Associations of the volume and proportion of vigorous-intensity physical activity with all-cause, cardiovascular, and cancer mortality: a systematic review and meta-analysis

Introduction

The role of physical activity intensity in promoting human health has aroused great attention (Ekelund et al., 2023; Fosstveit et al., 2024). Recent observational studies utilizing accelerometers to record the amount and intensity of physical activities have demonstrated the remarkable effectiveness of vigorous-intensity physical activity (VPA) in health promotion (García-Hermoso et al., 2021). For instance, as little as 4.4 min of VPA per day was associated with a 26% to 30% reduction in all-cause mortality risk and a 24% to 32% reduction in cardiovascular disease (CVD) mortality risk, in a sample of non-exercisers from the UK Biobank (n = 25,241) (Stamatakis et al., 2022). Similarly, in another UK Biobank subsample (n = 22,398), a median of 4.5 min of daily VPA was associated with a 31% reduced risk of cancer related to lack of physical activity (Stamatakis et al., 2023). These findings highlight that even short durations of VPA can provide substantial health benefits.

While with potentially greater variability, self-reported data allows large-scale investigations that include diverse populations, offering insights not easily captured by accelerometer-based research. Overall, individuals with self-reported daily VPA at low, moderate, or high volumes demonstrated a greater reduction in all-cause and cause-specific mortality risks compared to those engaging only in light-to-moderate intensity physical activities (LMPA) (Ding et al., 2022; Wang et al., 2021; Zhao et al., 2019). However, previous studies may have overlooked critical issues. First, the volume of VPA varies across studies, whether data were recorded using accelerometers or self-reports, making it unclear what the optimal volume of VPA is for reducing mortality risks. Second, whether there is a dose–response relationship between the volume of VPA and mortality risk is unclear. To address these issues, a systematic review and meta-analysis to examine the association between the VPA volume and these mortality risks, and the potential dose–response relationship between them is needed. This would also help determine the amount of VPA required to achieve substantial mortality risk reduction and identify whether an upper limit exists for causing health benefits.

In practice, people usually engage in both VPA and moderate-intensity physical activity (MPA) during one session. The World Health Organization (WHO) physical activity guidelines recommend adults accumulate at least 150 min of MPA per week, or 75 min of VPA per week, or an equivalent combination (where 1 min of VPA is considered equivalent to 2 min of MPA), specifying the required amount of moderate-to-vigorous intensity physical activity (MVPA) (Bull et al., 2020). These three activity patterns (MPA, VPA, and MVPA) can be considered as different proportions of VPA to MVPA, corresponding to 0% VPA, 100% VPA, and any other proportions of VPA between these two extremes. By adjusting for the total volume of MVPA, previous studies reported that a higher proportion of self-reported VPA to MVPA was beneficial for reducing all-cause and cause-specific mortality risks, but this might stand only within a certain range (Lee et al., 2022; Lopez et al., 2019; Mu et al., 2022). Beyond this specific range, increasing the proportion of VPA may lead to diminishing health benefits, while controversy exists regarding this specific range (Lee et al., 2022; Lopez et al., 2019; Mu et al., 2022; Rey Lopez et al., 2020). A dose–response meta-analysis of current large-scale observational studies is essential to further examine the relationship between the proportion of VPA (relative to MVPA) and these mortality risks, as well as to determine the optimal VPA proportion associated with the lowest mortality risks.

Therefore, the purpose of this systematic review and meta-analysis is twofold: (1) to assess the association between the volume of VPA and the risks of all-cause, CVD, and cancer mortality; (2) to explore the dose–response relationship between the proportion of VPA (relative to MVPA) and the risks of all-cause, CVD, and cancer mortality.

Methods

We registered the study protocol with PROSPERO (CRD42024525067), and the reporting adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (Liberati et al., 2009). Given that the included studies were observational and based on large sample sizes or nationally representative populations, we also followed the Meta-analysis of Observational Studies in Epidemiology (MOOSE) (Stroup et al., 2000) guidelines.

