The authors concluded that predispositional genetic testing in adults has no significant impact on psychological outcomes, behaviour or perceived risk. The review process was limited by methodological flaws in the areas of study selection and quality assessment that make it difficult to judge the reliability of the authors' conclusions.

To evaluate the perceived risks and psychological and behavioural impacts of predispositional genetic testing.

MEDLINE, EMBASE, CINAHL, PsycINFO and the Cochrane CENTRAL Register were searched from 2000 to 2006. Search terms were reported. The reference lists of relevant articles, reviews and the Science Citation Index were searched to identify additional studies. English language studies published in a peer-reviewed journal were eligible for inclusion.

Randomised controlled trials (RCTs) and prospective studies evaluating perceived risk, psychological (affective) and/or behavioural outcomes resulting from the genetic testing of adults with a family history of any multifactorial adult-onset genetic disorder were eligible for inclusion in the review. Studies had to provide a separate analysis of carriers and/or non carriers. Pre- and post-test data had to be available for perceived risk and affective outcomes; post-test data alone were sufficient for behavioural outcomes. Studies where genetic testing had not taken place, where participants were already affected by the disorder and for single gene disorders were excluded. The included studies tested for hereditary breast and ovarian cancer (HBOC), hereditary nonpolyposis colorectal carcinoma (HNPCC) and Alzheimer's disease (AD). Follow up ranged from baseline to five years. Where reported, most participants were self-selected educated white adults. Participants tested for HBOC were primarily but not exclusively female. Studies of HNPCC and AD included males and females. Affective outcomes (measured by a variety of listed instruments) included general and disorder-specific distress, general and state anxiety, depression and worry. Behavioural outcomes (measured largely by self report) related to screening, prophylactic surgery, use of chemotherapeutics and other health-related behaviours. Perceived risk was measured (largely by unvalidated instruments) and included standard genetic counseling. Titles and abstracts were screened by one reviewer. Two independent reviewers assessed the eligibility of full text articles. Discrepancies were resolved by discussion or by involving a third reviewer.

The authors stated that they chose not to perform a formal validity assessment.

Two reviewers independently extracted data on differences (including percentage changes) between the outcomes of interest in carriers and non carriers and between groups. Discrepancies were resolved by discussion.

Studies were combined in a narrative synthesis grouped by the outcomes of interest.

Thirty studies were included in the review.

Affective outcomes

Studies tested for: HBOC, 11 prospective studies (n=1,212); HNPCC, six prospective studies (n=578); HBOC and HNPCC, one prospective study (n=87); and AD, one RCT (n=86) and one prospective study (n=76).

Irrespective of disease, the majority of studies found no long term general distress for carriers or non carriers following genetic testing or disclosure of results. There were no studies of AD measuring general distress. In testing for HBOC, there was a trend towards lower short-term and long-term cancer-specific distress regardless of carrier status. Studies assessing cancer-specific distress following HNPCC found either no effect or increased short-term distress that returned to baseline for carriers; non carriers showed either a decrease or no change. There was no effect for specific distress in AD carriers over time. The majority of studies reported no long-term differences in anxiety or depression levels, irrespective of condition or carrier status. The results were conflicting in three studies measuring levels of worry.

Behavioural outcomes

Studies tested for: HBOC, seven prospective studies (n=1,516) and two cross-sectional studies (n=140); HNPCC, six prospective studies (n=529); and AD, one RCT (n=111).

Rates of screening behaviour were higher in carriers than non carriers of HBOC and HNPCC in terms of mammography, breast (self and clinical) examination, transvaginal ultrasound, Ca-125 and colonoscopy. Rates of prophylactic surgery for carriers of HBOC ranged from 0 per cent to 51 per cent for mastectomies and 13 per cent to 65 per cent for oophorectomies. One study reported that 9.5 per cent of female HNPCC carriers had undergone a hysterectomy within one year of disclosure. Following genetic testing, 20 per cent to 22.5 per cent of HBOC carriers were using chemotherapeutics to prevent breast cancer. AD carriers were more likely to engage in disease-preventative behaviour. Lifestyle behaviour changes were similar between carriers and non carriers.

Perceived risk

Studies tested for: HBOC, three prospective studies (n=422); and HNPCC, four prospective studies (n=449).

There were no long term differences in risk perception in relation to HBOC and HNPCC. Both carriers and non carriers had lower levels than prior to genetic testing.

Predispositional genetic testing for HBOC, HNPCC and AD in adults has no significant impact on psychological outcomes, behaviour or perceived risk.

The review question was clear. Inclusion criteria was adequately provided for study design, participants, intervention and outcomes. The search strategy covered several relevant electronic databases. The restriction to published English-language articles meant that language and publication biases could not be ruled out. The review process was conducted with some efforts to minimise error and bias, but the absence of two reviewers at the initial study selection phase and a lack of study quality assessment represent substantial limitations for judging the reliability of the review. Due to variations between the studies the method of data synthesis was appropriate. The authors' conclusions reflected the evidence presented, but without further detail on study quality and given the other methodological concerns it is difficult to judge their reliability.

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

Research: The authors stated that further research is required in terms of how to present information to patients undergoing genetic counseling and to test the multiple impacts of genetic testing for multifactorial disorders. Specific studies testing for Alzheimer's disease are warranted.

Heart and Stroke Foundation of Ontario Program. Grant number PRG#5513.

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.