Universal HIV screening of pregnant women in England would be cost-effective in areas of high prevalence

OBJECTIVE To estimate the cost-effectiveness of a voluntary HIV-screening programme for all pregnant women. DESIGN Cost-effectiveness analysis. Only costs to the National Health Service were considered. SETTING England. INTERVENTION Serological screening for HIV antibodies in pregnant women who arc unaware of being infected. The sensitivity and specificity of the initial test are 100 and 99.9%, respectively, with positive results confirmed with additional tests. Preventative interventions are offered to HIV-positive women. ASSUMPTIONS All women agree to be tested. Uptake of interventions in HIV-positive women: zidovudine 75%, elective cesarean section 40%, emergency cesarean section 15%, formula feeding 95%. Mother-to-child transmission rates: 18% for vaginal delivery- and 10% for cesarean section delivery in the absence of zidovudine treatment, 8 and 6%, respectively, with zidovudine treatment, and 14% for breast feeding. Life expectancy: normal 77 years, child with HIV 7 years. Mother will gain 1 year symptom-free with earlier treatment. Cost assumptions: screening (test £4 plus pre-test counselling) £40, post-test counselling of positive women £50, zidovudine treatment for mother (14 weeks) and baby (6 weeks) £600, vaginal delivery £400, elective cesarean section £1000, emergency cesarean section £1300, formula feeding £800, lifetime health-care costs for a child with HIV £178,300, post-partum treatment of mother £12,300 per year. MAIN OUTCOME MEASURES Net cost (costs of screening and preventative interventions minus costs of averted health care) per life year gained (mother and child). MAIN RESULTS At a prevalence of 15 women unaware of being HIV infected per 10,000 pregnant women, for each HIV-positive women detected, the cost of screening is £26,700, the cost of the extra interventions is £1300 and the cost of care for HIV-infected children is £29,100 less, compared to no screening. The net cost per life year gained is £3300. If the prevalence is lower at 1/10,000, the cost per life year gained increases to £114,000. If pre-test counselling is integrated into routine prenatal care, so that the cost of screening involves only the cost of the test, the cost per life year gained would be £7300 at the lower prevalence and cost saving at the higher prevalence. In London, where the prevalence of unaware HIV-infected pregnant women is 14/10,000, the costs of screening and prevention would equal the costs averted if the cost of screening could be reduced from £40 to £22 per person. CONCLUSION A universal HIV-screening programme for pregnant women would be cost-effective in areas of high prevalence if the pre-test counselling were integrated into routine prenatal care. © 2000 Harcourt Publishers Ltd.

Duke Scholars

Author Evan Robert Myers Obstetrics and Gynecology, Women's Community & Population He ...