This review assessed second-generation computer-tailored primary prevention interventions for dietary behaviour change and concluded that the evidence of effectiveness for these interventions was fairly strong. Given uncertainty over parts of the review process and the below-average quality of many included studies, the reliability of the authors' conclusions is unclear.

To evaluate the effectiveness of second-generation computer-tailored primary prevention interventions for dietary behaviour change and determine the key characteristics for success; this abstract is focused on the first objective.

MEDLINE, EMBASE, PsycINFO, CINAHL and EBM Reviews were searched for English-language studies from 1996 to January 2008; search terms were reported. Previous systematic reviews and reference lists of relevant articles were searched to identify additional studies.

Randomised controlled trials (RCTs) or quasi-experimental designs that assessed second- or third-generation computer-tailored dietary behaviour change interventions in adults with pre-test and post-test behavioural outcome data were eligible for inclusion. Conference abstracts were excluded, as were dissertations, commentaries, technology descriptions or information architecture and descriptions of the development of an intervention. Also excluded were studies in which the intervention was targeted at caregivers, health professionals and those either with a manifest chronic disease state or whose recruitment was based upon chronic disease registries. Studies of interventions that included individual therapist-generated feedback or that were described as maintenance strategy for a previous intervention were excluded. Interventions with limited interpersonal contact were included and these comprised: computer-tailored feedback through telephone or email; initial one-off face-to-face sessions for the purpose of instructing participants in the use of the technology or data collection; and interventions that had additional treatment arms such as face-to-face sessions. Interventions with significant face-to-face contact that involved counselling in one of the main treatment arms of the study were excluded. Dietary behaviours included reduced fat consumption, increased fruit and vegetable consumption or increased fibre intake. Delivery mechanisms for computer-tailored interventions included desktop computer programs, internet/intranet, telephone and multimedia. The primary outcome measure was dietary behaviour change; where multiple behaviours were assessed only dietary behaviour and change in body mass or weight outcomes were considered. The most frequently used outcome measure was the food frequency questionnaire. Most included studies involved self-selected healthy adults recruited through community, workplace and primary health care settings. Where stated, mean age ranged from 20.2 to 45.9 years and the proportion of females ranged from 43% to 100%.

The authors stated neither how the studies were selected for the review nor how many reviewers performed the selection.

Two reviewers independently assessed study quality using 17 criteria that assessed both internal validity (eight criteria on study design and methods, analysis and intervention design) and external validity (nine criteria on program reach and representativeness, outcomes for decision-making and maintenance of effects). Study quality was expressed as a percentage of maximum possible score. Disagreements were resolved by consensus.

Data were extracted as the direction of effect on the range of pre- and post-test behavioural outcomes with significant findings highlighted; numerical data were not provided to clarify statistical significance.

Two reviewers independently extracted data for the review; disagreements were resolved by consensus.

A narrative synthesis was provided and differences between studies were discussed in the text.

A total of 13 RCTs was included (n=at least 6,300, range 73 to 1,071); there was a discrepancy between the text and tables. Internal validity scores averaged 66% (range 50% to 88%). External validity scores averaged 52% (range 33% to 78%). Five studies had above average validity scores. Follow-up ranged from one to six months; one study continued to two years (without control group).

Significant short- to medium-term positive effects on dietary behaviour outcomes were reported for computer-tailored interventions (seven studies). Five studies reported a significant between-group effect on fat intake in favour of the computer-tailored intervention over a control group: three of these studies reported within-group positive effects on fat intake, but no significant between-group effects; and two found positive between-group effects for mediators such as knowledge, awareness, self efficacy, stage of change and intention to change. The computer-tailored intervention yielded positive effects over the control group for increasing fruit and vegetable intake (four studies) and for increasing fibre intake (three studies).

One study reported that the cost of setting up a website was at least $20,000 AUD.

The evidence of effectiveness for computer-tailored dietary behaviour change interventions was fairly strong. Such interventions had potential to reach large self-selected groups of people. There was uncertainty about whether the reported behaviour changes were generalisable or sustainable long term. The relative success of different components of efficacious interventions was unclear, as was the optimal intervention intensity.

The review question and inclusion criteria were clear. The search strategy was thorough, but restricted to publications in English and unpublished studies were not actively sought; publication and language biases may have been present and some studies may have been missed. It was unclear whether methods were used to reduce error and bias for study selection; data extraction and study quality were both assessed in duplicate. Study quality was assessed using appropriate criteria. A narrative synthesis was appropriate given variability across the studies. Given the uncertainty over the methods used for study selection, possibility of publication and language biases and below average quality of most included studies, the reliability of the authors' conclusions is unclear.

Practice: The authors stated that interventions should be based on theory and tailored to the individual.

Research: The authors stated that further research on computer-tailored dietary behaviour primary prevention interventions was required. Work was required to determine the optimal intensity for population-level interventions. Areas for investigation included: use of objective outcome measures; replication of successful efficacy trials in different settings and population groups; a review of the research on engagement and acceptability of such interventions; long-term post-intervention follow-up studies; and cost-effectiveness studies.

New South Wales Department of Health supplied funding for publication.

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.