A randomised controlled trial and cost-effectiveness evaluation of 'booster' interventions to sustain increases in physical activity in middle-aged adults in deprived urban neighbourhoods

Background - More evidence is needed on the potential role of 'booster' interventions in the maintenance of increases in physical activity levels after a brief intervention in relatively sedentary populations.

Objectives - To determine whether objectively measured physical activity, 6 months after a brief intervention, is increased in those receiving physical activity 'booster' consultations delivered in a motivational interviewing (MI) style, either face to face or by telephone.

Design - Three-arm, parallel-group, pragmatic, superiority randomised controlled trial with nested qualitative research fidelity and geographical information systems and health economic substudies. Treatment allocation was carried out using a web-based simple randomisation procedure with equal allocation probabilities. Principal investigators and study statisticians were blinded to treatment allocation until after the final analysis only.

Setting - Deprived areas of Sheffield, UK.

Participants - Previously sedentary people, aged 40-64 years, living in deprived areas of Sheffield, UK, who had increased their physical activity levels after receiving a brief intervention.

Interventions - Participants were randomised to the control group (no further intervention) or to two sessions of MI, either face to face ('full booster') or by telephone ('mini booster'). Sessions were delivered 1 and 2 months post-randomisation.

Main outcome measures - The primary outcome was total energy expenditure (TEE) per day in kcal from 7-day accelerometry, measured using an Actiheart device (CamNtech Ltd, Cambridge, UK). Independent evaluation of practitioner competence was carried out using the Motivational Interviewing Treatment Integrity assessment. An estimate of the per-participant intervention costs, resource use data collected by questionnaire and health-related quality of life data were analysed to produce a range of economic models from a short-term NHS perspective. An additional series of models were developed that used TEE values to estimate the long-term cost-effectiveness.

Results - In total, 282 people were randomised (control = 96; mini booster = 92, full booster = 94) of whom 160 had a minimum of 4 out of 7 days' accelerometry data at 3 months (control = 61, mini booster = 47, full booster = 52). The mean difference in TEE per day between baseline and 3 months favoured the control arm over the combined booster arm but this was not statistically significant (-39 kcal, 95% confidence interval -173 to 95, p = 0.57). The autonomy-enabled MI communication style was generally acceptable, although some participants wanted a more paternalistic approach and most expressed enthusiasm for monitoring and feedback components of the intervention and research. Full boosters were more popular than mini boosters. Practitioners achieved and maintained a consistent level of MI competence. Walking distance to the nearest municipal green space or leisure facilities was not associated with physical activity levels. Two alternative modelling approaches both suggested that neither intervention was likely to be cost-effective.

Conclusions - Although some individuals do find a community-based, brief MI 'booster' intervention supportive, the low levels of recruitment and retention and the lack of impact on objectively measured physical activity levels in those with adequate outcome data suggest that it is unlikely to represent a clinically effective or cost-effective intervention for the maintenance of recently acquired physical activity increases in deprived middle-aged urban populations. Future research with middle-aged and relatively deprived populations should explore interventions to promote physical activity that require less proactive engagement from individuals, including environmental interventions.