Results

Study screening

After starting with 1,922 studies, 694 were excluded due to duplication, and 886 were excluded based on title and abstract screening. Full texts were retrieved for 342 studies with the potential for inclusion. Twenty studies were finally included in the analysis (Arem et al., 2015; Ding et al., 2022; Duarte Jr et al., 2024; Evenson, Wen & Herring, 2016; Gebel et al., 2015; Kikuchi et al., 2018; Kushi et al., 1997; Lahti et al., 2014; Lee et al., 2022; Lee, Hsieh & Paffenbarger, 1995; Lee & Paffenbarger, 2000; Leitzmann et al., 2007; Lopez et al., 2019; Mu et al., 2022; Rockhill et al., 2001; Shiroma et al., 2014; Wang et al., 2021; Wen et al., 2011; Yu et al., 2003; Zhao et al., 2019). A PRISMA flow chart presenting this progress is shown in Fig. 1 (Liberati et al., 2009).

Meta-analysis

The volume of VPA and all-cause, CVD, and cancer mortality risks

Dose–response analyses.

A curvilinear inverse relationship was shown between the volume of VPA and the risks of all-cause, CVD, and cancer mortality (P for non-linearity < 0.001) (Fig. 2A, Table S7). With 0 min of VPA per week as the reference, any amount of VPA was associated with a lower risk of all-cause, CVD, and cancer mortality, with substantial mortality risk reduction occurring at 180 min per week (all-cause mortality HR = 0.78, 95% CI [0.76–0.79], I² = 88.46%; CVD mortality HR = 0.77, 95% CI [0.73–0.80], I² = 70.13%; cancer mortality HR = 0.86, 95% CI [0.82–0.90], I² = 46.64%). Further increase in the volume of VPA beyond 180 min per week yielded only modest additional benefits, and reduced benefits were observed when VPA volume exceeded 450 min per week in the dose–response curves for all-cause and CVD mortality risk. Funnel plots for the HRs related to all-cause, CVD, and cancer mortality risks, plotted against the standard error of the log (HR), were asymmetric (Fig. S2). Egger’s test yielded P-values of 0.037, 0.063, and 0.094 for these risks, indicating potential publication bias (Lau et al., 2006; Terrin et al., 2003). The decorrelated residuals versus exposure plots showed a less adequate fit at >400 min of VPA per week for all-cause and cancer mortality risk models (Fig. S3).

The highest volume of VPA vs. LMPA (no VPA).

Eight studies were included to compare the effects of the highest volume (the VPA volume was categorized into different levels) of VPA versus LMPA on mortality risks (Fig. 2B). Engaging in the highest volume of VPA was associated with reduced risks of all-cause (HR = 0.75, 95% CI [0.70–0.80], I² = 77.3%), CVD (HR = 0.71, 95% CI [0.63–0.81], I² = 64.8%), and cancer mortality (HR = 0.85, 95% CI [0.79–0.92], I² = 46.8%) compared to LMPA (Fig. 2B). Even though Egger’s test yielded P-values of 0.119, 0.082, and 0.221, the asymmetry observed in the funnel plot (Figs. S4–S6) suggests that publication bias could not be conclusively ruled out (Lau et al., 2006; Terrin et al., 2003). Sensitivity analyses were performed by removing one study at a time (Figs. S7–S9). Subgroup analyses were conducted post hoc based on sensitivity analyses and variations in study-level characteristics identified as potential contributors to heterogeneity, such as methodological differences and population variations. The subgroup analyses reduced heterogeneity and yielded consistent results with the overall study findings, with HR of 0.76 (95% CI [0.73–0.80], I² = 26.3%) for all-cause mortality risk, 0.73 (95% CI [0.68–0.79], I² = 0%) for CVD mortality risk, and 0.88 (95% CI [0.83–0.93], I² = 10.3%) for cancer mortality risk, respectively (Figs. S10–S12).

The proportion of VPA (relative to MVPA) and all-cause and CVD mortality risks

Dose–response analyses.

