Infectious Disease · Medical Literature
Frailty phenotype transitions and functional improvements during a supervised exercise trial in older people with HIV: results from the HEALTH Trial.
Abstract
Frailty and sarcopenia contribute to functional decline in older people with HIV (PWH), yet intervention data remain limited. We evaluated changes in frailty phenotype status, sarcopenia-related outcomes and functional performance during a supervised exercise trial and assessed associations between baseline frailty, study withdrawal and intervention response. The High-Intensity Exercise to Attenuate Limitations and Train Habits in Older Adults with HIV (HEALTH) study randomised sedentary PWH aged ≥50 years to 16 weeks of supervised high-intensity interval training (HIIT) or continuous moderate exercise (CME), both combined with progressive resistance training. Frailty was assessed using Fried's phenotype; sarcopenia using current consensus definitions and exploratory HIV-specific cut-points. Functional outcomes included 400-m walk performance and fatigue. Of 118 participants (median age 58 years; 85% male), 94 completed the intervention. Among completers, pre-frailty/frailty status decreased from 48.9% to 30.9% (P < .01), largely reflecting improvements in exhaustion and low activity, with no significant differences between HIIT and CME. Sarcopenia prevalence was low at baseline and changed minimally across definitions. Participants with baseline pre-frailty/frailty were more likely to withdraw (P = .03), yet among retained participants demonstrated greater improvements in 400-m walk performance than non-frail participants (-7.1% [95%CI -8.7, -5.4] vs -4.6% [95% CI -6.3, -2.8]). Fatigue improved among participants with baseline pre-frailty/frailty (-3.3 points [95% CI -5.7, -0.9]) but not in non-frail participants (-1.0 points [95% CI -3.4, 1.4]). During this supervised exercise trial, favourable frailty phenotype transitions and functional improvements were observed among older PWH, particularly in participants with baseline pre-frailty/frailty. Low sarcopenia prevalence limited conclusions regarding categorical sarcopenia outcomes. Strategies to improve retention among more vulnerable participants may enhance intervention reach and impact.
Clinical relevance
Among sedentary older people with HIV (PWH) aged ≥50 years, does a 16-week supervised exercise program (HIIT or CME combined with progressive resistance training) produce favorable frailty phenotype transitions, sarcopenia improvements, and functional gains, and does baseline frailty status predict withdrawal or differential intervention response? __SNAPSHOT__ {"studyDesign":"Randomized controlled trial (parallel-group, two-arm superiority design); secondary/sub-group analysis of frailty, sarcopenia, and functional outcomes from the HEALTH Trial (NCT04550676)","setting":"Two academic HIV care centers in the United States: University of Colorado Anschutz Medical Campus (Aurora, CO) and University of Washington (Seattle, WA)","sampleSize":"n=118 randomized (HIIT: n=57, CME: n=54 per the primary HEALTH trial analysis); 94 completed the 16-week intervention","population":"Sedentary PWH aged ≥50 years on stable antiretroviral therapy (ART), self-reporting fatigue, and virologically suppressed (HIV-1 RNA <200 copies/mL); median age 58 years; 85% male","intervention":"16 weeks of supervised high-intensity interval training (HIIT): 5 high-intensity bouts at 90% heart rate reserve (HRR) for 4 minutes alternating with 4 moderate-intensity bouts at 50% HRR for 3 minutes, plus progressive resistance training 3 days/week","comparator":"16 weeks of supervised continuous moderate exercise (CME): continuous walking or jogging at 60% HRR for 42 minutes, plus progressive resistance training 3 days/week","primaryOutcome":"Changes in frailty phenotype status (Fried criteria), sarcopenia-related outcomes (standard consensus definitions and exploratory HIV-specific cut-points), 400-m walk performance, and fatigue; associations between baseline frailty and study withdrawal or intervention response","followUp":"16 weeks (intervention phase); long-term durability of frailty transitions not assessed","fundingSource":"Not reported in abstract; the HEALTH trial principal investigator (K.M.E.) has disclosed grant funding from Gilead Sciences and consultancy for ViiV, Merck, and Gilead (paid to the University of Colorado)"}
Key findings
Among the 94 participants who completed the 16-week intervention, the prevalence of pre-frailty or frailty decreased from 48.9% to 30.9% (absolute reduction: 18.0 percentage points [calculated]; p<0.01), with improvements driven largely by reductions in exhaustion and low physical activity components of Fried's phenotype. There were no statistically significant differences in frailty phenotype transitions between the HIIT and CME arms. Sarcopenia prevalence was low at baseline and changed minimally regardless of which consensus definition or HIV-specific cut-points were applied. Regarding functional performance, participants with baseline pre-frailty or frailty who completed the trial demonstrated greater improvements in 400-m walk time than non-frail completers (−7.1% [95% CI −8.7, −5.4] vs. −4.6% [95% CI −6.3, −2.8]). Fatigue improved meaningfully among pre-frail/frail completers (−3.3 points [95% CI −5.7, −0.9]) but not among non-frail participants (−1.0 points [95% CI −3.4, 1.4]), whose confidence interval crossed zero. However, participants classified as pre-frail or frail at baseline were significantly more likely to withdraw from the study before completing the intervention (p=0.03), indicating that the most vulnerable individuals—who appeared to benefit most—were also least likely to be retained. NNT for frailty phenotype improvement is not calculable from the abstract (no comparator arm without exercise). No clinically important adverse events were specifically reported in the abstract.
