Skip to main content
BMJ Open logoLink to BMJ Open
. 2025 Oct 16;15(10):e098789. doi: 10.1136/bmjopen-2025-098789

Handgrip strength and walking pace negatively impact the incidents of diabetes mellitus among people with HIV: a prospective cohort study in China

Qiguo Meng 1,0, Congcong Guo 2,0, Yali Xie 1, Haijun Zhang 1, Xiaoxiao Chen 1, Shanling Wang 1, Tingting Hua 1, Yating Wang 1, Tingting Wang 1, Haijiang Lin 1,*, Na He 3,✉
PMCID: PMC12530415  PMID: 41101959

Abstract

Abstract

Background

We aimed to examine the associations of handgrip strength and walking pace with the risk of incident diabetes mellitus (DM) among people with HIV (PWH).

Design

A cohort study.

Setting

This study was conducted in Taizhou City Center for Disease Control and Prevention in Zhejiang Province, mainland China.

Participants

A total of 1916 participants (mean age 43.3±14.2 years, 77.7% male) without diagnosed or unknown DM at baseline from the Comparative HIV and Aging Research in Taizhou cohort were included.

Outcome measures

The primary outcome was incident DM, defined by fasting glucose >7.0 mmol/L, HbA1c concentration >6.5% or diagnosed with DM by a doctor during the period of follow-up.

Results

During a mean follow-up of 3.2 years, 184 incident DM cases were identified. Overall, there was a significant inverse association between both handgrip strength and walking pace with the risk of incident DM in PWH. Compared with participants with lower handgrip strength (<32.1 and <20.3 kg for males and females) or slower walking pace (<2.1 miles/hour), those with higher handgrip strength (≥32.1 and ≥20.3 kg for males and females) or faster walking pace (≥2.1 miles/hour) had a lower risk of incident DM, the adjusted HR (aHRs) (95% CI were 0.60 (0.42 to 0.85) and 0.59 (0.42 to 0.83), respectively. A higher handgrip strength and a faster walking pace were responsible for 38.0% and 39.7% of DM cases, respectively. Moreover, the lowest risk of incident DM was observed in participants with both higher handgrip strength and faster walking pace (aHR, 0.44; 95% CI 0.25 to 0.76).

Conclusion

Higher handgrip strength or faster walking pace was significantly associated with a reduced risk of incident DM among PWH.

Keywords: Diabetes Mellitus, Type 2; China; HIV & AIDS


STRENGTHS AND LIMITATIONS OF THIS STUDY.

  • The use of restricted cubic splines provides evaluation of non-linear relationships.

  • The use of subgroup analysis and interaction to reveal possible modifications of the association between absolute handgrip strength/walking pace with the risk of incident diabetes mellitus (DM).

  • Despite the comprehensive adjustments for the potential confounders, residual confounding by unmeasured variables cannot be entirely ruled out.

  • A single-timepoint measurement of handgrip strength and walking pace data does not reflect the impact of their dynamic changes on DM.

Introduction

With the advent of antiretroviral therapy (ART), the disparity in life expectancy between people with HIV (PWH) and those without HIV has significantly narrowed, transforming HIV into a manageable chronic condition.1 However, this extended lifespan brings increased susceptibility to chronic diseases, driven in part by persistent inflammation.2 3 Studies have demonstrated that PWH face a significantly higher risk of developing various chronic conditions, such as cardiovascular diseases (CVD), liver and kidney disorders, neurocognitive impairments and metabolic issues like diabetes mellitus (DM).4,7

DM poses a major global public health challenge due to its high rates of morbidity, associated disabilities and mortality.8 In 2021, the International Diabetes Federation estimated that approximately 537 million adults worldwide were affected by DM, a figure projected to rise to 783 million by 2045.9 Moreover, accumulating evidence suggests that PWH bear a disproportionately higher burden of DM compared with the general population.10,12 These findings underscore the urgent need for effective strategies to prevent DM, particularly among PWH.

Recent studies indicate that improved physical abilities, including brisk walking pace and enhanced handgrip strength, were associated with a reduced likelihood of adverse health outcomes, such as CVD, dementia and Parkinson’s disease.13,15 Emerging research also links faster walking pace and higher handgrip strength to a lower risk of developing DM. For instance, slow walking pace has been identified as an independent risk factor for DM,16 while a meta-analysis has demonstrated a negative association between muscular strength and DM incidence.17

From a physiological perspective, handgrip strength and walking pace could reflect distinct mechanisms: grip strength primarily measures muscle contraction power, whereas walking pace captures a combination of strength, balance and coordination. Consequently, the combined assessment of both parameters might provide a more comprehensive insight into health outcomes than considering each individually. However, evidence on the joint impact of handgrip strength and walking pace on DM risk remains limited. To date, only one study has explored this association, indicating that slower walking pace and weaker handgrip strength were associated with a higher DM risk.18 Notably, this study did not evaluate the dose–response relationship, which could provide more granular information. Furthermore, especially among PWH, little is known about this association.

