Episiotomy rates in primiparous women in Latin America: hospital based descriptive studyBMJ 2002; 324 doi: http://dx.doi.org/10.1136/bmj.324.7343.945 (Published 20 April 2002) Cite this as: BMJ 2002;324:945
- Fernando Althabe (), researcher in perinatology,
- José M Belizán, director,
- Eduardo Bergel, epidemiologist
- Latin American Center for Perinatology, Pan American Health Organization, World Health Organization, Casilla de Correos 627, Montevideo 11000, Uruguay
- Correspondence to: F Althabe
- Accepted 6 December 2001
Current scientific evidence shows that routine episiotomy is not justified: it has no benefit for mother or infant, increases the need for perineal suturing and the risk of complications to the healing process at seven days post partum, produces unnecessary pain and discomfort, and has potentially harmful long term effects.1 2 3 We report rates of episiotomy in primiparous women in Latin American hospitals according to characteristics of hospitals and caregivers.
Participants, methods, and results
We conducted a hospital based descriptive study based on data routinely collected in a perinatal information system.4 We analysed data from 122 hospitals in 16 Latin American countries that had reported 416 852 deliveries between 1995 and 1998. We selected hospitals reporting more than 35 spontaneous vaginal deliveries in primiparous women, which is the sample size required to give a 95% confidence interval of 10% either way for an episiotomy rate of 90%. This selection comprised 105 hospitals in 14 countries, which reported 94 472 spontaneous vaginal deliveries in primiparous women. We report episiotomy rates by hospital, with medians (interquartile ranges) as a summary measure.
In 91 hospitals (87%) episiotomy rates were higher than 80% and in 69 hospitals (66%) they were higher than 90%. The overall median rate was 92.3%, and median rates by country varied between 69.2% and 96.2% (table). Episiotomy rates were similar in primary, secondary, and tertiary hospitals (89.8%, 91.6%, and 92.7%, respectively) and for public, private, and social security hospitals (90.2%, 96.4%, and 95.6%, respectively). The rates were also similar according to who attended the delivery (doctors in 91.4%, midwives or nurses in 93.6%, and students in 93.7%).
Nine in every 10 primiparous women who gave birth spontaneously in hospitals in Latin America between 1995 and 1998 had an episiotomy. This figure was similar in public and private hospitals, primary care and referral hospitals, and deliveries attended by doctors or midwives. If a rate of 92% is applied to the 2.35 million primiparous women giving birth spontaneously in Latin American hospitals per year, this means that 2.17 million primiparous women per year receive an episiotomy.
The results were obtained from a database in which routine data are collected, and therefore have some potential limitations. The rates might have been affected by different outcome definitions among hospitals, but in maternity health services, episiotomy has a unique definition. In view of the high rates, it is possible that episiotomy rates are over-reported by recording perineal tears as episiotomies. But this is unlikely in the perinatal information system, because outcomes are marked separately in the data collection form. The rate of missing values in this dataset was below 1%.
Seventy one per cent of hospitals in the database were located in Argentina and Uruguay. The results might therefore represent standards of care in hospitals in those two countries rather than elsewhere. The similarities between hospitals in rates of episiotomy are, however, unlikely to be indicative of bias in hospital selection but of a common standard practice in the use of the procedure in most of Latin America's hospitals.
This situation is inadmissible in the light of the current evidence. The challenge now is to design and test an original intervention directed to women and caregivers to change the use of episiotomy in Latin American hospitals.
We thank all health workers and health related workers in many settings of Latin America and the Caribbean for their efforts to collect and send the data from the perinatal information system.
Contributors: FA and JMB participated in the design, execution, and analysis of the study. EB participated in the analysis of the study. The manuscript was prepared by FA, JMB, and EB. Roberto Porro and Luis Mainero provided technical assistance with the perinatal information system database. FA is the guarantor.
Editorial by Langer and p 942
Funding Latin American Center for Perinatology, Pan American Health Organization, World Health Organization.
Competing interests None declared.