A U-shaped inverse relationship between the proportion of VPA (relative to MVPA) and all-cause as well as CVD mortality risk (P for non-linearity < 0.001) was observed (Fig. 3A and Table S8). Using 0% of VPA to MVPA as the reference level, the lowest HRs were observed when VPA comprised 37.5% of MVPA for all-cause (HR = 0.90, 95% CI [0.88–0.93]) and CVD mortality risk (HR = 0.88, 95% CI [0.83–0.94]), with only minor variations in HRs within 30–60% of VPA to MVPA. However, a further higher proportion of VPA beyond 60% was associated with diminished effectiveness in reducing mortality risk; and at the right end of the shapes, engaging in VPA alone (100% VPA to MVPA) showed significantly reduced benefits for all-cause (HR = 0.95, 95% CI [0.90–1.01]) and CVD mortality risks (HR = 0.97, 95% CI [0.87–1.08]). In contrast, the optimal combination of VPA and MPA, which was 37.5% of VPA to MVPA, produced better outcomes with a 10% reduction in all-cause and a 12% reduction in CVD mortality risk compared to MPA alone, and with a 5% reduction in all-cause and a 9% reduction in CVD mortality risk compared to VPA alone.

There was a relatively high between-study variance (I² = 82.4% for all-cause mortality risk and I² = 70.7% for CVD mortality risk) in this analysis. The funnel plot was symmetric, with P-values of 0.512 and 0.499 from Egger’s test for all-cause and CVD mortality risk, respectively (Fig. S13). The decorrelated residuals versus the proportion of VPA to MVPA plot showed a good fit between the residuals and exposure for the all-cause and CVD mortality risk models (Fig. S14). Additionally, two studies (Lopez et al., 2019; Wang et al., 2021) that assessed the proportion of VPA (relative to MVPA) and cancer mortality risk found that the risk initially decreased but subsequently increased as the proportion rose, suggesting that the optimal benefit was confined to a specific range.

The highest proportion of VPA to MVPA vs. the lowest.

We calculated the HRs for the highest versus lowest proportion of VPA (to MVPA) for all-cause and CVD mortality (Fig. 3B). Engaging in the highest proportion of VPA was associated with a reduced risk of all-cause (HR = 0.93, 95% CI [0.85–1.01], I² = 88.3%) and CVD mortality risk (HR = 0.90, 95% CI [0.79–1.02], I² = 64.8%). However, a substantial between-study variance was observed, with the midpoint of the range of the highest proportion of VPA (relative to MVPA) varying significantly across original studies (ranging from 0.65 to 1), which might have impacted the statistical power of the analysis. Funnel plots for both all-cause and CVD mortality risk suggest no publication bias (Fig. S15), further supported by Egger’s test (P = 0.086 and 0.083, respectively).

Discussion

This study is the first to investigate the association of VPA volume and proportion (relative to MVPA) with health outcomes by comprehensively analyzing results from large-scale studies with long-term follow-up. For the first purpose, we systematically evaluated the relationship between the volume of VPA and the risks of all-cause, CVD, and cancer mortality, and compared the effects of the highest volume (when the VPA volume was categorized into different levels) of VPA versus LMPA (no VPA). The results revealed a non-linear relationship between the volume of VPA and the mortality risks, with even a small amount of VPA associated with significant mortality risk reduction. Notably, substantial benefits were observed at 180 min of VPA per week, associated with a 22%, 23%, and 14% reduction in all-cause, CVD, and cancer mortality risk, respectively. A notable decline in benefits was observed when VPA volume exceeded 450 min per week in the dose–response curves for all-cause and CVD mortality risk. However, limited data on VPA exposure beyond 400 min per week may weaken the statistical power of this finding. Moreover, the reductions in all-cause and CVD mortality observed at the highest VPA volumes vs. LMPA (no VPA) were greater than those estimated from the dose–response analysis (180 min of VPA per week), with reductions of 25% vs. 22% for all-cause mortality, and 29% vs. 23% for CVD mortality. For these reasons, we can’t rule out the effectiveness of VPA volume exceeding 450 min in reducing the mortality risks. However, it is clear that beyond 180 min of VPA per week, the additional benefits are modest, accompanied by increasing uncertainty, as reflected by the wide confidence intervals. These findings suggest that individuals can incorporate 180 min or more of VPA per week to efficiently minimize the risk of all-cause, CVD, and cancer mortality.