Based on these studies, we hypothesise HIV-infected individuals with higher handgrip strength and faster walking pace have a lower risk of DM, and among those with both high handgrip strength and fast walking pace, the risk is the lowest. In order to verify our hypothesis, we utilised data from the Comparative HIV and Aging Research in Taizhou (CHART) cohort to examine the associations between handgrip strength, walking pace and the risk of incident DM among PWH.

Methods

Study population

Details of the study design and sampling methods of the CHART have been described elsewhere.19,21 In brief, the CHART is an ongoing prospective cohort study of HIV and age-related comorbidities among PWH and HIV-negative adults being done in local Centers for Disease Control and Prevention in Taizhou prefecture, Zhejiang province, China. In 2017–2019 (baseline), 2892 PWH adults aged 18 years or older who live in Taizhou prefecture were recruited and enrolled. During the same period, 5412 HIV-negative individuals were recruited by multistage random sampling according to the frequency matching principle of age (±5 years) and sex 1:2 from six local communities (Yuhuan, Luqiao, Tiantai, Huangyan, Linhai and Sanmen), which were randomly selected using random sampling within each county or district in Taizhou prefecture. In 2020–2022, these participants completed the first follow-up survey (visit 1). Depending on resources, standardised screening for age-related comorbidities will be conducted every 3 years thereafter.

Of the 2892 HIV-positive participants initially recruited, we applied the following eligibility criteria for analysis: (1) inclusion required complete baseline data on handgrip strength and walking pace, plus follow-up DM status; (2) exclusion criteria comprised loss to follow-up, death during study, pre-existing DM at baseline (n=853), missing baseline physical function data (n=41) or unavailable follow-up DM data (n=82). After exclusions, 1916 participants formed the primary analytic cohort (online supplemental Figure 1).

Patient and public involvement

Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Assessment of handgrip strength and walking pace

Handgrip strength was measured by a Camry EH101 electronic hand dynamometer, and its measurement unit used was kilogram. Specifically, participants sat upright, elbow adjacent to their torso and forearm positioned on an arm rest with the thumb facing upward and are asked to squeeze the dynamometer ‘as hard as you can’ using their dominant hand. A single maximum score from participants’ self-reported dominant hand was used for the analysis.22 Continuous and sex-specific quartiles of handgrip strength were used in the main analysis. In addition, normalised handgrip strength, which is defined as handgrip strength (kg)/body mass index (BMI), was used in the sensitivity analysis.23

To objectively quantify walking pace, participants were asked to walk a standard 4.57 m course at their normal, self-selected speed. A trained staff used a stopwatch to record the time taken to complete this distance. This time measurement was subsequently converted into a standardised measure of speed (mph) to facilitate interpretation and comparison.24 Continuous and quartiles of walking pace were used in our analysis.

Assessment of covariates

All covariates were assessed at baseline. Demographics were available through standardised questionnaires that included age (years), sex (male; female), smoking (never or ever smoked; current smoking), drinking alcohol regularly (yes and no), educational levels (primary school and below; junior high school; high school and above), income (<¥5000/year; ¥5000–11000/year; ≥¥11 000/year), physical activities (min/week). Neurocognitive performances were assessed using the Chinese version of the Mini-Mental State Examination (MMSE), a 30-point test with a high sensitivity (90.8%) and specificity (93%) in the detection of dementia in the Chinese populations.25 Depression score was measured using the modified version of the Zung Self-Rating Depression Scale, which consists of 9 items, the total score of the 9 items was multiplied by 1.25.26 Height (m) and weight (kg) were measured by an electronic measuring instrument, and BMI, kg/m2, was calculated by dividing weight by square of standing height. Systolic blood pressure (SBP, mm Hg) and diastolic blood pressure (DBP, mm Hg) were measured by an electronic sphygmomanometer, repeat the measurement twice, each time at an interval of 1–2 min and record the average value of the two readings. High-density lipoprotein cholesterol and low-density lipoprotein cholesterol were detected by an automatic analyser (ADVIA 2400, Siemens Healthcare Diagnostics, Tarrytown, New Jersey). HIV-related variables included CD4+cell count (cells/mm3), information of ART and transmission categories (heterosexual; homosexual; others) were extracted through the national HIV/AIDS Comprehensive Response Information Management.

Study outcome

Fasting glucose was measured by an automatic analyser (ADVIA 2400, Siemens Healthcare Diagnostics, Tarrytown, New Jersey). Haemoglobin A1c (HbA1c) concentration was measured using an automatic HbA1c analyser (TOSOH G8, Yokkaichi, Mie Prefecture, Japan). The outcome of interest was incident DM, defined as fasting glucose >7.0 mmol/L, HbA1c concentration >6.5% or diagnosed with DM by a doctor during the follow-up period.

Statistical analysis

Population characteristics were presented as mean±SD for continuous variables and proportions for categorical variables. Differences in population characteristics by sex-specific quartiles of absolute handgrip strength and quartiles of walking pace were accordingly compared using analysis of variance (ANOVA) or χ2 tests.

