Maternal anthropometry‐based screening and pregnancy outcome: a decision analysis

Abstract
To assess the impact of screening and intervention based on maternal height, prepregnancy weight and weight during weeks 16-19 or 24-27 in reducing adverse pregnancy outcomes (IUGR, preterm birth and assisted delivery) in developing country settings. Decision analysis based on a recent multicentre WHO collaborative study of maternal anthropometry and pregnancy outcomes and meta-analyses of controlled clinical trials of balanced energy/protein supplementation (for IUGR and preterm birth) and support from caregivers during labour (for assisted delivery). Subjects for the analysis comprised pregnant women from Cali, Colombia (1989, n = 4598); urban and rural Pune, India (1990, n = 4307); and urban and rural Myanmar (1981-82, n = 3542) followed until delivery. Seven to 45% of pregnant women had positive screens, with preventive fractions (PFs) ranging from 0.034 to 0.109 for IUGR, 0.027-0.082 for preterm birth and 0.011-0.105 for assisted delivery. Screening prevention ratios (SPRs = ratios of the number of women treated to the number of cases of adverse outcome prevented) are high in all three study settings for preterm birth and assisted delivery (range 22.8-115.7) and low in settings with a high prevalence of the adverse outcome and high specificity of the anthropometric measure (India for IUGR, range 7.0-8.0). Sensitivity analyses demonstrate a marked linear fall in PF and an exponential rise in the SPR as the relative risk associated with intervention increases (i.e. as the protective benefit of intervention decreases) from 0.60 to 0.95. A maternal anthropometry-based 'risk approach' is unlikely to result in a major reduction in adverse pregnancy outcomes in developing country settings. For risk-free and inexpensive interventions (e.g. caregiver support during labour), a better strategy would be to forego screening and instead treat all pregnant women.