Prevalence and predictors of treatment dropout from physical activity interventions in schizophrenia: a meta-analysis

Abstract

Objective

Physical activity interventions have been shown to improve the health of people with schizophrenia, yet treatment dropout poses an important challenge in this population, and rates vary substantially across studies. We conducted a meta-analysis to investigate the prevalence and predictors of treatment dropout in physical activity interventions in people with schizophrenia.

Method

We systematically searched major electronic databases from inception until August 2015. Randomized controlled trials of physical activity interventions in people with schizophrenia reporting dropout rates were included. Two independent authors conducted searches and extracted data. Random-effects meta-analysis and meta-regression analyses were conducted.

Results

In 19 studies, 594 patients with schizophrenia assigned to exercise interventions were investigated (age=37.2 years, 67.5% male, range=37.5%–100%). Trim and fill adjusted treatment dropout rate was 26.7% [95% confidence interval (CI)=19.7%–35.0%], which is more than double than in nonactive control interventions (odds ratio=2.15, 95% CI=1.29–3.58, P= .003). In the multivariate regression, qualification of the professional delivering the intervention (β=−1.06, 95% CI=−1.77 to − 0.35, P= .003) moderated treatment dropout rates, while continuous supervision of physical activity approached statistical significance (P= .05).

Conclusions

Qualified professionals (e.g., physical therapists/exercise physiologists) should prescribe supervised physical activity for people with schizophrenia to enhance adherence, improve psychiatric symptoms and reduce the onset and burden of cardiovascular disease.

Introduction

Compared with the general population, life expectancy in people with schizophrenia is shortened by 10 to 20 years [1], [2], with evidence indicating that the mortality gap still is growing [3]. The underlying causes for the increased risk for premature mortality in this population are complex and multifactorial. It is well established that people with schizophrenia are at an increased risk for cardiovascular diseases [4], metabolic syndrome [5], diabetes [6] and respiratory diseases [7]. Although genetic factors [8], psychotropic medication use [9] and shared pathophysiological mechanisms [10] contribute significantly to this high-risk profile, unhealthy lifestyle habits such as smoking [11], [12], poor diet [13] and low levels of physical activity [14] play a prominent role.

Increasing premature mortality and increased prevalence of unhealthy lifestyle choices of people with schizophrenia suggest that this population has not yet fully benefited from the health care improvements implemented in the general population. Although in previous years there has been a greater focus on improving the health status of people with schizophrenia [15], [16], there are still important health inequalities present even at the very early stages of the disease [17]. A primary reason for this is that, despite empirical evidence of moderate efficacy of pharmacotherapy [18], [19], [20], psychotherapeutic treatments [21], [22] and exercise [23], a large proportion of people with schizophrenia fail to respond to these treatment modalities. In pharmacological treatments for schizophrenia, it is not uncommon that the treatment dropout rate exceeds 50% [24]. Dropout represents a major barrier to the achievement of a successful treatment outcome. Patients who drop out generally experience worse clinical outcomes [25]. At societal level, dropout and lack of adherence to treatment are of major concern for budget holders and policy makers as they are associated with greater risk of rehospitalization and thus greater resource utilization [26]. Patients who fail to complete study protocols can affect statistical analyses, research outcomes and interpretation of results [27]. For example, the current research evidence for physical activity interventions in patients with schizophrenia is mainly based on data from participants completing the interventions [28]. This may skew results, favoring individuals who fully engage with physical activity [28]. Only a single trial [29] to date compared per-protocol and intention-to-treat analyses and found that significant improvements in physical fitness, psychiatric symptoms and overall functioning only occurred in participants who attended greater than 50% of the exercise sessions.

Some features common to randomized controlled trials (RCTs) investigating physical activity interventions in people with schizophrenia, such as the use of predetermined manualized protocols e.g., in terms of frequency, intensity, time, type) and specific eligibility criteria [e.g., inclusion of only patients with a higher body mass index (BMI)] have the potential to affect dropout rates. This suggests the need to consider dropout from RCTs separately from pragmatic or real-world interventions [30]. To date, no meta-analysis has examined the prevalence and predictors of dropout data from physical activity RCTs in people with schizophrenia. The current evidence on adherence to physical activity is mainly based on cross-sectional, prospective and qualitative research. Correlates that have been consistently associated with lower physical activity participation are the presence of negative symptoms and cardiometabolic comorbidity. Side effects of antipsychotic medication, lack of knowledge regarding cardiovascular disease risk factors, no belief in health benefits, lower self-efficacy, other unhealthy lifestyle habits and social isolation have also been associated with lower physical activity participation [31]. Qualitative research has identified that the most frequently cited barriers towards participation in physical activity for people with schizophrenia are patients' lack of motivation and a lack of priority given to physical activity by health care professionals. The most frequently cited facilitators include the provision of esteem support by health care professionals and the promotion of enjoyment and autonomy for the patient [32]. Given the increasing burden of cardiovascular diseases among people with schizophrenia and the potential for exercise to ameliorate this leading cause of mortality, understanding the prevalence and predictors of dropouts in physical activity RCTs is an important research question. Moreover, an empirically derived estimate of typical frequency of dropout from RCTs and an exploration of its moderators would inform the design of new RCTs, in addition to informing clinical practice and policy.

Recognizing that dropout from treatment protocols is high among people with schizophrenia, the current meta-analysis had the following aims: (a) to establish the prevalence of dropout in physical activity RCTs among people with schizophrenia; (b) to compare the prevalence of dropout from physical activity with the dropout in nonactive control conditions; (c) to identify predictors that may influence dropout such as demographics (e.g., mean age, % male), illness-related factors (psychiatric symptoms, illness duration, psychotropic medication use) and physical activity intervention parameters (e.g., frequency, intensity, time and type of physical activity) and the professional qualifications of the person delivering the intervention.