This review assessed the impact of computer- and videotape-based patient education on knowledge, satisfaction and decision-making in patients with cancer. The authors concluded that education technologies increase knowledge and satisfaction. Anxiety does not increase with these interventions. Given the presence of clinical and statistical variation between the studies, it is unclear how reliable these conclusions are.

To evaluate the impact of videotape- and computer-based patient education on knowledge, satisfaction and decision-making in patients with cancer.

MEDLINE (1966 to week 3, April 2006), EMBASE (1980 to week 18, 2006), CINAHL (1982 to week 3, April 2006), the Cochrane Database of Systematic Reviews (week 3, April 2006) and the Cochrane CENTRAL Register (week 3, April 2006) were searched; the search terms were reported. Reference lists of identified studies were screened and an expert panel was consulted regarding ongoing research and recently published articles. It was not clear whether there were any language restrictions.

Study designs of evaluations included in the review

Studies of randomised controlled trials were eligible for inclusion.

Specific interventions included in the review

Studies of videotape- or computer-based patient education focusing on diagnostic screening were eligible for inclusion. Studies involving hypothetical choices, informed consent for clinical trials or public health screening were excluded. Studies of interventions in areas other than treatment decision-making or consent were excluded. Interventions using the internet or written information were excluded. The interventions included were videotape and computer education packages focusing on choosing treatment options, empowering and managing one's care, and facilitating consent. Control groups received written information and/or verbal consultation.

Participants included in the review

Studies of adults with a diagnosis of cancer were eligible for inclusion. Inclusion criteria for the type of cancer and stage of treatment were not stated. The studies included patients with breast cancer, patients awaiting colonoscopy and patients with different types and diagnoses of cancer.

Outcomes assessed in the review

Studies of knowledge, satisfaction, decision-making, treatment choice and care management were eligible for inclusion. The outcomes included varied between studies and covered patient knowledge, information-seeking, involvement in decision-making, interaction with doctors, satisfaction, treatment preference, anxiety and depression.

How were decisions on the relevance of primary studies made?

The authors did not state how the studies were selected for the review, or how many reviewers performed the selection. However, reasons for exclusion were given.

Two authors independently carried out the validity assessment using Rinck's system for assessing the validity of studies in palliative cancer care. This index assesses the validity of studies according to accrual, homogeneity, randomisation, attrition, intervention, outcome assessment and results. Each item was assigned a score of 1, 1/2 or 0, with a maximum total score of 7.

The authors did not state how many reviewers performed the data extraction.

The estimated mean difference or difference in proportion and sample standard deviation (SD) were extracted to calculate effect sizes. Changes in outcomes were extracted to calculate mean differences or odds ratios (ORs) with 95% confidence intervals (CIs). Effect sizes were calculated for all outcomes by dividing the estimated mean difference or the difference in proportions by the sample SD.

How were the studies combined?

A random-effects meta-analysis was conducted. A weighted mean difference (WMD) with 95% CI was calculated for continuous variables and an OR with 95% CI for dichotomous variables.

How were differences between studies investigated?

A chi-squared test and I-squared test for heterogeneity were carried out. Videotape and computer interventions were reported separately in the text and tables.

Nine studies (n=1,678) were included. Three studies (n=486) were of videotape interventions and 6 studies (n=1,192) were of computer interventions.

The majority of the studies were rated as high quality. Two scored the maximum of 7, six scored from 5 to 6 and one scored 4. Results for knowledge were equivocal. There was a significant improvement in knowledge in the intervention group when measured using continuous outcomes (WMD 5.95, 95% CI: 5.34, 6.56, p<0.00001) but not when dichotomous outcomes were used (OR 1.36, 95% CI: 0.83, 2.24, p=0.23). There was evidence of a significant level of heterogeneity for the analysis using continuous outcomes (chi-squared 2,379.35, d.f.=3, p<0.00001) but not for the analysis using dichotomous outcomes. Effect sizes for individual studies on knowledge ranged from 0.12 to 1.03. However, only 2 studies had an effect size of more than 0.5, indicating that the effect was generally moderate. Results for satisfaction were equivocal. Patient education interventions showed a significant increase in satisfaction when dichotomous outcomes were used (OR 1.67, 95% CI: 1.02, 2.74, p=0.04) but not when continuous outcomes were used (WMD 0.17, 95% CI: -0.02, 0.37, p=0.08). There was no evidence of heterogeneity in either analysis. Effect sizes for individual studies on knowledge were small to moderate, ranging from -0.05 to 0.4. In 1 study satisfaction was correlated with lower levels of psychological morbidity (no p-value reported). Other outcomes were not measured using a meta-analysis but the results were discussed individually in the text. In studies where psychological morbidity was measured, there was no difference between groups in the majority of studies. Two studies demonstrated a reduction in anxiety in the intervention group (p=0.001) and 1 study demonstrated a significant reduction in depression in the intervention group (p=0.001). A shift in doctor-patient interaction to a shared decision-making model was reported by 1 study (no statistical information provided). One study reported increased levels of social support in the intervention group and greater benefits of the intervention for disadvantaged groups (no statistical information provided).

No formal cost-benefit analysis was conducted. However, 1 study reported that, in the long-term, a personalised information system was no more costly than general computer information and was half the cost of full patient access to written information.

Patient education technology was as effective as traditional patient education and in many outcomes superior. Video and computer technology improved knowledge, were well-accepted by patients, and increased satisfaction with information and the decision-making process. Education technology did not increase anxiety. The impact of these technologies on patient decision-making was variable. The method of delivering interventions and degree of contact with a health professional affected the impact of the interventions.

The inclusion criteria for interventions, outcomes and design were clearly stated; those for the participants were less clear. A comprehensive search for published trials was carried out. However, it is unclear whether language restrictions were applied, which could result in relevant studies being omitted, and there does not appear to have been a search for unpublished studies, which may result in publication bias and an overestimation of intervention effects. A relevant quality assessment was carried out and the studies included were generally of good quality. However, the included studies only covered North America, the UK, and Australia, which may limit the applicability of the results. There was insufficient information on the study selection and data extraction processes to rule out the possibility of error or bias. Given the significant differences in intervention between videotape and computer technology, restricting the analysis to a narrative synthesis would have been more appropriate. The presence of both statistical and clinical heterogeneity limits the ability to draw general conclusions. Given these limitations, the authors' conclusions should be treated with caution.

Practice: The authors stated that technological devices should be used as an aid to and not as a substitute for interaction with a health professional.

Research: The authors did not state any implications for further research.

National Institute of Clinical Excellence; Cicely Saunders Foundation.

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.