The impact of physical activity and sedentary behaviors on frailty levels

Key points

1. 01Prospective cohort data consistently show a dose-response relationship: more physical activity correlates with lower frailty levels and slower frailty progression.
2. 0212 clinical trials reviewed: 10 of 12 showed frailty was reduced more in exercise groups than in controls; multicomponent interventions combining exercise with nutrition or cognitive training showed the strongest effects.
3. 03In NHANES data (n=2,569), frailty severity was strongly tied to less time in physical activity and more time sedentary: severely frail individuals spent 70% of waking hours sedentary versus 57% in non-frail individuals.
4. 04Prolonged, uninterrupted sedentary bouts (at least 30 min) are independently associated with higher frailty and raised mortality risk in vulnerable-to-severely frail groups, but non-bouted sedentary time was not associated with mortality.
5. 05Only 1-3% of severely frail adults met WHO physical activity guidelines (150 min/week MVPA in 10-min bouts); light-intensity activity may be a more feasible target for the frailest individuals.

How it was conducted

Design

Narrative review of epidemiological and clinical trial evidence combined with novel cross-sectional and survival analysis of NHANES accelerometry data

NHANES sample

2,569 community-dwelling adults aged 50 years or older from the 2003-2004 and 2005-2006 cycles

Frailty measure (NHANES)

46-item frailty index (range 0-1); non-frail 0-0.10 (n=347), vulnerable 0.10-0.20 (n=909), mildly frail 0.20-0.30 (n=682), moderately frail 0.30-0.40 (n=370), severely frail >0.4 (n=261)

Physical activity measurement

Objective accelerometry; cut-points: sedentary 0-100 counts/min, light 101-2019 counts/min, MVPA >=2020 counts/min

Prospective studies reviewed

16 cohort studies (Table 1) examining physical activity and frailty incidence or progression

Intervention trials reviewed

12 clinical trials (Table 2) using exercise as single or multicomponent intervention to prevent or reverse frailty

What they found

• Odds of incident frailty (phenotypic model) reduced by 6% for every additional 1 MET-h/week increase in self-reported physical activity: OR 0.94 (95% CI: 0.92-0.96) (Garcia-Esquinas et al., 2015).
• Sedentary lifestyle versus high activity was associated with OR 9.54 (95% CI: 4.94-18.42) for being frail at follow-up (Wade et al., 2016b).
• Inactive older adults had OR 3.0 (95% CI: 2.3-4.1) for transitioning to a worse frailty state and OR 0.3 (95% CI: 0.2-0.4) for improving frailty state compared with active individuals (Ahmad et al., 2018).
• NHANES data: proportion spending waking hours sedentary increased across frailty groups - non-frail 57%, vulnerable 61%, mildly frail 64%, moderately frail 68%, severely frail 70% (p<0.05 for all comparisons).
• Proportion in prolonged sedentary bouts (>=30 min) increased with frailty: non-frail 15%, vulnerable 17%, mildly frail 19%, moderately frail 21%, severely frail 24% of awake time (p<0.05 for all comparisons).
• Prolonged sedentary bouts associated with higher frailty scores independent of age, sex, demographics, total sedentary time, and MVPA: beta-coefficient 1.14 (SE: 0.30), p<0.001 (Kehler et al., 2018b).
• Replacing 30 min/day of sedentary time with 30 min MVPA was associated with lower frailty score: beta-coefficient -2.46 (95% CI: -3.78 to -1.14), p<0.01 (Manas et al., 2018).
• Replacing 30 min sedentary time with light-intensity physical activity associated with lower odds of phenotypic frailty: OR 0.85 (95% CI: 0.80-0.92), p<0.05 (Nagai et al., 2018b).
• Higher sedentary time was not associated with mortality in the least frail group (HR: 0.90, 95% CI: 0.70-1.14, p=NS) but was associated with mortality in vulnerable (HR: 1.14, 95% CI: 1.01-1.28), mildly frail (HR: 1.31, 95% CI: 1.14-1.50), and moderately-to-severely frail individuals (HR: 1.35, 95% CI: 1.21-1.50) (Theou et al., 2017).
• LIFE-P trial: frailty prevalence at 12-month follow-up was 10.0% in exercise group versus 19.1% in control group (p=0.01); mean difference in frailty criteria -0.48 (95% CI: -0.62 to -0.33) (Cesari et al., 2015).
• In the larger LIFE study (n=1,635), no significant difference in frailty prevalence between intervention and control groups over 3.5-year follow-up (Trombetti et al., 2018).
• Multicomponent intervention (Jang et al., 2018): OR for reducing frailty at 6 months 0.08 (95% CI: 0.03-0.18) and at 12 months 0.06 (95% CI: 0.02-0.16).
• Non-bouted sedentary time was not associated with mortality risk in any frailty group (NHANES Cox-proportional hazards analysis).

Limitations

• Most prospective studies relied on self-reported physical activity, which tends to overestimate true activity levels compared to accelerometry.
• No clinical trials to date have examined the impact of specifically reducing sedentary time on frailty outcomes, limiting causal inference for the sedentary behavior findings.
• Significant variability in exercise type, intensity, duration, and frailty measurement tools across intervention trials makes it difficult to draw firm recommendations for optimal exercise prescription.
• Most intervention trials used the phenotypic frailty model, which captures mainly physical frailty and may not fully reflect multi-system frailty complexity.

Why it matters

For patients

Older adults who move more and sit less, especially avoiding long unbroken sitting periods, are likely to age with lower frailty and better survival odds, and even light activity such as short walks may help those who are already frail.

For clinicians

Exercise interventions (moderate intensity, at least 2-3 times per week for 6 months or longer, combining aerobic and resistance training) reduce frailty in clinical trials, and clinicians should also counsel frail patients to break up prolonged sitting independently of overall activity targets.

For readers

This review synthesizes 16 cohort studies and 12 trials alongside novel NHANES survival data to show that both insufficient activity and prolonged sedentary bouts independently drive frailty and mortality, but no trials have yet tested sedentary-reduction as a standalone frailty intervention.