to be feasible and readily adaptable, may be implemented by staff with varying levels of expertise, and is often tested in multiple settings. Based on these distinctive features, the trial by Vass et al. should be considered an effectiveness study. The study population included a large, representative sample of community-living persons who were drawn from 34 different municipalities. The intervention was relatively brief, included procedures that could be easily adapted and implemented across different settings, and targeted clinicians from several disciplines. Although effectiveness studies often pose special challenges, they should adhere whenever possible to the same methodological standards that have been established for efficacy studies,31 including ensuring that important prognostic factors are equally balanced between the intervention and control groups. In efficacy studies, this is usually accomplished through randomization of individual participants, but randomizing participants may not always be feasible when studies are conducted in real-world settings.23 In the current study, municipalities were randomized rather than individuals. Although the authors report that there were no major differences in baseline characteristics, participants in the control group had a higher rate of mobility disability (29.3% vs 32.5%) than participants in the intervention group (personal communication with the authors). When randomization is not feasible or cannot be conducted at the individual level, it is essential that such baseline differences be adequately addressed in the statistical analysis. Although desirable, it may not be possible to isolate the effect of the intended intervention in an effectiveness study. The benefit of the educational intervention in the current study, for example, could be attributable to an increase in the number of home visits rather than to an improvement in the quality of the visits. Of course, this problem of attribution can also arise in efficacy trials when an intervention includes multiple components.32 Although the effect size of the educational intervention was modest in relative terms, it could have a notable effect in absolute terms. From a public health perspective, a small effect across a large population will often outweigh a large effect in a small population. Because the authors appropriately analyzed and reported their results according to the intention-to-treat principle, the effect size was likely diminished because only 60% of the study population actually accepted and received at least one preventive home visit. Although prone to bias, evaluating results according to the actual receipt of the intervention, in a secondary set of analyses, could have provided valuable complementary information.33 Remarkably, follow-up data on the primary mobility outcome was obtained from more then 95% of the nondecedents. Bias from differential attrition is reduced when losses to follow-up are minimized. The ultimate goal of preventive home visits and other preventive interventions is to optimize the health, well-being, and functional status of older persons. As history can attest,30 the completion of high-quality efficacy studies is often insufficient for informing health policy and shaping clinical practice. When well designed and executed, effectiveness studies can greatly facilitate the translation of research into practice. The current study should remind us that the job is not done when the results of an efficacy study are reported. Barriers to implementing an evidence-based intervention into clinical practice should be identified and addressed,34,35 and high-quality effectiveness studies should subsequently be conducted. Alternatively, pragmatic studies that combine the best features of efficacy and effectiveness studies, such as a recently completed trial of multidimensional assessment,36 should be considered. By effectively translating research into practice, this valuable information will enhance the likelihood that innovative, evidence-based interventions will be adopted within the healthcare system.
No takes yet. Share an insight, caveat, or question.
Thomas M. Gill (2005) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: