Advancing understanding of effective exercise on referral: a mixed methods evaluation of the Northumberland scheme

Abstract

Exercise on Referral Schemes (ERS) are a widespread community intervention in which health professionals refer patients to a programme of supervised exercise at leisure facilities. National guidance states routine data should be collected and made available for analyses, and that there is a need to better understand what elements of ERS work, for which subgroups of the population. This thesis examines what elements of behaviour change provision within ERS work, for whom and in what circumstances, in order to gain a better understanding of what influences referral to, engagement with, and adherence to such schemes.

First the thesis presents a quantitative evaluation study of 2233 referrals to the 24-week Northumberland ERS in nine leisure facilities between July 2009 and September 2010. Main outcome measures were uptake, 12-week adherence, 24-week completion, and changes in self-reported physical activity, blood pressure, body mass index (BMI) and waist circumference. Two qualitative studies follow, one examining pre-scheme perceptions of 15 referrals and the second following them through the scheme. Data from semi-structured interviews conducted in both studies are presented as three narrative typologies of the referral journey.

This research demonstrated that demographics and other factors related to referral minimally increased ability to predict engagement. Completion resulted in significant increases in self-reported physical activity and significant, but small, reductions in BMI and waist circumference. Participants had complex social circumstances, multiple personal reasons for referral and high expectations of positive health changes. Staff and peer support were influential to success, especially if expectations were not met. The narrative typologies help to identify those for whom ERS currently works well, those for whom ERS works but who may struggle with sustained behaviour change, and those for whom it does not work. This novel approach to classifying likelihood of success is used to discuss potential improvements to ERS.