A restricted cubic spline function was applied to demonstrate the dose–response relationship of absolute handgrip strength and walking pace with the risk of incident DM. Moreover, the association between absolute handgrip strength and walking pace with the risk of incident DM was estimated using Cox proportional hazards models (HR and 95% CI) without and with adjustments for demographics (age, sex, educational level and income), anthropometric and lifestyle factors (BMI, SBP, DBP, smoking status, alcohol consumption, physical activities, walking pace and handgrip strength), MMSE score, depression score (model 1). The population attributable fraction (PAF) was calculated to estimate the proportion of DM events attributable to the exposures.27 In addition, we also assessed the joint effect of handgrip strength and walking pace on incident DM.

Possible modifications of the association between absolute handgrip strength/walking pace with the risk of incident DM were also assessed for the following variables: sex (male and female), age (median), BMI (<25 and ≥25 kg/m2), smoking status (never or ever smoked and current smoking), drinking alcohol regularly (yes and no), educational levels (primary school and below, junior high school and high school and above), income (<¥5000/year and ≥¥5000/year), regular exercise (yes and no), MMSE scores (median) and depression (yes and no, defined as depression points ≥18). Interactions were examined by including interaction terms into the multivariable models using likelihood ratio testing, with adjustments for all aforementioned confounders.

To assess the robustness of the findings, several sensitivity analyses were conducted. First, handgrip strength (kg)/BMI was calculated to assess the association between normalised handgrip strength and the risk of DM.23 Second, multivariate imputation by chained equation was used to handle the missing values in covariates. Third, the HIV-related covariates including CD4+cell count, time for ART and transmission categories were further adjusted based on model 1.

A two-sided p<0.05 was considered as statistically significant. All statistical analyses were performed using R software (V.4.4.1).

Results

Baseline characteristics of participants

Of the 1916 participants, the mean age was 43.3 (SD, 14.2) years, with 1488 (77.7%) being male. The mean handgrip strength was 38.0 (SD, 8.2) for males and 24.2 (SD, 6.8) kg for females.

The baseline characteristics of the study participants, stratified by sex-specific handgrip strength quartiles, were presented in table 1 and online supplemental Table 1. Participants with higher handgrip strength and faster walking pace were younger, more likely to be heterosexual and exhibited lower depression scores. They also had higher BMI, DBP, educational levels, income levels, MMSE scores and CD4+ cell count, alongside longer durations for ART.

Table 1. Baseline characteristics of people with HIV according to handgrip strength or walking pace.

Sex-specific quartiles of handgrip strength*, kg Q1 Q2 Q3 Q4 P value
N 479 479 479 479
Age, years 48.7±16.4 43.0±14.3 41.6±12.8 39.7±11.2 <0.001
BMI, kg/m2 21.3±3.0 21.7±2.8 22.3±2.9 22.7±3.1 <0.001
SBP, mm Hg 123.6±14.6 124.0±15.1 123.5±14.7 125.4±14.3 0.155
DBP, mm Hg 75.8±10.0 76.2±9.7 76.6±10.0 77.9±10.0 0.010
Smoking status 0.231
 Never or ever smoked 281 (58.7) 308 (64.3) 305 (63.7) 290 (60.5)
 Current smoking 198 (41.3) 171 (35.7) 174 (36.3) 189 (39.5)
Drinking alcohol regularly 226 (47.2) 219 (45.7) 213 (44.5) 217 (45.3) 0.862
Education levels <0.001
 Primary school and below 233 (48.6) 164 (34.2) 156 (32.6) 140 (29.2)
 Junior high school 127 (26.5) 161 (33.6) 182 (38.0) 174 (36.3)
 High school and above 119 (24.8) 154 (32.2) 141 (29.4) 165 (34.4)
Income <0.001
 <¥5000 /year 359 (75.1) 287 (59.9) 279 (58.2) 270 (56.7)
 ¥5000–11 000/year 92 (19.2) 140 (29.2) 144 (30.1) 142 (29.8)
 ≥¥11 000 /year 27 (5.6) 52 (10.9) 56 (11.7) 64 (13.4)
Physical activities, min/week 4.3±7.1 4.9±7.9 4.7±6.7 4.8±6.9 0.601
HDL, mmol/L 1.1±0.3 1.1±0.4 1.1±0.3 1.1±0.3 0.759
LDL, mmol/L 2.5±0.8 2.4±0.7 2.5±0.7 2.4±0.8 0.371
MMSE scores, points 24.7±5.9 27.0±4.2 27.7±3.5 28.2±3.0 <0.001
Depression scores, points 17.8±5.7 17.4±5.9 16.8±5.6 16.0±5.3 <0.001
Time of ART, year 2.4±2.3 2.7±2.4 2.8±2.4 2.9±2.5 0.027
CD4+ cell count, cell/mm3 376.0±223.9 391.5±223.1 437.5±203.5 459.6±260.6 <0.001
Transmission category 0.119
 Heterosexual 344 (71.8) 309 (64.5) 306 (63.9) 307 (64.2)
 Homosexual 128 (26.7) 158 (33.0) 164 (34.2) 162 (33.9)
 Others 7 (1.5) 12 (2.5) 9 (1.9) 9 (1.9)

Values are presented as means±SD or frequencies (%).

*

Sex-specific quartiles of absolute handgrip strength (kg): Q1: <32.1; Q2: 32.1 to<38.0; Q3: 38.0 to<42.5 Q4: ≥42.5 for males; Q1: <20.3; Q2: 20.3 to<24.2; Q3: 24.2 to<28.5 Q4: ≥28.5 for females.

