This review of second- and third-generation computer-tailored interventions for physical activity behaviour change concluded that evidence for effectiveness was inconclusive. Potential bias in the search and selection of studies may limit the reliability of the authors’ conclusions.

To conduct a narrative systematic review of the range, quality, effectiveness and key success factors of second- and third-generation computer-tailored primary prevention interventions for physical activity behaviour change.

MEDLINE, EMBASE, PsycINFO, CINAHL and EBM Reviews were searched from January 1996 to January 2008 for English-language studies that were published in a scientific journal. Search terms were reported. Reference lists of reviews and included studies were searched.

Randomised controlled trials and pre-post studies that evaluated second- or third-generation computerised interventions that provided tailored physical activity advice were included. Second-generation interventions were those delivered by interactive technology or desktop applications. Third generation interventions were delivered by relatively new mobile and remote devices. Interventions were considered tailored if they provided advice or feedback specific for the characteristics and assessment of the individual. Physical activity or fitness levels had to be described. Studies were excluded if the intervention involved significant face-to-face-counselling or if the target group were caregivers, health professionals or people with a chronic disease.

All included studies described second-generation interventions, which included internet, email, computer program and telephone technology; one study also included a third-generation intervention of a mobile phone. The reported outcomes were physical activity, dietary behaviour and weight reduction. The authors stated that most participants were Caucasian, female and well educated. The method used to select studies was not reported.

Two authors assessed the validity of each study with quality criteria based on the Australian National Public Health Partnership and outlined by Glasgow and Emmons. Criteria included randomisation methods, inclusion criteria, baseline comparability, validity of outcomes, details of follow-up, reporting and analytical methods. Disagreements were resolved by consensus.

Each article was reviewed by two authors. Any disagreements were resolved by consensus.

Data was synthesised narratively and in table format. Studies were categorised as physical activity interventions only or a combined intervention that consisted of a physical activity and a dietary behaviour or weight reduction intervention or both. Whether or not there was a significant difference between the intervention and comparator group was listed in the table for each outcome of dietary behaviour, physical activity and weight reduction. Percentage scores for internal and external validity were listed. In the narrative synthesis of the data, the authors reported the number of studies that reported short-, medium- and long-term benefits for each of the outcomes; it was unclear how these follow-up periods were defined. Differences between studies were discussed with respect to study quality, intervention characteristics and use of theory.

Seventeen publications that reported 16 studies were included. It was not clear how many studies were randomised controlled trials and how many were pre-post studies. The number of participants ranged from 31 to 1,071. Internal validity scores ranged from 44% to 89% and external validity scored ranged from 33% to 78%.

Results for physical activity were reported for all studies, dietary behaviour in six studies and weight reduction in six studies. Significant differences for any of these outcomes were reported in nine studies in the table (reported as 10 in the abstract).

For seven of 16 studies that reported physical activity, there was a significant difference between the intervention and comparator group; this included four of eight studies that used a physical activity intervention only.

Three of the six studies (all combination interventions) that reported dietary behaviour were significantly different and two of the six studies (all combination interventions) that reported weight reduction were significantly different.

The evidence for effectiveness of computer-tailored interventions for physical activity behaviour change was inconclusive.

The inclusion criteria were clearly stated (although second- and third-generation interventions were not defined). The search strategy appeared appropriate, but the limitation to English-language articles published in journals may have caused studies to be missed. There may have been a bias in study selection as the methods used were not reported. Validity was assessed with appropriate criteria and methods designed to reduce reviewer error and bias. Details of included studies were limited. More information on study participants, interventions and reporting of other outcome data in addition to p-values could have strengthened this review. The narrative synthesis of data appeared appropriate given the apparent heterogeneity of interventions. Potential bias in the search and selection of studies may limit the reliability of the authors’ conclusions.

Research: The authors stated that to better determine generalisability, future research should replicate efficacy trials in different populations and settings. Studies with follow-up periods of at least a year were needed, as were studies that reported cost-effectiveness data, optimal intensity of interventions, effectiveness and timing of tailoring and the best way of delivering interventions that target more than one behaviour change.

Practice: The authors did not state any implications for practice.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.