Abstract
STUDY OBJECTIVE: To assess the predictive utility of obstetric risk factors for identifying before the onset of labour those women at high risk of obstetric complications in a developing world setting, where home deliveries predominate and emergency transport is scarce. DESIGN: Risk factors derived from two population based, case-control studies (one of cephalopelvic disproportion and one of post partum haemorrhage), carried out in Zimbabwe were used to construct weighted and unweighted scores, a variety of screening algorithms, and sets of probabilities estimated from logistic regression models. These screening tests were evaluated for sensitivity, specificity, positive predictive value, and "cost" (the proportion of the population testing positive). Each complication was evaluated separately and the two were then pooled. PARTICIPANTS: All were Harare residents with singleton, vertex deliveries and spontaneous onset of labour. A total of 201 experienced cephalopelvic disproportion, 150 had post partum haemorrhage, and 299 had normal, unassisted deliveries. MEASUREMENTS AND MAIN RESULTS: Largely because of the very low incidence of the two complications studied (1% or less), positive predictive values were low (less than 7%). Holding "cost" constant at 10%, a screening test for cephalopelvic disproportion could predict 42.3% of cases compared with only 35.0% of those with post partum haemorrhage. Weighted scores had little advantage over unweighted ones, and probabilities from the logistic regression models did not differentiate cases from controls very well. CONCLUSIONS: With simple algorithms based on maternal height, parity, and obstetric history, more than one third of women at risk for potentially fatal complications could be identified at relatively small cost to themselves or the health care system.