We thank Dr Alouini for his comments and are grateful for the opportunity to clarify a number of points from our work, ‘Familial risk of obstetric anal sphincter injuries: registry-based cohort study’. In his Letter to the Editor, Dr Alouini points out several factors involved in the risk of obstetric anal sphincter injuries (OASIS) that we did not consider in our paper. These risk factors include the mode of delivery, the anatomy of the ano-vulvar region, the type of episiotomy and the change in maternity units between mothers and their daughters or between sisters.
However, as we are discussing the impacts of different confounders, it is most important to have in mind that a confounder is not only associated with the outcome (OASIS in the current birth), but also the exposure (history of OASIS in a relative).
The mode of delivery: All of our analyses were adjusted for instrumental delivery. We did not specify which type of instrument (forceps or vacuum) was used because it is unlikely that it would influence or change the relative risk. After reanalysing the risk of OASIS in daughters whose mothers had OASIS, adjusting separately for vacuum and forceps, the results were exactly the same (data not shown).
It is possible that, in families with common use of forceps, the percentage of OASIS would be higher, without affecting the relative risk. In Figure 1, the use of forceps and vacuum in Norway during the study period is demonstrated. The use of forceps in Norway has decreased considerably in the second generation.
The ano-vulvar distance: Our database did not include information about the ano-vulvar distance. However, the association between the perineal body length and sphincter injuries has not been reported consistently by previous studies.[2, 3]
Episiotomy: Mediolateral episiotomy has traditionally been used in Norway. The angle of the cut in episiotomy is strongly associated with the risk of OASIS. In a reanalysis, adjusting for episiotomy had no effect on intergenerational recurrence (data not shown).
The risk of OASIS in different maternity units: Dr Alouini is correct when he suggests that the obstetric routine influences the risk of OASIS. In a previous study, we reported that hospitals with a number of deliveries of more than 3000 per year showed a higher incidence of OASIS. It is possible that, if the mother had given birth in a low-incidence maternity unit and the daughter had given birth in a high-incidence maternity unit, it would interfere with our results. For this purpose, we additionally adjusted the risk of OASIS between mothers and their daughters for the change in maternity unit between generations. The adjustment did not alter the results.
As we pointed out in our conclusion, the observed risk of OASIS between generations and siblings should be interpreted cautiously, as bias from unmeasured confounding factors may have an impact on the findings. However, we are confident that the type of instrumental delivery, the angle of episiotomy and the change in maternity unit between generations do not represent bias in our study.