Practice implications
This study provides encouraging evidence that a structured, supervised exercise program—either HIIT or CME combined with resistance training—can reduce the burden of pre-frailty and frailty in older PWH over 16 weeks, with functional and fatigue benefits appearing most pronounced in those with baseline frailty. Critically, the two exercise modalities produced equivalent frailty and functional outcomes, supporting patient-preference-driven exercise prescription rather than mandating high-intensity formats. However, several important caveats limit generalizability: (1) Only abstract-level data are available to this reviewer, which precludes full assessment of methods, statistical models, and covariate adjustment. (2) The 20% dropout rate (118 randomized, 94 completers) and the finding that frail participants were more likely to withdraw introduces significant attrition bias—those who stood to gain most are underrepresented among completers, likely inflating observed benefit. (3) The study population was 85% male, predominantly non-Hispanic white based on the primary trial demographics, and required stable ART and virologic suppression, limiting applicability to women, virologically unsuppressed patients, and more diverse HIV populations. (4) No non-exercise comparator arm exists within this analysis, so improvements cannot be fully disentangled from regression to the mean, seasonal effects, or the Hawthorne effect. (5) Sarcopenia prevalence was low at baseline, constraining the ability to detect sarcopenia-related changes. (6) Long-term durability of frailty phenotype transitions was not assessed. This trial does not establish superiority of either exercise modality for frailty outcomes and does not confirm that improvements are sustained beyond the supervised intervention period. __LIMITATIONS__ 1. Attrition bias: 20% of randomized participants (24/118) did not complete the intervention, and pre-frail/frail individuals—the subgroup with the greatest apparent benefit—were significantly more likely to withdraw (p=0.03), potentially overestimating observed functional gains among completers. 2. No non-exercise control arm: frailty improvements cannot be attributed exclusively to the intervention without ruling out regression to the mean or spontaneous improvement. 3. Only abstract available: full statistical modeling, covariate adjustment strategy, and sensitivity analyses cannot be assessed. 4. Demographic homogeneity: 85% male and conducted at two US academic centers with predominantly virologically suppressed patients on stable ART, limiting generalizability to women, patients with complex ART regimens, or underserved populations. 5. Low baseline sarcopenia prevalence: limits power to detect sarcopenia-related changes, potentially due to non-representative sampling or definition used. 6. Short follow-up (16 weeks): durability of frailty phenotype transitions and functional improvements after the supervised program ends is unknown. 7. Surrogate/functional endpoints: frailty phenotype categorization and 400-m walk improvement are clinically meaningful but do not establish downstream impact on mortality, hospitalization, or AIDS-related events. 8. Industry financial conflicts of interest in the primary HEALTH trial team may warrant scrutiny of conclusions, though the direction of potential bias is unclear. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ A 16-week supervised exercise program—either HIIT or CME combined with resistance training—produced a clinically meaningful reduction in pre-frailty and frailty prevalence (~18 absolute percentage points, p<0.01) and superior functional and fatigue improvements in older PWH who were pre-frail or frail at baseline, with no significant difference between exercise modalities. However, pre-frail and frail patients were more likely to drop out, introducing attrition bias that may inflate observed benefits. These findings support incorporating supervised, structured exercise into routine care for older PWH with pre-frailty or frailty, while underscoring the urgent clinical challenge of retaining the most vulnerable patients in exercise programs.