"Consider this: Dr. Kenneth Cooper (author of Aerobics, The New Aerobics, Aerobics for Women), the U.S. Air Force Cardiologist who coined the term 'aerobics" (meaning a form of exercise) and has promoted their use for over 25 years, now admits that he was wrong! According to Dr. Cooper, further research has shown that there is no correlation between aerobic endurance performance and health, longevity, or protection against heart disease. He will admit, however, that such activities do carry with them a great risk of injury. Further, he admits that gross-overuse activities such as running are damaging to the body." – Ken Hutchins, SuperSlow Exercise Guild
The baseline characteristics of the study participants are presented in Table 1. No differences between the training groups existed at study entry. In both groups, men spent more hours in sedentary behavior and had significantly higher weight, height, and VO2peak compared to women. Contrary, more women than men performed at least 30 min of daily physical activity (Table 1). The included participants had higher VO2peak (11%) compared to those with no exercise logs. They also had higher VO2peak (17%) and height (1.7%) compared to dropouts, but a lower BMI (3.7%) (p < 0.05). A higher proportion of the included participants performed 30 min of daily physical activity compared to the dropouts (77.3% vs. 66.1%, p < 0.05).
One way repeated ANOVA was used to compare time to exhaustion between sessions (S1, S2 and S3). Relative reliability was calculated with the intraclass correlation (ICC) model (3, 1) . Absolute reliability was calculated with the typical error of measurement (the standard deviation of the change scores divided by [28, 29]). Bland and Altman’s 95% limits of agreement were also used (calculated for S1 vs S2, S1 vs S3 and S2 vs S3) as an additional representation of measurement error and to identify the presence of heteroscedasticity . As data were heteroscedastic, both raw data and log transformed Bland and Altman’s plots are presented. Limit of agreement ratio (LOA) was also calculated from the log transformed data as follow: LOA = (1.96 × SDdiff / grand mean) × 100; where “SDdiff” represents the SD of the differences between tests (S1 vs S2, S1 vs S3, S2 vs S3) and “grand mean” represents (mean S1 + mean S2 + mean S3)/3. As time to exhaustion data were heteroscedastic, we also calculated the coefficient of variation (CV) for each subject as follow: CV = 100×(SD of the three measurements)/(mean of the three measurements). Mean CV for all subjects were also calculated. We also calculated the smallest worthwhile change (0.2 × between subjects SD) .
It is well known that exercise in the older population may prevent several diseases [1–4]. Reduced physical activity impairs the quality of life in elderly people with Alzheimer's Disease , Parkinson's Disease , and Depressive Disorders . Moreover, musculoskeletal, cardiopulmonary, and cerebrovascular decline are associated with poor physical fitness because of the cumulative effects of illness, multiple drug intake, fatigue, and bed rest [7, 8]. The effects of physical activity and exercise programs on fitness and health-related quality of life (HRQOL) in elderly adults have been widely studied by several authors [9–11]. De Vries et al.  conducted a meta-analysis focusing on elderly patients with mobility problems and/or multimorbidity. Eighteen articles describing a wide variety of actions were analyzed. Most used a multicomponent training program focusing on the combination of strength, balance, and endurance training. In 9 of the 18 studies included, interventions were supervised by a physical therapist. Intensity of the intervention was not reported and the duration of the intervention varied from 5 weeks to 18 months. This meta-analysis concluded that, considering quality of life, the exercise versus no-exercise studies found no significant effects. High-intensity exercise appears to be somewhat more effective in improving physical functioning than low-intensity exercise. These positive effects are of great value in the patient population but the most effective type of intervention remains unclear. Brovold et al.  recently examined the effects of high-intensity training versus home-based exercise programs using the Norwegian Ullevaal Model  on a group of over-65-year-olds after discharge from hospital. These authors based their study on the Swedish Friskis-Svettis model  which was designed by Johan Holmsater for patients with coronaropathy to promote their return to work and everyday activities and improve their prognoses. This model includes three intervals of high intensity and two intervals of moderate intensity, each one lasting for 5 to 10 minutes. Included in each is coordination. Exercises consist of simple aerobic dance movements and involve the use of both upper and lower extremities to challenge postural control . Exercise intensity was adjusted using the Borg Rating of Perceived Exertion (RPE) Scale. Moderate intensity was set between 11 and 13, and high intensity was set between 15 and 17 on the Borg Scale.