A couple of studies have reported supportive results to our findings. A large-scale study involving 136,766 participants across 21 countries found that, at each level of the total volume of physical activity, participants engaging in VPA exhibited lower CVD mortality risk compared to those without engaging in VPA (Lear et al., 2023). Another study found that approximately 30–35 min of VPA per week and 500 min of MVPA per week provided comparable benefits in mitigating the association between abdominal obesity and incident cardiovascular disease (Sanchez-Lastra et al., 2024). Together with the findings of the present study, the crucial role of engaging in VPA for efficient health promotion is highlighted, which extends beyond merely considering the volume of physical activity. In addition, results from the original studies have generally indicated that the greatest reduction in all-cause and CVD mortality risk was achieved when participants engaged in 150–300 min of MPA and at least 150 min of VPA per week (Lee et al., 2022; Mu et al., 2022; Rockhill et al., 2001; Wang et al., 2021), which aligns with our results demonstrating a substantial reduction in mortality risks associated with VPA volume exceeding 180 min per week. Collectively, these findings underscore the importance of VPA and its volume in maximizing health benefits.

Another important finding of the present study is the association between the proportion of VPA (relative to MVPA) and the mortality risks. WHO physical activity guidelines propose three intensity patterns (MPA, VPA, and MVPA), with 1 min of VPA equivalent to 2 min of MPA for daily accumulation of physical activity (Bull et al., 2020). Based on this principle, we systematically reviewed studies examining the association between the proportion of VPA (relative to MVPA) and mortality risks, and observed a U-shaped inverse relationship between the proportion of VPA (relative to MVPA) and all-cause and CVD mortality risk. Our findings also demonstrate that engaging in MVPA, with VPA comprising 30–60% (of the total MVPA), is more effective in reducing all-cause and CVD mortality risks compared with MPA alone (0% VPA to MVPA) or VPA alone (100% VPA to MVPA).

Considering the fact that achieving a 100% VPA proportion within the total MVPA was usually not common, we summarized HRs for the highest VPA proportions (not necessarily 100%) versus the lowest (MPA only). The summary HRs for all-cause mortality (HR = 0.93, 95% CI [0.85–1.01]) and CVD mortality (HR = 0.90, 95% CI [0.79–1.02]) showed wide confidence intervals, indicating uncertainty regarding their protective effects. Furthermore, in the original studies, although the highest VPA proportions ranging from 65% to 100% showed significant variability in reducing mortality risks, they were generally associated with reduced benefits compared to the optimal VPA proportion in each study. Together with the results in the present study, incorporating 30–60% VPA in MVPA appears optimal for reducing all-cause and CVD mortality risks, while a further increase in the proportion of VPA shows reduced benefits. This finding suggests that practitioners advocate a balanced approach to physical activity, incorporating 30–60% VPA in daily MVPA to efficiently lower the risks of all-cause and CVD mortality.

The 2020 WHO guidelines recommend that adults engage in at least 150–300 min MPA, or at least 75–150 min of VPA, or an equivalent combination of the two per week for substantial health benefits (Bull et al., 2020). The guidelines also note that exceeding 300 min MPA, or 150 min of VPA, or the equivalent combination, can provide increasing general health benefits (Bull et al., 2020). Our findings that 180 min of VPA per week for individuals engaging in only LMPA is associated with substantial reductions in all-cause, CVD, and cancer mortality risks (with only modest additional benefits observed beyond this threshold), are consistent with the guidelines, and offer a valuable supplement by addressing specific mortality outcomes. Moreover, although the WHO guidelines allow substitution between MPA and VPA for general health benefits, our findings emphasize that having VPA contribute 30–60% of total MVPA is associated with maximal reductions in all-cause and CVD mortality risks, underscoring the particular importance of VPA.

