Understanding control group improvements in physical activity intervention trials

Lauren Waters (2011). Understanding control group improvements in physical activity intervention trials PhD Thesis, School of Population Health, The University of Queensland.

       
Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s4009527_phd_revised.pdf Thesis revised application/pdf 20.37MB 16
Author Lauren Waters
Thesis Title Understanding control group improvements in physical activity intervention trials
School, Centre or Institute School of Population Health
Institution The University of Queensland
Publication date 2011-12
Thesis type PhD Thesis
Supervisor Professor Elizabeth Eakin
DR Marina Reeves
DR Brianna Fjeldsoe
Professor Neville Owen
Total pages 203
Total colour pages 6
Total black and white pages 197
Language eng
Subjects 111716 Preventive Medicine
111799 Public Health and Health Services not elsewhere classified
Abstract/Summary Abstract Background Broad reaching interventions aimed at increasing population levels of physical activity are a population health priority. In a number of recent randomised controlled intervention trials, the control group has demonstrated meaningful and unexpected improvements in physical activity despite not receiving the intervention. A number of potential explanations for control group improvements have been proposed, but very little empirical research has been directed towards understanding this phenomenon. Aim The primary aim of this research is to gain an increased understanding of the factors associated with improved physical activity among control group participants in physical activity intervention trials. The set of studies included in this thesis examine specific factors that have been hypothesised to explain control group improvements in physical activity intervention trials (control group treatment, participant characteristics, pre-trial physical activity screening and measurement error). Methods The studies in this thesis comprise of a systematic review, and three empirical investigations. The studies in Chapters 3 and 4 use data from a completed cluster-randomised physical activity and diet intervention trial (the Logan Healthy Living Program) in which a control group improvement was observed. Data from an ongoing randomised controlled trial targeting weight loss (Living Well with Diabetes) was used for the study described in Chapter 5. The analyses in Chapters 3-5 are secondary in nature. Results The results of the studies included in this thesis provide evidence that control group improvements occur in approximately 28% of physical activity intervention trials. Hypotheses proposed to explain this phenomenon were broadly grouped into those related to the control group treatment, the characteristics of the enrolling participants and the way in which physical activity is measured. Control group treatment: No evidence was found to suggest that the treatment of the control group impacts on control group outcomes. Participant characteristics: The systematic review in Chapter 2 showed that trials that targeted a population who were at risk of a chronic disease; used pre-trial screening to select inactive participants; and where participants’ mean body mass index was in the overweight range reported control group improvements more frequently than trials without these characteristics. The study in Chapter 3 revealed that, for control group participants in the Logan Healthy Living Program, those who were unmarried, retired, had completed high school, and who had a higher body mass index were more likely than those without these characteristics to report an increase of 60 minutes or more at the end of the intervention. The study included in Chapter 4 found that the amount of improvement reported by the control group in the Logan Healthy Living Program would have been minimally impacted upon had pre-trial screening been used to purposively select inactive participants. Measurement: Due to the unanticipated absence of a control group improvement in the Living Well with Diabetes trial, the study included in Chapter 5 was not able to examine whether control group improvements represent real behaviour changes or merely reflect changes in the way behaviour is reported or recorded. The bias between self-reported and objectively measured physical activity increased as the amount of physical activity increased, therefore measurement error may contribute to an explanation for control group improvements in physical activity intervention trials where a control group improvement occurs. Conclusions Control group improvements in physical activity intervention trials are likely the result of a number of interconnected mechanisms. Control group treatment is often cited as an explanation for control group improvements, but there is currently insufficient evidence to support this hypothesis. The characteristics of the participants enrolled in the trial appear to be important predictors of whether or not a control group improvement is observed. However, further research is needed to determine whether the individual participant characteristics identified in this thesis as being important predictors of control group improvements, are generalisable to other populations. Measurement error cannot be excluded as a possible explanation for this phenomenon based on the results of this research. Further empirical studies aimed at understanding control group improvements in physical activity intervention trials are warranted. Understanding why control group improvements occur in physical activity intervention trials will aid in the interpretation of null or reduced intervention effects occurring as a result of concurrent control and intervention group improvements. Such research has the potential to provide an important contribution to the progression of physical activity intervention research.
Keyword Behaviour
Usual care
Measurement
Participant characteristics
Screening
Research methods
Diabetes
Additional Notes Colour pages (pdf numbers): 28, 37,83, 95, 123, 134 Landscape pages (pdf numbers): 11-16, 121, 182-196

 
Citation counts: Google Scholar Search Google Scholar
Access Statistics: 90 Abstract Views, 16 File Downloads  -  Detailed Statistics
Created: Wed, 16 May 2012, 14:01:11 EST by Lauren Waters on behalf of Library - Information Access Service