Abstract Objective Anal incontinence occurs as a result of damage to pelvic floor and the anal sphincter. In women, vaginal delivery has been recognized as the primary cause. To date, figures quoted for overt third degree anal sphincter tear vary between 0% and 26.9% of all vaginal deliveries and the prevalence of anal incontinence following primary repair vary between 15% and 61%. Our aim was to analyse the long-term (minimum 10 years post primary repair) anorectal function and quality of life in a cohort of women who suffered a third degree tear (Group 1) and compare the results with a cohort of women who underwent an uncomplicated vaginal delivery (Group 2) or an elective caesarean delivery (Group 3). Method In all, 107 patients who suffered a third degree tear between 1981 and 1993 were contacted with a validated questionnaire. The two control groups comprised of 125 patients in each category. Those who responded to the questionnaire were invited for anorectal physiology studies and endoanal ultrasound. Results Of the total number contacted, 54, 71 and 54 women from the three groups returned the completed questionnaire. In the three groups, a total of 28 (53%), 13 (19%) and six (11%) complained of anal incontinence (P < 0.0001) respectively. Comparison of quality of life scores between the groups showed a poorer quality of life in those who suffered a tear (P < 0.0001). In addition, in spite of primary repair, 13 (59%) patients in group 1 showed a persistent sphincter defect compared to one (4%) occult defect in Group 2 and none in Group 3. Conclusion Our study indicates that long-term results of primary repair are not encouraging. It therefore emphasizes the importance of primary prevention and preventing further sphincter damage in those who have already suffered an injury (during subsequent deliveries).