ART, antiretroviral therapy; BMI, body mass index; DBP, diastolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SBP, systolic blood pressure.

Associations of handgrip strength and walking pace with the risk of incident DM

Over a mean follow-up of 3.2 (±0.68) years, 184 (9.6%) incident DM cases occurred.

 Overall, handgrip strength showed a significant inverse association with the risk of incident DM in PWH (figure 1 and table 2). As there was no significant interaction (p for interaction=0.981) for sex, we used sex-specific quartiles of handgrip strength in the following analysis. Compared to participants in the lowest quartile (quartile 1), the adjusted HRs (95% CI) for incident DM among those in quartile 2, 3 and 4 were 0.53 (0.34 to 0.82), 0.72 (0.48 to 1.11) and 0.56 (0.35 to 0.89), respectively (table 2). Based on these results, quartiles 2 through 4 were merged into a single group for further analysis, resulting in an adjusted HR (95% CI) of 0.60 (0.42 to 0.85) when compared with quartile 1 (table 2). Additionally, the PAF (95% CI) for incident DM associated with handgrip strength was −38.0% (−68.8% to 7.3%), indicating that stronger handgrip strength (quartiles 2–4) contributed to reduce the risk of developing DM by 38.0% (table 2).

Figure 1. The dose–response relationship between handgrip strength and walking pace with the risk of incident DM. *Adjusted for age, sex, educational level, income, BMI, SBP, DBP, smoking status, alcohol consumption, physical activities, walking pace/handgrip strength, MMSE score and depression score. BMI, body mass index; DBP, diastolic blood pressure; DM, diabetes mellitus; MMSE, Mini-Mental State Examination scores; SBP, systolic blood pressure.

Figure 1

Table 2. The associations between handgrip strength and walking pace with incident DM.

N Events (rate) Crude model Model 1*
HR (95% CI) P value HR (95% CI) P value
Handgrip strength, kg
Sex-specific quartiles †
 Q1 479 63 (13.2) ref ref
 Q2 479 36 (7.5) 0.53 (0.35 to 0.80) 0.003 0.53 (0.34 to 0.82) 0.004
 Q3 479 44 (9.2) 0.69 (0.47 to 1.02) 0.062 0.72 (0.48 to 1.11) 0.135
 Q4 479 41 (8.6) 0.56 (0.38 to 0.84) 0.004 0.56 (0.35 to 0.89) 0.013
Categories
 Q1 479 63 (13.2) ref ref
 Q2–4 1437 121 (8.4) 0.59 (0.44 to 0.81) <0.001 0.60 (0.42 to 0.85) 0.004
PAF, % −41.7 (−68.2 to 15.1) −38.0 (−68.8 to 7.3)
Walking pace, miles/h
Quartiles
 Q1 (<2.1) 430 59 (13.7) ref ref
 Q2 (2.1–<2.3) 509 34 (6.7) 0.42 (0.27 to 0.64) <0.001 0.48 (0.31 to 0.74) <0.001
 Q3 (2.3–<2.6) 178 19 (10.7) 1.01 (0.60 to 1.69) 0.975 0.86 (0.48 to 1.54) 0.612
 Q4 (≥2.6) 799 72 (9.0) 0.52 (0.37 to 0.74) <0.001 0.63 (0.43 to 0.92) 0.017
Categories
 Q1 (<2.1) 430 59 (13.7) ref ref
 Q2-4 (≥2.1) 1486 125 (8.4) 0.53 (0.39 to 0.72) <0.001 0.59 (0.42 to 0.83) 0.002
PAF, % −54.7 (−84.3 to 25.0) −39.7 (−68.5 to 10.9)
*

Model 1: Adjusted for age, sex, educational level, income, BMI, SBP, DBP, smoking status, alcohol consumption, physical activities, walking pace/handgrip strength, MMSE score and depression score.

†

Sex-specific quartiles of handgrip strength (kg): Q1: <32.1; Q2: 32.1 to <38.0; Q3: 38.0 to <42.5 Q4: ≥42.5 for males; Q1: <20.3; Q2: 20.3 to<24.2; Q3: 24.2 to<28.5 Q4: ≥28.5 for females.

BMI, body mass index; DBP, diastolic blood pressure; DM, diabetes mellitus; MMSE, Mini-Mental State Examination scores; PAF, population attributable fraction; SBP, systolic blood pressure.

There was also a significant inverse association between walking pace with the risk of incident DM in PWH (figure 1 and table 2). Compared with participants in the slowest walking pace quartile (<2.1 miles/hour), the adjusted HR (95% CI) for incident DM among those in quartile 2 (2.1–<2.3 miles/hour), quartile 3 (2.3–<2.6 miles/hour) and quartile 4 (≥2.6 miles/hour) were 0.48 (0.31 to 0.74), 0.86 (0.48 to 1.54) and 0.63 (0.43 to 0.92), respectively (table 2). When quartiles 2 through 4 were further merged into a single group, the adjusted HR (95% CI) was 0.59 (0.42 to 0.83) compared with quartile 1 (table 2). Additionally, the PAF (95% CI) for incident DM associated with walking pace was −39.7% (−68.5% to 10.9%), indicating that faster walking pace (quartiles 2–4) was associated with a 39.7% reduction in the risk of developing DM (table 2).

