Characterizing Differences in Mortality at the Low End of the Fitness Spectrum

Sandra Mandic; Jonathan N. Myers; Ricardo B. Oliveira; Joshua P. Abella; Victor F. Froelicher


Med Sci Sports Exerc. 2009;41(8):1573-1579. 

In This Article


Demographic Characteristics

During a mean ± SD follow-up period of 8.7 ± 5.3 yr, there were a total of 546 (13%) deaths from any cause and 109 (2.5%) cardiovascular deaths. Significant trends were observed for decreasing age and for both all-cause and cardiovascular mortalities from the least-fit (Q1) to the most-fit (Q5) quintiles (Table 1). Compared to the fittest subjects (fifth quintile), the least-fit subjects (Q1) had a fourfold increased risk of both all-cause and cardiovascular mortalities, and the risk was progressively reduced with increasing quintiles of fitness (Figure 1). These trends are evident from the Kaplan-Meier survival curves (Figure 2). However, the reduction in mortality risk between quintiles was not linear; a nearly twofold increase in age-adjusted relative risk of mortality occurred between the least-fit (Q1) and the next-least-fit (Q2) quintile of fitness, with smaller differences observed between the other quintiles (Figure 1). Mortality remained significantly higher in Q1 compared with that in Q2 when adjusted for differences in clinical characteristics or when patients with less than 3 yr of follow-up were excluded.

Figure 1.

Hazard ratios for all-cause and cardiovascular mortalities across quintiles of fitness in healthy individuals. CV = cardiovascular; Q = quintile; Q1 = least-fit; Q5 = most-fit.

Figure 2.

Kaplan-Meier curves for all-cause mortality across quintiles of fitness in healthy individuals. Quintile 1 = least-fit; quintile 5 = most-fit.

Trends Across Quintiles of Fitness

Use of cardiovascular medications and prevalence of risk factors were significantly lower among the lowest compared with the highest quintiles of fitness (Table 1). By ANOVA, significant main effects were observed for decreased resting HR and blood pressure and increased peak exercise HR (absolute and age-predicted), systolic blood pressure, and absolute and age-predicted exercise capacity from the lowest to the highest quintile of fitness (Table 2).

The observed main effects in recreational physical activity patterns across quintiles of fitness (Figure 3) were largely attributable to a decreased proportion of less active individuals (energy expenditure: < 1000 kcal·wk-1) and an increased proportion of active individuals (energy expenditure: >2000 kcal·wk-1) from the first to the fifth quintile, whereas the proportion of moderately active individuals (energy expenditure: 1000-2000 kcal·wk-1) was relatively constant across quintiles of fitness (Figure 4).

Figure 3.

Age-adjusted weekly energy expenditure in recent (upper panel) and lifetime (lower panel) recreational physical activity for quintiles of fitness in healthy individuals. Q1 = least-fit quintile; Q5 = most-fit quintile. *P < 0.05 versus Q5; P < 0.05 versus Q1.

Figure 4.

Interactions between fitness, physical activity patterns, and mortality risk in healthy individuals. All-cause = all cause mortality; CV = cardiovascular mortality; CR = cardiorespiratory; Mod. active = moderately active; Q1 = least-fit quintile; Q5 = most-fit quintile.

Comparison Between Least-fit and Next-least-fit Quintiles

Subjects in Q1 were older, had more extensive use of ACE inhibitors, calcium channel blockers, and other antihypertensive agents, and had a higher prevalence of diabetes compared with Q2 (Table 1). The prevalence of dyslipidemia was lower in Q1 versus Q2. No other differences in risk factors or medication use were found between the groups. Exercise test responses (maximum HR and systolic blood pressure, exercise capacity) were significantly lower in Q1 versus Q2 (Table 2).

Energy expenditure expressed as age-adjusted recent recreational physical activity (in the last year) was significantly lower in Q1 (n = 142) versus Q2 (n = 152) (1120 ± 1529 vs 1566 ± 1987 kcal·wk-1, respectively, P = 0.03). Lifetime adulthood recreational physical activity pattern was not different between the groups.

Fitness as a Multivariate Predictor of Mortality

After adjustment for age, risk factors, and cardiovascular medications, reduced peak exercise capacity was the strongest multivariate predictor of all-cause (hazard ratio: 0.89 (95% confidence interval (CI): 0.86-0.91)) and cardiovascular mortalities (hazard ratio: 0.87 (95% CI: 0.82-0.93)). Each 1-MET increase in exercise capacity conferred an 11% risk reduction for all-cause mortality and a 13% risk reduction for cardiovascular mortality in the total population. When only Q1 and Q2 were considered, the risk reductions per MET were 21% for all-cause mortality and 24% for cardiovascular mortality.

Comparison of demographic and clinical characteristics of the cohort with physical activity data versus the entire cohort is presented in Table 3. Because physical activity data were collected on patients in more recent years, this subgroup had a significantly shorter follow-up period and lower mortality rates compared with the entire cohort. In addition, higher usage of ACE inhibitors, anticoagulants and statins, and higher prevalences of dyslipidemia and smoking were observed in the physical activity subgroup versus the remainder of the cohort. Prevalence of other risk factors and exercise capacity were not different between the cohorts.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.