Interventions to keep older people out of nursing homes

1 billion people in the world are aged 60 or over, and by 2030 this number will increase by almost 40%. These metrics require attention to different aspects of care, with public health interventions being the culprit to sustain healthy ageing and/or prevent progression to fragile states.

In this context, interventions that keep older people out of nursing homes are extremely important both from the perspective of personal and population health. A recent meta-analysis published in the Journal of the American Geriatrics Society (JAGS) by Joseph Gaugle and colleagues provides a useful synthesis of evidence on this topic based on studies including 203,735 older persons.

The authors found that three types of interventions substantially reduce the transition to nursing homes: specialty geriatrics care, multicomponent interventions, and cognitive stimulation.

The main manuscript does not contain details about the performed interventions, and a reader needs to refer to the Supplementary to know more about the exact individual interventions (or about the lack of this information).

Specialty geriatrics care

Specialty geriatrics care includes a heterogeneous interventions such as inpatient units that offer comprehensive geriatrics assessment and care planning/support, often staffed by multidisciplinary teams of geriatricians, nurses, social workers, and/or other professionals. Overall, they were related to a summary OR 0.77 (95% CI, 0.60–0.99), but demonstrated high heterogeneity across studies (I2?=?79%).

Multicomponent interventions

Multicomponent interventions combined one or more discernible intervention elements into a single program, and demonstrated a summary OR of 0.82 (95%, 0.67–0.99) with high heterogeneity (I2?=?62%). However, the categorization of “multicomponent” was not distinctive in the meta-analysis.

Cognitive stimulation/reminiscence therapy

Cognitive stimulation/reminiscence therapy were related to even lower OR?0.60 (95% CI 0.38–0.96) and were less heterogeneous (I2?=?5.5%), but these results were based only on three studies.

Of note, cognitive stimulation therapy is the only nonpharmacological therapy recommended for the treatment of dementia by the UK National Institute for Health and Clinical Excellence.

Other interventions with a positive effect

The systematic review also obtained publications with other intervention types represented by sporadic literature evidence. Nevertheless, these interventions also demonstrated a substantial effect on preventing attendance to nursing homes: home-based interventions (OR?=?0.85; 95% CI 0.68–1.06; I2?=?63%) and inpatient/discharge management (OR?=?0.81; 95% CI 0.62–1.05; I2?=?26%).

Interventions without proven effect

Many evaluated interventions did not demonstrate a statistically significant effect on preventing institutionalization of older adults. Among them were exercise programmes, caregiver skills-building interventions, web-based tools for care, protein energy supplement, and community-based integration.

Limitations of this (and other) meta-analysis on the topic

As the authors rigorously mentioned (and the rest of the sentence is the exact citation from the paper), almost half of the studies do not consider nursing home admission as a primary or secondary outcome of interest and in some instances it was reported as loss to follow-up or in similar fashion as part of study flow. Although these studies were included in meta-analysis, there likely exist other intervention studies that (a) treated nursing home admission as a reason for loss to follow-up rather than an outcome but (b) did not report data on nursing home admission in their samples.

Moreover, over 75% of the studies included in the meta-analysis did not distinguish between short-term, rehabilitative stays, or long-stay admissions, or between persons entering nursing homes or assisted living facilities. This non-uniformity in the definition of outcomes and merging completely different populations in a single group substantially diminishes the ability to provide definitive recommendations to health authorities, physicians, and citizens.

The meta-analysis and other review articles on this topic pose a dozen of questions for different stakeholders.

Questions to regional health authorities:

1. Which of the interventions with proven effect are implemented at the local level?
2. Whether their experience (positive or negative) from different interventions has been published in the literature?
3. In case of lack of resources to analyse the locally implemented interventions, whether the anonymized data were put in the open access, or at least whether the meta-data about them are available to stimulate further research efforts on this highly relevant topic?
4. Whether health authorities provide funds and education for the analysis of routine clinical practice and real world data that are generated each day but have never been analysed?
5. Who should care about the growing elderly population? The authors indicated that in the USA the availability of geriatricians is almost three times lower than the number required to address the current health needs. With the global number exceeding 1 billion older people and the lack of health care resources in many countries, what should be the effective model to provide better health and prevent age-related complications? Whether community health workers have an appropriate role in our societies?

Questions to researchers:

1. Whether studies including elderly population consistently report functional and cognitive baseline characteristics that allow to adequately group patients in meta-analyses and to adjust the resulting metrics to patient characteristics?
2. Whether anonymized data from individual studies are available in open access, or at least whether the meta-data about them are available in public repositories to facilitate more precise individual-level analyses?
3. Whether enough details are provided in the publications to understand exactly which intervention has been performed, with the major goal to disseminate and scale the experience?

Questions to developers of electronic health records:

1. Whether EHR allows to add custom fields in case they are required by health professionals, and do this via a handy user interface without referring to the EHR developer?
2. Whether EHR database tables and fields are appropriately documented that allow to extract data for a customer-need analysis without referring to the EHR developer?
3. Whether EHR allows to account for multiple (possibly recurring) longitudinal outcomes that are relevant for different analyses?

Questions to publishers:

1. Why studies not following the reporting guidelines are still published in scientific journals that makes the evidence synthesis weaker and frequently impossible to make a conclusion about which interventions to implement?

?? LinkedIn's algorithm responds this way: reacting??, commenting??, or sharing?? this post will help it reach the right person. Let's make this happen, and thank you for supporting these efforts!