When handgrip strength and walking pace were analysed jointly, compared with those with both lower handgrip strength (quartile 1) and slower walking pace (quartile 1), the lowest risk of incident DM was observed in participants with both higher handgrip strength (quartile 2–4) and faster walking pace (quartile 2–4) (adjusted HR, 0.44; 95% CI 0.25 to 0.76) (table 3).

Table 3. Joint effect of handgrip strength and walking pace on incident DM.

Handgrip strength, kg* Walking pace, mile/h N Events (rate) Crude model Model 1†
HR (95% CI) P value aHR (95% CI) P value
Q1 Q1 (<2.1) 164 21 (12.8) ref ref
Q2–4 Q1 (<2.1) 266 38 (14.3) 0.92 (0.54 to 1.57) 0.766 0.92 (0.51 to 1.66) 0.789
Q1 Q2–4 (≥2.1) 315 42 (13.3) 0.80 (0.47 to 1.35) 0.402 0.87 (0.50 to 1.52) 0.633
Q2–4 Q2–4 (≥2.1) 1171 83 (7.1) 0.42 (0.26 to 0.68) <0.001 0.44 (0.25 to 0.76) 0.003
*

Sex-specific quartiles of handgrip strength: Q1: <32.1; Q2-4: ≥32.1 for males; Q1: <20.3; Q2-4: ≥20.3 for females.

†

Model 1: Adjusted for age, sex, educational level, income, BMI, SBP, DBP, smoking status, alcohol consumption, physical activities, MMSE score and depression score.

BMI, body mass index; DBP, diastolic blood pressure; DM, diabetes mellitus; MMSE, Mini-Mental State Examination scores; SBP, systolic blood pressure.

In sensitivity analyses, consistent results were observed for normalised grip strength (handgrip strength/BMI) (online supplemental Table 2). Moreover, the findings remained robust after addressing missing covariate values (online supplemental Tables 3 and 4). Further adjustments for the HIV-related covariates also did not materially alter the results (online supplemental Tables 5 and 6).

Stratified analyses by potential effect modifiers

There was a significantly stronger inverse relationship between handgrip strength and the risk of incident DM in the undepressed participants (p value for interactions=0.026) (online supplemental Figure 2).

None of the other variables, including sex, BMI, smoking status, drinking alcohol regularly, educational level, income, regular exercise and MMSE scores, significantly modified the association between handgrip strength and the risk of DM in PWH (online supplemental Figure 2). In addition, no variables were found to modify the association between walking pace with the risk of DM in PWH (online supplemental Figure 3).

Discussion

Consistent with our hypothesis, in this large prospective cohort study focusing on PWH, straightforward indicators such as handgrip strength and walking pace were significantly associated with a lower risk of incident DM, independently of demographics, anthropometric and lifestyle factors as well as HIV-related factors. A higher handgrip strength (≥32.1 kg and ≥20.3 kg for males and females) and a faster walking pace (≥2.1 miles/hour) were attributed to reducing 38.0% and 39.7% of DM cases, respectively. Furthermore, the combination of stronger handgrip strength and faster walking pace demonstrated the greatest reduction in DM risk. In addition, the depression status significantly modified the association between handgrip strength and the risk of DM, highlighting the potential influence of psychological factors in this association.

Several studies have suggested that a stronger handgrip strength or a faster walking pace was associated with a lower risk of incident DM.1617 28,31 In 194 304 White-European participants, those with obesity who walked briskly were at an approximately 10- to 12-fold higher risk of type two diabetes, whereas those with obesity and walked slowly had an approximately 12- to 15-fold higher risk compared with brisk-walking participants with normal BMI.16 In 2623 Chinese middle-aged and older adults, compared with participants in low normalised grip strength, those in high normalised grip strength had decreased odds of progression to diabetes.30 Moreover, one study reported that joint effect of handgrip strength and walking pace on incident DM was slightly lower than the sum of their independent associations in 205 738 participants free of DM using data from the UK Biobank study.18 However, this study only considered classified handgrip strength and walking pace, without exploring the dose–response relationship, which could provide more granular information. In addition, the combined association of handgrip strength and walking pace with the risk of incident DM in PWH remains largely unexplored. Our current study, leveraging data from the CHART cohort to examine these associations in PWH, addresses this critical gap in knowledge and provides timely insights into the interplay of these physical abilities in reducing DM risk.

