Methodological Issues in Exercise Intervention Studies: Attrition and Adherence

Purpose: While the benefits of exercise interventions in people undergoing HCT have been systematically reviewed elsewhere, subject attrition rates and adherence to exercise interventions following HCT have not. This systematic review of the literature examined subject attrition rates and reasons for attrition as well as adherence to exercise interventions following BMT/HCT. 

Methods:  Publications were identified through literature searches of PubMed, CINAHL, and Embase as well as hand searching the references of retrieved studies.  Studies published between January 1985 and July 2013 that prospectively tested an exercise intervention in BMT/HCT patients were included.  All retrieved abstracts were initially classified as (1) exercise intervention studies; (2) studies related to exercise and/or BMT or HCT that did not prospectively test an exercise intervention; (3) integrative or systematic reviews of exercise intervention studies in BMT or HCT patients; or, (4) other types of reviews articles (i.e., clinical reviews, case reports, etc.).  Studies that prospectively tested an exercise intervention were further evaluated to determine (1) the type of exercise modality employed; (2) subject attrition rates and reasons for attrition; (3) the amount of supervision required to implement the intervention; (4) timing of the intervention; and, (5) exercise adherence rates.

Results:  Twenty studies met the inclusion criteria.  The majority of studies tested an aerobic exercise intervention (n=7; 35%) or a combination of aerobic and strength training (n=7; 35%). The aerobic exercise interventions varied and included activities such as stationary bicycling, walking, or treadmill walking. Other exercise modalities tested included: (1) strength training (n=4; 20%) or (2) combination of stretching, aerobic and strength training (n=2; 10%).  Supervised exercise sessions (55%; n = 11) were more commonly used than unsupervised sessions (20%; n= 4). Five studies (25%) used a combination approach by supervising some sessions with study participants completing other sessions unsupervised. The overall attrition rate was 18% (180/998 subjects). Major reasons for attrition included death, change in health status, protocol issues, personal issues, and lost to follow-up/no reason provided.  Supervised exercise programs rarely published exercise adherence information.  Unsupervised exercise program relied mainly on self-report to document adherence. Adherence rates in these studies ranged from 50-100%.

Conclusion:  Questions regarding subject attrition and adherence to exercise interventions must be addressed to identify those interventions that are likely to be successful when translated into clinical practice.  Subject attrition from exercise studies following BMT/HCT is relatively low. Adherence information for exercise interventions needs to be regularly addressed.