Background: Research suggests communication skills do not reliably improve with experience. Considerable effort is dedicated to courses improving communication skills for health professionals. Evaluation of such courses is important to enable evidence-based teaching and practice.
Objectives: To assess whether communication skills training is effective in changing health professionals' behaviour in cancer care with regard to communication/interaction with patients.
Search strategy: We searched CENTRAL (Cochrane Library Issue 3 2001), MEDLINE (1966 to November 2001), EMBASE (1980 to November 2001), PsycInfo (1887 to November 2001), CINAHL (1982 to November 2001), AMED (1985 - October 2001), Dissertation Abstracts International (1861 to March 2002) and EBM Reviews (1991 to March/April 2001). Reference lists of relevant articles were searched. Three further studies were detected in November 2003.
Selection criteria: Randomized controlled trials or controlled before and after studies of communication skills training in cancer health professionals, measuring changes in behaviour/skills using objective and validated scales.
Data collection and analysis: Two reviewers independently assessed trials and extracted data.
Main results: Of 2824 references, 3 trials involving 347 health professionals were included. One provided an intensive 3 day course then assessed oncology doctors interacting with 640 patients; a second provided a modular course then assessed role plays with oncology nurses; the third was modular and assessed outcomes with clinical and simulated interviews and patient questionnaires. In one trial, course attendees used more focused questions (probability < 0.005), focused and open questions (p = 0.005), expressions of empathy (p < 0.005) and appropriate cue responses (p < 0.05) at follow up than non-attendees. No significant differences were found between attendees and non-attendees for leading questions. From baseline to follow up, attendees had significantly different changes in rates of leading questions (p < 0.05), focused questions (p < 0.005), open questions (p < 0.05) and empathy (p = 0.005). The only observed significant difference in the second trial was that trained doctors controlled the follow-up interview more than untrained doctors (p < 0.05). Neither studies found differences in summarising, interrupting and checking. The third trial found trained nurses used more emotional speech than untrained counterparts, particularly regarding anxiety and distress. Patients interviewed by trained nurses used more emotional terms, but no differences emerged in questionnaires.
Reviewers' conclusions: Training programmes assessed by these trials appear to be effective in improving some areas of cancer care professionals communication skills. It is unknown whether this training would be effective if taught by others, nor the comparative efficacy of these programmes.