Interventions for Frailty
Given the general impact of frailty on the individual and society, it is clear that interventions, both preventive and therapeutic, are required. Interventional studies where a change in frailty status is the primary outcome are limited. Due to the lack of research with these clear outcomes, it is necessary to consider interventional studies using measurements of an element of frailty such as muscle strength, or walking speed as a primary outcome. The interventions are often specifically exercise and nutrition based, or pharmacological. However some trials use multifactorial interventions and the use of the Comprehensive Geriatric Assessment.
Exercise and Nutritional Based Interventions
When considering exercise and nutritional based interventions in frailty there is conflicting evidence within the available literature. To some extent this may be due to the variability in outcomes and inclusion criteria. Clinical studies have both analysed the independent and combined effect of exercise programmes and nutrition on frailty and the components of frailty.
A recent meta-analysis and four individual systematic reviews have found beneficial evidence of exercise programmes on selected physical and functional ability [80-83]. Improvements have been demonstrated in gait speed [80, 84], balance , performance in activities of daily living (ADL) [80, 85] and the Short Physical Performance Battery (SPPB) . However, there was substantial variability in the method of frailty assessment in these studies, with the meta-analysis including only participants classified as frail according to a recognised frailty definition (eight RCTs, 1068 participants classified as frail according to Fried Frailty Index, Speechley and Tinetti criteria and Edmonton Frail Scale) [80 ] . A recent systematic review reported that exercise programmes appeared to be more effective in studies which did not assess frailty using an operational definition. The authors suggested that the participants in these studies may have been more responsive to exercise due to better baseline health or actually being non-frail . This suggests that exercise may be less effective once clinical frailty is evident. A further review found that although five out of six exercise intervention studies reported benefit in at least one performance measure in ‘frail‘ individuals , the one study reporting no benefit was the only one assessing frailty using an operational definition . This adds evidence that exercise interventions may have no positive effect in operationally defined frail individuals. Another recent review of 19 RCTs that included recruited participants defined as, or considered as, frail reported variable benefits of an exercise programme; evidence of benefit was shown for gait speed and short physical performance battery but no consistent effect on ADL or balance was demonstrated .
Epidemiological studies have shown associations between protein intake and frailty: The InCHIANTI study reported that the risk of frailty (Fried frailty criteria) was twice as high in those in the lowest quintile of protein intake compared with those with higher intakes . These data were supported by a later study of 24,417 participants (frailty-free at baseline) in the Women’s Health Initiative Observational Study which showed that a higher baseline protein intake at baseline was associated with a reduction of frailty risk after 3 years of follow up . A more recent crosssectional study of 194 community dwelling adults older than 75 years reported an association between frailty (Fried frailty criteria) and distribution of daily protein intake  . Frailer individuals did not eat less protein compared with non-frail counterparts (median daily intake 1.07 g kg-1) but they had a more uneven distribution of protein. There are very few studies solely using nutrition as an intervention: one study of 87 older adults classified as frail according to the Fried frailty criteria and taking a commercial supplement containing additional calories, protein, and essential amino acids twice daily for 12 weeks, showed improvements in selected functional outcomes compared with controls: improved physical function and timed get up and go, a stable SPPB score compared with a decrease SPPB score in the control group and attenuated reduction in gait speed when compared with the control group . However, no benefit in muscle strength or function from a protein- energy supplement was reported in an earlier trial that did not use a standardised assessment of frailty . In a further trial participants identified as frail on the basis of physical inactivity and weight loss showed no significant effects on fitness or disability measures . The same study included an exercise-only group and showed benefit in functional performance. One of the major methodological challenges in nutritional supplementation studies with frail adults is attempting to control for baseline nutritional status, weight and body composition, as well as accounting for adherence to supplementations in the real world. This may explain the lack of major effects to date.
Other studies have combined nutrition with physical activity and exercise training . A recent RCT in 62 institutionalised older adults classified as frail using the Fried frailty criteria reported that protein supplementation combined with resistance exercise training increased muscle mass compared with a training-only group . In the Singapore Frailty Intervention Trial, 246 frail older adults were randomised into five groups: 48 received just physical training and 49 received just nutritional supplementation for 6 months. The other three groups received cognitive training, a combined intervention and usual care respectively. At both 6 and 12 month follow-up a significant reduction in frailty scores compared with control was shown for both interventions when delivered individually as well as in combination with cognitive training . A pilot RCT study of 117 pre-frail and frail Taiwanese older adults reports that those randomised to a 3-month combination of nutritional consultation and exercise training showed a short-term beneficial effect on frailty status at 3 months, but not at 6 or 12 months compared with those receiving the interventions individually . The nutritional component in this study however was educational in nature rather than a direct intervention using dietary modification or supplementation. Results from the LIFE-P trial comparing a physical activity intervention with a successful lifestyle (educational) intervention in 424 community dwelling older adults showed a significant difference in frailty (Fried criteria) prevalence after 12 months in the physical activity group .
Read together these reports highlight a major unmet need for trials that examine exercise interventions, and nutritional supplements to specifically improve physical function in frail older adults. With the recent recommendation to increase protein intake in older age , the relative lack of nutrition and exercise intervention study data to improve frailty status is surprising.