Several mechanisms may account for these observed associations between advanced muscle fitness and reduced DM risk. First, increased muscle strength and mass were inversely associated with systemic inflammation, mechanistically, skeletal muscle acted as an endocrine organ, secreting anti-inflammatory myokines that suppressed proinflammatory cytokines.32 Greater lean mass exhibits lower concentrations of CRP, IL-6 and Tumor Necrosis Factor (TNF)-α—biomarkers directly implicated in pancreatic β-cell dysfunction and insulin resistance through NF-κB pathway activation.33 This inflammation-muscle relationship may be particularly relevant in HIV populations, where chronic immune activation exacerbates metabolic risk. Second, the metabolic benefits of physical activity extend beyond energy expenditure. Regular muscle contraction stimulates activated protein kinase (AMPK)-mediated glucose transporter 4 (GLUT4) translocation to myocyte membranes, enhancing insulin-independent glucose uptake.34 Concurrently, hepatic insulin sensitivity improves through reduced ectopic lipid deposition, as demonstrated by MRI studies showing 15%–25% decreases in liver fat content following resistance training interventions.35 These adaptations collectively lower fasting glucose and reduce postprandial hyperglycaemia. Third, muscular fitness modulates redox homeostasis through multiple pathways: (1) increased mitochondrial efficiency reduces reactive oxygen species (ROS) production during oxidative phosphorylation, (2) strength-trained muscles upregulate antioxidant enzymes (superoxide dismutase (SOD), catalase) by 2–3 fold, and (3) myokines like irisin activate Nrf2-mediated oxidative stress defenses.36 This is clinically significant given that HIV infection and ART independently increase oxidative stress burden, potentially accelerating diabetes progression through protein kinase C activation and advanced glycation end-product formation.

It is noteworthy that handgrip strength showed a stronger protective effect against DM in participants without depression. This difference may stem from depression’s established links to elevated inflammation and oxidative stress37 38—pathways that overlap with and may partially counteract the metabolic benefits of muscle strength. While these findings suggest depression could attenuate the DM-protective effects of muscular fitness, further research should clarify whether this reflects biological interactions or behavioural factors. Furthermore, previous studies reminded that several clinically relevant factors (age, education, income and physical activity) might be potential effect modifiers,39 while formal interaction tests did not reveal statistically significant modifications (all p interaction >0.05). This suggests that the protective association between higher handgrip strength and faster walking pace with lower diabetes risk may be relatively consistent across these population subgroups in our cohort. However, single-timepoint measurements of covariates like physical activity may not reflect long-term exposure patterns that truly modify DM risk, thus larger studies with repeated measures of these covariates would help clarify whether truly differential effects exist across these subgroups.

Our study has several limitations that should be acknowledged. First, despite the comprehensive adjustments for the potential confounders, residual confounding by unmeasured variables such as hormonal levels and body composition which were associated with sarcopenia and osteoporosis in PWH40 41 cannot be entirely ruled out. Second, as the study population was exclusively drawn from Taizhou, Zhejiang Province, the findings cannot be directly generalised to other regions or nations, and the data cannot be used to estimate disease prevalence and incidence rates on a broader scale. Third, since we only used a single-timepoint measurement of handgrip strength and walking pace data, which does not reflect the impact of their dynamic changes on DM. Thus, further studies in diverse populations are needed to validate and extend our findings.

Conclusion

In conclusion, our study demonstrated that stronger handgrip strength and faster walking pace were significantly associated with a lower risk of incident DM among PWH, a higher handgrip strength and a faster walking pace were responsible for 38.0% and 39.7% of DM cases, respectively. The results hold important implications for clinical practice and public health strategies. Given that walking pace and handgrip strength were parameters that can be easily assessed during clinical consultations and are modifiable through practical interventions, evaluating these attributes could aid in pinpointing individuals at increased risk of developing DM. Consequently, this targeted approach could facilitate the delivery of comprehensive interventions to mitigate the risk of DM onset among PWH.

Supplementary material

online supplemental file 1
bmjopen-15-10-s001.docx (5.3MB, docx)
DOI: 10.1136/bmjopen-2025-098789

Acknowledgements

The authors thank the staff and participants of the CHART study for their important contributions.

Footnotes

Funding: This study was supported by the Natural Science Foundation of China (grant number 82173579), Natural Science Foundation of Zhejiang Province (TGY23H260003), Zhejiang Science and Technology Plan for Disease Prevention and Control (Project Number 2025JK130), Taizhou City Foundation for Talents (TZ2022-2), Shanghai Municipal Health Commission (grant number GWVI-11.1-05; GWVI-11.1-03) and Yi-Wu Research Institute (grant number KCF201512) of Fudan University.

Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-098789 ).

Provenance and peer review: Not commissioned; externally peer-reviewed.

Patient consent for publication: Consent obtained directly from patients.

Ethics approval: This study was conducted in accordance with the Helsinki Declaration and was approved by the Institutional Review Board of the School of Public Health at Fudan University in China (IRB#2017-07-0629-S). Participants gave informed consent to participate in the study before taking part.

Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Data availability statement

No data are available.

