What are the most effective interventions to improve physical performance in pre-frail and frail adults? It’s surprisingly hard to tell.

As populations age and become frail, evidence-based interventions to improve health outcomes are urgently needed. A newly published systematic review in BMC Geriatrics set out to identify interventions that were most effective in improving physical performance in older adults at increased risk of negative outcomes. In this blog post, the authors of the review discuss why this was harder than expected and what can be done to design better interventions.

Frailty: costs and prevention

There are increasing concerns regarding the pressure on UK and US healthcare systems associated with the rapidly increasing number of frail older adults. Frailty – identified by impaired mobility, loss of balance, muscle strength, and endurance – is associated with increased risk of hospital admission, increased duration of stay, and a need for home care. Unplanned hospital admissions, including of frail older adults, cost the NHS >£13 billion, and the US healthcare economy >$1.1 trillion per year.

Physical activity, even minimal amounts, is associated with numerous protective benefits such as reducing the risk of developing diabetes, stroke, hip fractures, improving mental well-being and delaying physical deterioration in older adults. Evidence suggests that targeted interventions to improve balance and muscle strength can help reduce the risk of falls and subsequent hospital admission.

The goal of our review

The objective of our review was to build on existing work by focusing exclusively on randomized controlled trials aimed at improving physical symptoms of frailty in older adults, in order to allow for greater ease of comparison across studies. In the review, we describe the active components of the interventions to try and identify what works well, for whom, and under what conditions. This might sound like a relatively straightforward task.

It was not easy

We repeatedly came up against methodological issues that compromised our ability to make more definite recommendations and conclusions.

First and foremost was the lack of consensus on the definition of frailty or how best to assess it. We found that participants who were pre-frail, frail, and non-frail were often put into the same group and the intervention success was decided on the group as a whole, rather than the sum of its parts. This means we cannot know with any certainty if the intervention worked as well for those who were severely frail versus those who were less so.

Second, behavior change theories have shown predictive and explanatory power for increasing physical activity, and have been used to develop maintenance plans for continued activity; however, there was little evidence that the interventions used in the studies we found were grounded on any formalized behavior change theory.

The limitations we faced in this review are common problems across many long-term health population research fields.

Moreover, most studies were relatively brief in delivery and very few had longer-term follow-up to examine if behavior change was being maintained. Given the growing demand for service provision on the one hand and the budget cuts across health and social care in the UK on the other, interventions must be able to demonstrate benefits that prove to be cost effective in the long term. So while it appears that physical activity may prevent/reduce physical signs of frailty in the short term, it needs to be maintained and integrated into normal daily life for the benefits to be felt both by the individual and by the health and social care services.


It may surprise some to learn that the limitations we faced in this review are common problems across many long-term health population research fields. Increasingly, there have been growing calls for core measures to be introduced and implemented by healthcare professionals, researchers, and funders. This has proven successful in other conditions such as diabetes and some cancers, and could be helpful in frailty assessment as well.

Other action we can take is to ensure that the future interventions we design are fit for purpose, are grounded in theory, and can promise to deliver real long-term health benefits in terms of frailty prevention or symptom reduction.

View the latest posts on the BMC Series blog homepage