In Latin America, where the overall rate of cesarean section is among the highest in the world, cesarean rates at 17 hospitals that mandated a second opinion for nonemergency procedures decreased by 7% in relation to those at comparison hospitals.1 The benefit was due mainly to a 13% relative reduction in the rate of intrapartum cesarean sections (those performed during labor). Implementation of the second opinion policy was not associated with any adverse changes in measures of maternal and neonatal well-being. In addition, women delivering at hospitals with the second opinion policy were as satisfied with their care as were women delivering at hospitals following usual policy.
Hospitals were eligible for the study if they had a cesarean section rate of at least 15% and had more than 1,000 deliveries per year. Matched hospitals were randomly assigned to an intervention group or a control group. Hospitals in the intervention group implemented a mandatory policy that when an attending physician decided a woman needed a nonemergency cesarean section, the physician had to obtain a second opinion from another physician of equal or higher clinical status. The consulting physician applied evidence-based guidelines and discussed the case with the attending physician, who made the final decision. Changes in group outcomes between the six-month periods preceding and following implementation of the intervention were compared.
Analyses were based on data from 149,276 women who delivered at 34 hospitals in Argentina, Brazil, Cuba, Guatemala and Mexico between October 1998 and June 2000. Nearly all of the hospitals were public or nonprofit. Data collected during the first six-month period revealed that intervention and control hospitals had similar overall cesarean section rates (26% and 25%, respectively) and proportions of women who had had a previous cesarean section (14% at each), but intervention hospitals had a higher overall rate of intrapartum cesarean section (17% vs. 15%) and a higher proportion of women having their first birth (38% vs. 34%).
Hospitals in the intervention group experienced a small but significant reduction in the rate of all nonemergency cesarean sections relative to that of hospitals in the control group (relative rate reduction, 7%). This reduction reflected a relative reduction of 13% for the intervention hospitals in the rate of intrapartum cesarean sections and a relative increase of 2% in elective procedures. Relative reductions also varied according to the reason for the cesarean section and were greatest for those done for maternal health (29%), fetal distress (22%) and slow progression of labor (20%).
Intervention hospitals did not differ from control hospitals with respect to changes in rates of stillbirth, neonatal mortality, perinatal mortality or admission of neonates to an intensive care unit for more than a day. Similarly, differences between the two groups in changes in rates of operative vaginal delivery, maternal postpartum intensive care unit admission for more than a day and maternal death were statistically indistinguishable.
In hospitals in the intervention group, second opinions were obtained for 88% of nonemergency cesarean sections after implementation of the policy. The consulting physician agreed with the attending physician 96% of the time. Overall, the second opinion led to a change in the attending physician's initial decision to perform a cesarean section in only 2% of cases.
Of women at intervention and control hospitals whose physicians had initially scheduled them for a cesarean section, similar proportions were told or saw that their physician consulted another physician; 90% of these women reported that knowing of this consultation made them feel "better," with no difference between groups. The vast majority of women in both groups said they would use the same hospital for future deliveries (88% and 87%, respectively) and would recommend it to other pregnant women (91% and 93%).
Without knowledge of the study's results, 54% of physicians at intervention hospitals rated the mandatory second opinion policy as "effective" or "very effective" in reducing the rate of cesarean section. Furthermore, 87% thought it would be a feasible strategy for public hospitals, and 91% would recommend its use in such institutions; those figures were 41% and 65% in regard to private hospitals.
"The implementation of a mandatory second opinion policy in public hospitals on an indication of intrapartum caesarean section could prevent 22 caesarean sections for every 1,000 women in labour without harmful effects [to] the baby or the mother," the researchers argue. They note that hospital staff and policymakers will need to weigh the benefits and costs of such a strategy to determine if it is appropriate for their institution.
1. Althabe F et al., Mandatory second opinion to reduce rates of unnecessary caesarean sections in Latin America: a cluster randomised controlled trial. Lancet, 2004, 363(9425):1934-1940.