This review suggested that physical activity and counselling improved daily step activity in (breast) cancer survivors and effectiveness was increased when a step goal was defined. The authors advised that their findings should be interpreted with caution due to clinical and statistical heterogeneity. This caution appeared justified, especially given the paucity of data and concerns regarding the review methodology.

To determine the effect of physical activity interventions on daily walking activity enhancement in cancer survivors.

MEDLINE, The Cochrane Library and PEDro were searched up to November 2009. Search terms were reported. Reference lists of retrieved articles were checked for further studies. Conference proceedings were screened. Only studies written in English were eligible for inclusion in the review.

Randomised controlled trials (RCTs) that assessed the effects of a physical activity intervention on daily walking activity in cancer patients were eligible for inclusion in the review. Physical activity interventions were defined as one or more of walking, physical exercise, counseling, yoga and relaxation. Studies were excluded if they evaluated the effectiveness of drugs, nutrition, transcutaneous electrical stimulation and mineral/vitamin supplements; studies of hospitalised patients were excluded. Eligible studies had to monitor walking activity using pedometers, step counters or accelerometers.

Included studies assessed supervised exercise programmes, behaviour change programmes and telephone advice regarding standard public health recommendations. Various physical activity modalities were used and included combined supervised endurance training (swimming, aerobics, other forms or a combination), individual exercises at a health club and home exercise. Three included studies compared the combined effect of physical activity, counselling (face to face/telephone) and goal setting against usual care. Four trials included physical activity counseling, with a median of four counseling sessions (range four to eight sessions). Intervention durations varied from six weeks to six months for supervised physical activity and from six weeks for supervised and 10 weeks for home-based activity. Frequency and duration of sessions varied between studies. Mean age of participants was 53.6 (standard deviation 4.2) years; all studies included participants older than 60 years. Most studies assessed participants with breast cancer and enrolled only female participants; only 11.7% of the included participants were male. One study assessed patients who had undergone a haematopoietic stem cell transplantation. On average, 71% of the included participants were white (standard deviation 32%). All of the included studies assessed participants who wore activity monitors for seven consecutive days. Physical activity was measured using an Actigraph MTI model 71256, CT1 M Actigraph accelerometer, Cymatech step activity accelerometer SAM3 and New Lifestyles Digiwalker SW-200. In two studies patient activity was recorded using a step diary. Three studies were performed in USA, one in Canada and one in Switzerland.

The authors did not state how papers were selected for the review.

Two reviewers independently assessed the methodological quality of the studies using a modified PEDro scale (nine out of 11 criteria used). Criteria included eligibility, use of randomisation, concealment of treatment allocation, baseline comparability, blinding of outcome assessors, use of intention-to-treat analysis, reporting of point estimates, measures of variability of the primary outcome, completeness of follow-up and use of between-group statistical comparisons. Discrepancies were resolved through consensus.

One reviewer extracted the change in the number of steps per day for the intervention and control groups. Data were used to calculate effect sizes (ES) with 95% confidence intervals (CIs).

A pooled effect size with 95% CI was calculated for three studies that assessed similar interventions and populations (random-effects model). Statistical heterogeneity was assessed using the I² statistic and Q-statistic (I²≥50%) or more was considered to indicate substantial heterogeneity. Publication bias was assessed using the fail-safe N.

Five RCTs (n=660) were included in the review; three (n=152) were included in the pooled analysis. The median score PEDro score was 8 out of 11 (range 7 to 9). Randomisation was adequate in all of the studies. All studies except one concealed the treatment allocation. Two studies reported blinding the outcome assessor. Dropout rates from baseline ranged from zero to 13%.

Mean change in daily step activity for patients in the intervention groups was 526 daily steps (standard deviation 537, range -92 to 1,299; five RCTs). The effect of combined physical activity and counselling on daily walking activity in breast cancer patients was superior to usual care (pooled ES 0.4, 95% CI 0.0 to 0.7); there was evidence of significant statistical heterogeneity (I² =79%).

Evidence suggested that physical activity and counselling improved daily step activity in (breast) cancer survivors and effectiveness is increased in studies that defined step goals; however, these findings should be interpreted with caution due to clinical and statistical heterogeneity.

This review answered a clearly defined research question. Literature searches were limited to articles written in English and so there was a risk of language bias. Unpublished studies were included in the review, but the authors acknowledged there was a risk of publication bias as specific attempts to locate unpublished data were limited. Attempts were made to reduce reviewer error and bias during quality assessment of the included studies, but not when extracting study data. It was unclear whether similar precautions were taken when selecting studies for inclusion. The methodological quality of the studies was assessed using relevant criteria. General study quality appeared to be relatively good. The authors acknowledged that there was evidence of both clinical and statistical heterogeneity. Although three of the included studies were judged to be sufficiently similar clinically, tests suggested evidence of statistical heterogeneity that may have effected the reliability of the pooled effect size. The authors acknowledged that a paucity of data limited their findings. Given these limitations, the authors cautious interpretation of their findings appeared justified, especially in light of the paucity of data and concerns regarding the reliability of the review methodology.

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

Research: The authors stated that further large randomised controlled trials were needed to evaluate the short- and long-term effects of goal-targeted physical activity, with or without counseling, on daily walking in various cancer populations. Trials that evaluated the effectiveness of social cognitive theory in cancer survivors were required.

None 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.