References

  • 1.Katz IT, Maughan-Brown B. Improved life expectancy of people living with HIV: who is left behind? Lancet HIV. 2017;4:e324–6. doi: 10.1016/S2352-3018(17)30086-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Deeks SG, Tracy R, Douek DC. Systemic effects of inflammation on health during chronic HIV infection. Immunity. 2013;39:633–45. doi: 10.1016/j.immuni.2013.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Maciel RA, Klück HM, Durand M, et al. Comorbidity is more common and occurs earlier in persons living with HIV than in HIV-uninfected matched controls, aged 50 years and older: A cross-sectional study. Int J Infect Dis. 2018;70:30–5. doi: 10.1016/j.ijid.2018.02.009. [DOI] [PubMed] [Google Scholar]
  • 4.Verheij E, Boyd A, Wit FW, et al. Long-term evolution of comorbidities and their disease burden in individuals with and without HIV as they age: analysis of the prospective AGEhIV cohort study. Lancet HIV. 2023;10:e164–74. doi: 10.1016/S2352-3018(22)00400-3. [DOI] [PubMed] [Google Scholar]
  • 5.Swanepoel CR, Atta MG, D’Agati VD, et al. Kidney disease in the setting of HIV infection: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2018;93:545–59. doi: 10.1016/j.kint.2017.11.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Mitra P, Jain A, Kim K. HIV and AIDS in Older Adults: Neuropsychiatric Changes. Curr Psychiatry Rep. 2022;24:463–8. doi: 10.1007/s11920-022-01354-z. [DOI] [PubMed] [Google Scholar]
  • 7.Wong C, Gange SJ, Moore RD, et al. Multimorbidity Among Persons Living with Human Immunodeficiency Virus in the United States. Clin Infect Dis. 2018;66:1230–8. doi: 10.1093/cid/cix998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.World health organization noncommunicable diseases: key facts. [18-Mar-2022]. https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases Available. Accessed.
  • 9.International Diabetes Federation Diabetes atlas 2021. [18-Mar-2022]. https://diabetesatlas.org/ Available. Accessed. [PubMed]
  • 10.Zhang E, Chai JC, Deik AA, et al. Plasma Lipidomic Profiles and Risk of Diabetes: 2 Prospective Cohorts of HIV-Infected and HIV-Uninfected Individuals. The Journal of Clinical Endocrinology & Metabolism. 2021;106:e999–1010. doi: 10.1210/clinem/dgab011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Nansseu JRN, Bigna JJR, Kaze AD, et al. Rates and drivers of progression to pre-diabetes and diabetes mellitus among HIV-infected adults on antiretroviral therapy: a global systematic review and meta-analysis protocol. BMJ Open. 2016;6:e012852. doi: 10.1136/bmjopen-2016-012852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Sarkar S, Brown TT. Diabetes in People with HIV. Curr Diab Rep. 2021;21:13. doi: 10.1007/s11892-021-01382-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Petermann-Rocha F, Ho FK, Welsh P, et al. Physical capability markers used to define sarcopenia and their association with cardiovascular and respiratory outcomes and all-cause mortality: A prospective study from UK Biobank. Maturitas. 2020;138:69–75. doi: 10.1016/j.maturitas.2020.04.017. [DOI] [PubMed] [Google Scholar]
  • 14.Liu M, He P, Ye Z, et al. Association of handgrip strength and walking pace with incident Parkinson’s disease. J Cachexia Sarcopenia Muscle. 2024;15:198–207. doi: 10.1002/jcsm.13366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.He P, Zhou C, Ye Z, et al. Walking pace, handgrip strength, age, APOE genotypes, and new-onset dementia: the UK Biobank prospective cohort study. Alz Res Therapy. 2023;15 doi: 10.1186/s13195-022-01158-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Boonpor J, Parra-Soto S, Gore J, et al. Association between walking pace and incident type 2 diabetes by adiposity level: A prospective cohort study from the UK Biobank. Diabetes Obes Metab. 2023;25:1900–10. doi: 10.1111/dom.15053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Tarp J, Støle AP, Blond K, et al. Cardiorespiratory fitness, muscular strength and risk of type 2 diabetes: a systematic review and meta-analysis. Diabetologia. 2019;62:1129–42. doi: 10.1007/s00125-019-4867-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Boonpor J, Parra‐Soto S, Petermann‐Rocha F, et al. Combined association of walking pace and grip strength with incident type 2 diabetes. Scandinavian Med Sci Sports . 2022;32:1356–65. doi: 10.1111/sms.14197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Lin H, Ding Y, Ning C, et al. Age-specific associations between HIV infection and carotid artery intima-media thickness in China: a cross-sectional evaluation of baseline data from the CHART cohort. Lancet HIV. 2019;6:e860–8. doi: 10.1016/S2352-3018(19)30263-2. [DOI] [PubMed] [Google Scholar]
  • 20.Shi R, Chen X, Lin H, et al. Association of HIV infection with metabolic syndrome among normal or underweight young adults: evidence from the CHART cohort. Biosci Trends. 2021;14:450–6. doi: 10.5582/bst.2020.03351. [DOI] [PubMed] [Google Scholar]
  • 21.He J, Lin H, Ding Y, et al. Genome-wide associated variants of subclinical atherosclerosis among young people with HIV and gene-environment interactions. J Transl Med. 2022;20 doi: 10.1186/s12967-022-03817-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Yuan S, Chen X, Lin H, et al. Interaction of declined handgrip strength and HIV infection on neurocognitive impairment. J Neurovirol. 2022;28:217–24. doi: 10.1007/s13365-021-01036-1. [DOI] [PubMed] [Google Scholar]
