The difficulty of recruiting older people to clinical trials is well known. However, there is limited information about the potential barriers to implementation of physical post-hospitalization interventions (10) and objectively measured clinical characteristics. The present study reveals that older adults have low interest (10.8%) in participating in a post-hospitalization physical exercise program. Non-participation was associated with physical, nutritional, and social parameters. Likewise, higher age, poor nutritional status, and home accessibility problems were strong independent predictors of non-participation.
Barriers to participation in the exercise program
Most post-hospitalization physical exercise programs have been carried out in patients undergoing cardiac or lung rehabilitation or after hip or knee repair. Participation rates reported in these programs (30–70%) are higher than in our program (11, 25). These differences may be due to the fact that physical exercise is considered an essential part of treatment after the above-mentioned diseases or conditions. Moreover, physical exercise after hospitalization is strongly recommended by the European Society of Cardiology (26), American Thoracic Society, and European Respiratory Society (27), and it is usually well-accepted by patients. However, geriatric patients with broader reasons for hospitalization similar to those in our study (mainly infections) often do not consider exercise as a useful intervention after discharge. Higher participation rates were reported in previous post-hospitalization physical exercise programs for older adults. However, the participants in those interventions were slightly younger than in the present study (10, 28). Nevertheless, the participation rate in our study was similar to other physical exercise programs for community-dwelling older adults, with rates of 7.3–13% in studies with similar participant characteristics (9).
It is remarkable that the main reason for non-participation for almost half of the patients was a lack of interest in physical exercise. This lack of interest has also been described in other older adult populations who do not recognize positive health-related effects of physical activity and do not believe that being physically active is helpful (29, 30). In this sense, health professionals may play a fundamental role in informing patients about physical and cognitive deterioration associated with hospitalization (1, 2) and the benefits of physical exercise after discharge to revert these deleterious effects (7). For these reasons, it is pertinent to actively encourage patients to get involved in physical exercise programs as part of a structured post-hospitalization treatment.
We also found that poorer accessibility at home was an independent factor for non-participation in the physical exercise program. This result agrees with other qualitative studies in which accessibility or environmental barriers hinder access to exercise programs (31). Usually, poorer accessibility is associated with lower socioeconomic status, which is one of the most important predictors of negative health outcomes (32). Therefore, special efforts are needed to provide accessible transport and improve incorporation of people with lower economic status in physical exercise programs and to encourage other healthy habits.
Participant characteristics
People who refused to participate were older than those who accepted the exercise program. Logistic regression models showed that the probability of non-participation increased by 13% with each additional year of age. This finding agrees with post-hospitalization cardiac rehabilitation programs, where participation declines significantly after 70 years of age and non-participation rates are even lower at 80 years of age (33). Similar results have been found in lung rehabilitation post-hospitalization programs (34).
We did not find gender-related differences in participation. However, a previous study found significantly lower participation of women after a cardiac event (11). These differences should be assessed with caution because there may be age, social, or disease-related differences among participants. For instance, older women in the Basque population are as physically active as older men (35), which could account for the lack of gender-related differences in our study.
Additionally, patients who participated in the program had better nutritional status. Interestingly, nutritional status was included in the last equation of the logistic regression model as an independent predictor of participation, which indicates strength of the relationship between nutrition and participation in physical exercise. Malnutrition and poor functional performance are closely interrelated—malnutrition is associated with a higher risk of sarcopenia (36) and, consequently, poorer functional status. Malnutrition in older people is due to multiple factors (37), led by medical illnesses, mental health conditions, psychological causes, or social isolation (38). Additionally, malnutrition or risk of malnutrition in older people is directly associated with multiple negative outcomes, including increased mortality (39) longer hospital stays, and worse quality of life (40, 41).
Interventions to improve nutritional and physical status regularly include exercise. Our findings suggest that it is necessary to implement strategies to reduce barriers to developing exercise interventions and to improve older patients’ participation in post-hospitalization exercise programs, especially for patients who generally refuse to participate. Further, lower participation of patients with worse nutritional status is worrying because these individuals are more prone to functional deterioration.
The strength of this study is its objective clinical measurement of reasons for non-participation in a post-hospitalization physical exercise program and its evaluation of nutritional and functional variables. To our knowledge, few studies have analyzed participation in post-hospitalization physical exercise programs in internal medicine and neurology departments. Further, those studies have usually analyzed qualitative and socioeconomic parameters for older adults’ non-participation (12, 13). Nonetheless, our study also has some limitations. First, it is limited by its cross-sectional nature, which excludes any ability to determine temporality and causality. Second, because this was a secondary analysis of a trial, some variables that could be relevant have not been assessed, such as social support, transport availability, or socio-economic status, which could have conditioned non-participation. Finally, the fact that there were only 55 participants could reduce the statistical power of the results and can also be considered a limitation of the study.