Gamification for health promotion: systematic review of behaviour change techniques in smartphone apps.

OBJECTIVE: Smartphone games that aim to alter health behaviours are common, but there is uncertainty about how to achieve this. We systematically reviewed health apps containing gaming elements analysing their embedded behaviour change techniques. METHODS: Two trained researchers independently coded apps for behaviour change techniques using a standard taxonomy. We explored associations with user ratings and price. DATA SOURCES: We screened the National Health Service (NHS) Health Apps Library and all top-rated medical, health and wellness and health and fitness apps (defined by Apple and Google Play stores based on revenue and downloads). We included free and paid English language apps using 'gamification' (rewards, prizes, avatars, badges, leaderboards, competitions, levelling-up or health-related challenges). We excluded apps targeting health professionals. RESULTS: 64 of 1680 (4%) health apps included gamification and met inclusion criteria; only 3 of these were in the NHS Library. Behaviour change categories used were: feedback and monitoring (n=60, 94% of apps), reward and threat (n=52, 81%), and goals and planning (n=52, 81%). Individual techniques were: self-monitoring of behaviour (n=55, 86%), non-specific reward (n=49, 82%), social support unspecified (n=48, 75%), non-specific incentive (n=49, 82%) and focus on past success (n=47, 73%). Median number of techniques per app was 14 (range: 5-22). Common combinations were: goal setting, self-monitoring, non-specific reward and non-specific incentive (n=35, 55%); goal setting, self-monitoring and focus on past success (n=33, 52%). There was no correlation between number of techniques and user ratings (p=0.07; rs=0.23) or price (p=0.45; rs=0.10). CONCLUSIONS: Few health apps currently employ gamification and there is a wide variation in the use of behaviour change techniques, which may limit potential to improve health outcomes. We found no correlation between user rating (a possible proxy for health benefits) and game content or price. Further research is required to evaluate effective behaviour change techniques and to assess clinical outcomes. TRIAL REGISTRATION NUMBER: CRD42015029841.

Cerebral function tests reveal differences in HIV-infected subjects with and without chronic HCV co-infection.

Neurocognitive impairment (NCI) remains prevalent in HIV-infected subjects despite effective combination antiretroviral therapy (CART). In subjects without evidence of hepatic decompensation, NCI is also a feature of chronic HCV infection. The present study aimed to examine cerebral function and establish differences between HIV-HCV co-infected (HCVco) and HIV mono-infected (HIVmo) individuals. Neurologically asymptomatic subjects with chronic HCVco were eligible and underwent computerized neurocognitive testing (CogState; CogState Ltd, Melbourne, Australia), a dementia assessment [International HIV Dementia Scale (IHDS)] and memory assessment [the Prospective and Retrospective Memory Questionnaire (PRMQ)]. Historic control data were available for 45 HIVmo individuals and differences between study groups were assessed. Twenty-seven HCVco subjects were recruited. Plasma HIV RNA was <50 copies/mL in 25/27 of HCVco subjects and all HIVmo subjects and nadir CD4+ cell count (mean ± SD) was 214 ± 166 cells/µL and 180 ± 130 cells/µL, in HCVco and HIVmo subjects, respectively. No statistically significant differences in neurocognitive parameters or PRMQ scores were observed between groups. However, a trend towards poorer executive function score was observed in HCVco subjects (p 0.106). IHDS score (mean ± SD) was poorer in HCVco subjects (10.48 ± 1.25) vs. HIVmo subjects (11.51 ± 0.76), (p <0.001). In a multivariate model, increasing age and HCVco were the only factors significantly associated with poorer IHDS scores (p 0.039 and <0.001, respectively). In HIV-infected subjects stable on CART, statistically significantly poorer performance in the IHDS score was observed in subjects with HCVco, although no differences were observed after neurocognitive testing or memory assessment.