Date Presented 04/05/19
This research considered attendance in the Well Elderly II intervention to determine the dosage effect on depression scores. In particular, how many hours of OT sessions are necessary to produce meaningful reductions in depressive symptoms? Our analysis suggests that the small improvements in depression scores do not appear to be associated with the dosage of the intervention received; therefore, any preventative effects of the intervention need to be reconsidered.
Primary Author and Speaker: David Schelly
Additional Authors and Speakers: Ramona Nadres, Alisha Ohl
PURPOSE: Lifestyle Redesign has influenced a number of preventative occupational therapy (OT) interventions in community dwelling populations, where the aim is to modify participants’ activities to prevent age-related decline. Juang et al. (2018) explored the mechanisms of change for Lifestyle Redesign and suggested that the Well Elderly II intervention (Clark et al., 2012), which included 6 months of weekly small group and individual OT sessions, led to increased activity frequency and a more positive perception of activity significance, which ultimately reduced depressive symptoms. However, like other Lifestyle Redesign inspired interventions (e.g., Mountain et al., 2017), attendance in the OT sessions was variable, yet no authors have considered attendance in their analyses. The dosage of OT—in this case attendance hours—is not only relevant as a covariate but is a necessary consideration when trying to determine the effectiveness of an intervention. The purpose of this research was to consider attendance in the Well Elderly II intervention to determine the dosage effect on depression scores. In particular, how many hours of occupational therapy sessions are necessary to produce meaningful reductions in depressive symptoms?
DESIGN: The Well Elderly II trial randomly assigned 460 community dwelling older adults (60+ years old) to a treatment and control group. After attrition, the treatment group included 187 participants – the majority female (71.7%), nonwhite (59.4%), and with at least a high school degree (also 71.7%). In the original study, statistically significant improvements were reported in several domains, but the largest effect size was on the Center for Epidemiologic Studies Depression Scale (CES-D) (Schelly & Ohl, in press).
METHOD: This study used multiple linear regression to predict post-test CES-D scores in the 187 treatment group participants. Covariates included individual and group participation hours, baseline CES-D, income, education, and several activity and social network variables. We also graphed mean participation hours for individuals with low (quartile 1) and high (quartile 4) baseline scores for the same variables.
RESULTS: In multiple regression, the coefficient for individual hours indicates that receiving additional hours of OT intervention predicts worsening depression. In the simplest model, 6 hours of individual intervention predicts worsening of depression by 1 standard error of measurement (SEM) on the CES-D. The coefficient for group hours is in the expected direction, but the effect is small: between 275 and 395 hours of group intervention are needed to improve depression by 1 SEM. Importantly, individuals in the first quartile of depression scores at baseline (i.e., more depressive symptoms) received more than double the hours of individual OT intervention compared to those in the fourth quartile.
CONCLUSION: The undesired direction of the effect of individual OT hours, where additional treatment is associated with depressive symptoms at post-test, may be due to a selection effect: individuals are more likely to receive OT treatment if they have depressive symptoms at baseline, and the most likely individuals to have depressive symptoms at post-test are those who had them at baseline. This analysis suggests that the small improvements in the CES-D do not appear to be associated with the dosage of intervention received; any preventative effect needs to be reconsidered.
IMPACT STATEMENT: This proposal is extremely important to scientific practice because Well Elderly inspired randomized controlled trials (RCTs) are not considering dosage in their analyses, suggesting the need to reinterpret previous results. Dosage should be better controlled in community-based OT trials.
References
Clark, F., Jackson, J., Carlson, M., Chou, C.-P., Cherry, B. J., Jordan-Marsh, M., et al. (2012). Effectiveness of a lifestyle intervention in promoting the well-being of independently living older people: Results of the Well Elderly 2 Randomised Controlled Trial. Journal of Epidemiology & Community Health, 66(9), 782-790. http://dx.doi.org/10.1136/jech.2009.099754
Juang, C., Knight, B. G., Carlson, M., Schepens Niemiec, S. L., Vigen, C., & Clark, F. (2018). Understanding the mechanisms of change in a lifestyle intervention for older adults. The Gerontologist, 58(2), 353-361. https://doi.org/10.1093/geront/gnw152
Mountain, G., Windle, G., Hind, D., Walters, S., Keertharuth, A., Chatters, R., et al. (2017). A preventative lifestyle intervention for older adults (Lifestyle Matters): A randomised controlled trial. Age and Ageing, 46(4), 627-634. https://doi.org/10.1093/ageing/afx021
Schelly, D., & Ohl, A. (in press). Examining clinical meaningfulness in randomized controlled trials: Revisiting the Well Elderly II. American Journal of Occupational Therapy, 71(1).
