Objective To test the effect of training residents in interpersonal and communication skills on women's satisfaction with patient–doctor relationship in labour and delivery rooms.

Design A stepped-wedge cluster randomised trial.

Setting Maternity wards in four teaching hospitals in Damascus and homes of participating women.

Participants Women delivering in the hospitals under study, via vaginal delivery with living baby, who consent to participate in the study. Difficult labour and high-risk pregnancies are excluded. Residents working in the study hospitals during the implementation phase who agree to take part in the study.

Main Outcome Measure Women's satisfaction with interpersonal relationships in labour and delivery rooms measured via a series of questions on a Likert scale and based on the Medical Interview Satisfaction Scale.

Results Women were on average 25 years old, 95% were homemakers, 76% had primary education, 48% lived in shared accommodation and 26% were nulliparous. At the individual level, the mean score of overall satisfaction of women was 68.66 (SD=14.24) out of a possible score of a 100 in the control group and 70.79 (SD=13) in the treatment group. At the hospital level, the mean scores of overall satisfaction of women were 70 (SD=4.70) and 70.99 (SD=4.85) for the control and treatment group, respectively. Using (generalised) linear mixed models approach to account for the study design, we were not able to detect a difference between the treatment and control group on the overall satisfaction of women.

Conclusion The training package does not seem to be associated with higher overall satisfaction scores.