  • 23.Hamasaki H. What can hand grip strength tell us about type 2 diabetes?: mortality, morbidities and risk of diabetes. Expert Rev Endocrinol Metab. 2021;16:237–50. doi: 10.1080/17446651.2021.1967743. [DOI] [PubMed] [Google Scholar]
  • 24.Li J, Chen X, Lin H, et al. Associations between HIV infection and frailty status and its individual components: Are frailty components disproportionally affected? HIV Med. 2023;24:533–43. doi: 10.1111/hiv.13429. [DOI] [PubMed] [Google Scholar]
  • 25.Zhang Z-X, Zahner GEP, Román GC, et al. Socio-demographic variation of dementia subtypes in china: Methodology and results of a prevalence study in Beijing, Chengdu, Shanghai, and Xian. Neuroepidemiology. 2006;27:177–87. doi: 10.1159/000096131. [DOI] [PubMed] [Google Scholar]
  • 26.Dong R, Lin H, Chen X, et al. Gut Microbiota and Fecal Metabolites Associated With Neurocognitive Impairment in HIV-Infected Population. Front Cell Infect Microbiol. 2021;11:723840. doi: 10.3389/fcimb.2021.723840. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Mansournia MA, Altman DG. Population attributable fraction. BMJ. 2018;360:k757. doi: 10.1136/bmj.k757. [DOI] [PubMed] [Google Scholar]
  • 28.Chen S, Yan S, Aiheti N, et al. A bi-directional Mendelian randomization study of sarcopenia-related traits and type 2 diabetes mellitus. Front Endocrinol. 2023;14 doi: 10.3389/fendo.2023.1109800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ye C, Kong L, Wang Y, et al. Causal associations of sarcopenia‐related traits with cardiometabolic disease and Alzheimer’s disease and the mediating role of insulin resistance: A Mendelian randomization study. Aging Cell. 2023;22:e13923. doi: 10.1111/acel.13923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Qiu S, Cai X, Yuan Y, et al. Muscle strength and prediabetes progression and regression in middle-aged and older adults: a prospective cohort study. J Cachexia Sarcopenia Muscle. 2022;13:909–18. doi: 10.1002/jcsm.12905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Vaishya R, Misra A, Vaish A, et al. Hand grip strength as a proposed new vital sign of health: a narrative review of evidences. J Health Popul Nutr. 2024;43:7. doi: 10.1186/s41043-024-00500-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Tuttle CSL, Thang LAN, Maier AB. Markers of inflammation and their association with muscle strength and mass: A systematic review and meta-analysis. Ageing Res Rev. 2020;64:101185. doi: 10.1016/j.arr.2020.101185. [DOI] [PubMed] [Google Scholar]
  • 33.Rohm TV, Meier DT, Olefsky JM, et al. Inflammation in obesity, diabetes, and related disorders. Immunity. 2022;55:31–55. doi: 10.1016/j.immuni.2021.12.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Colberg SR, Sigal RJ, Yardley JE, et al. Physical Activity/Exercise and Diabetes: A Position Statement of the American Diabetes Association. Diabetes Care. 2016;39:2065–79. doi: 10.2337/dc16-1728. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Bacchi E, Negri C, Targher G, et al. Both resistance training and aerobic training reduce hepatic fat content in type 2 diabetic subjects with nonalcoholic fatty liver disease (the RAED2 Randomized Trial) Hepatology. 2013;58:1287–95. doi: 10.1002/hep.26393. [DOI] [PubMed] [Google Scholar]
  • 36.Yuan T, Yang T, Chen H, et al. New insights into oxidative stress and inflammation during diabetes mellitus-accelerated atherosclerosis. Redox Biol. 2019;20:247–60. doi: 10.1016/j.redox.2018.09.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Beurel E, Toups M, Nemeroff CB. The Bidirectional Relationship of Depression and Inflammation: Double Trouble. Neuron. 2020;107:234–56. doi: 10.1016/j.neuron.2020.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Bhatt S, Nagappa AN, Patil CR. Role of oxidative stress in depression. Drug Discov Today. 2020;25:1270–6. doi: 10.1016/j.drudis.2020.05.001. [DOI] [PubMed] [Google Scholar]
  • 39.Hoogendijk EO, Afilalo J, Ensrud KE, et al. Frailty: implications for clinical practice and public health. Lancet. 2019;394:1365–75. doi: 10.1016/S0140-6736(19)31786-6. [DOI] [PubMed] [Google Scholar]
  • 40.De Vincentis S, Greco C, Fanelli F, et al. Sarcopenic obesity and reduced BMD in young men living with HIV: body composition and sex steroids interplay. J Endocrinol Invest. 2024;47:2715–30. doi: 10.1007/s40618-024-02375-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.De Vincentis S, Decaroli MC, Fanelli F, et al. Health status is related to testosterone, estrone and body fat: moving to functional hypogonadism in adult men with HIV. Eur J Endocrinol. 2021;184:107–22. doi: 10.1530/EJE-20-0855. [DOI] [PubMed] [Google Scholar]

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    online supplemental file 1
    bmjopen-15-10-s001.docx (5.3MB, docx)
    DOI: 10.1136/bmjopen-2025-098789

    Data Availability Statement

    No data are available.


    Articles from BMJ Open are provided here courtesy of BMJ Publishing Group

    RESOURCES