A potential issue of the findings is about the effects of VPA proportion (relative to MVPA) across different levels of MVPA volumes. Lee et al. (2022) analyzed the impacts of VPA within stratified amounts of total MVPA on all-cause and CVD mortality risks and found that increasing the VPA proportion was associated with lower mortality risk when MVPA was less than 300 min per week, but showed diminished benefits when MVPA exceeded 300 min. Supportively, a recent study (Strain et al., 2020) involving 96,476 participants demonstrated that a higher proportion of energy expenditure from MVPA in total physical activity was associated with lower all-cause mortality risk, particularly at total physical activity with lower levels of energy expenditure, while the benefits diminished at higher levels of energy expenditure. Together, these findings imply that the benefits of increasing VPA volume may be underrecognized at lower volumes of physical activity, but may be overestimated at higher volumes. Therefore, our findings regarding the 30–60% VPA proportion may be more reliable in the context of lower total volumes of physical activity. Based on our first conclusion that 180 min of VPA per week (for individuals engaging only in LMPA) yields substantial benefits, the potential gain from maintaining this proportion by increasing LMPA remains uncertain. Further research is needed to explore the association between VPA proportions and mortality risks across varying levels of physical activity volume.

The strength of this study lies in the quantification of the association of varying volumes and proportions of VPA (relative to MVPA) with the risks of CVD, cancer, and all-cause mortality through non-linear modeling. A common limitation in previous studies examining the effects of physical activity intensity on health outcomes is the failure to account for the volume of physical activity, which would diminish the credibility of the results. The present research addresses this limitation by considering the proportion of VPA (relative to MVPA) as a dose, allowing analysis of the effects of VPA versus MPA as well as the joint effects of VPA and MPA on mortality risks, thereby enabling more robust and reliable conclusions. Publication bias was evaluated by funnel plots and Egger’s regression test in each analysis to ensure the robustness of the findings.

There are several limitations in this study. Firstly, all included studies relied on self-reported physical activity data, which introduces potential measurement bias. A potential limitation of self-reported data is the tendency to overestimate the volume of VPA, and individual variability in reporting, which may result in an underestimation of the relationship between VPA and health outcomes (Prince et al., 2008; Ramakrishnan et al., 2021). However, the self-reported measurement allows for large-sample research, which is important for investigating the association with the mortalities. Moreover, a fixed-effects model was used to analyze the association between the volume of VPA and the outcomes. While this approach allowed for stable estimation, it may lead to underestimation of uncertainty in the pooled estimates, and this limitation should be considered when interpreting and applying the findings. Furthermore, since the studies included in our research predominantly used statistical models that adjusted for total MVPA volume, few have directly examined the relationship between VPA proportion and mortality risk across different levels of MVPA volume. In addition, considerable between-study heterogeneity was observed in certain analyses, while attempts to identify the sources of heterogeneity through subgroup analyses were limited by the small number of studies, which may reduce the statistical power.

Conclusion

This meta-analysis demonstrated a curvilinear inverse dose–response relationship between the volume of VPA and the risks of all-cause, CVD, and cancer mortality. Engaging in at least 180 min of VPA per week was associated with substantial reductions in these mortality risks. Additionally, a VPA proportion of 30–60% of total MVPA appears to be associated with maximal reductions in all-cause and CVD mortality risks, as indicated by a U-shaped inverse dose–response relationship between the proportion of VPA (relative to MVPA) and all-cause as well as CVD mortality risk. Overall, our findings address the importance of the VPA volume and its proportion (relative to MVPA) in health promotion, suggest incorporating a certain amount of VPA (>180 min per week) into daily physical activities, and specify an optimal combination of VPA and MPA (30–60% VPA to MVPA) in reducing all-cause, CVD, and cancer mortality risks. Future research is needed to further investigate the effects of VPA proportion on mortality risks by analyzing these effects across different levels of MVPA volume.

Supplemental Information