Self-sampling for group B streptococcus in women 35 to 37 weeks pregnant is accurate and acceptable: a randomized cross-over trial.
OBJECTIVE: In Canada, screening for group B Streptococcus (GBS) in pregnant women is recommended at 35 to 37 weeks’ gestation. Since there is normally no other indication for pelvic examination at this stage of pregnancy, women may be more comfortable performing the test themselves. We assessed the accuracy of self- sampling versus clinician sampling for GBS and women’s preference for each collection method. METHODS: Consecutive patients presenting between October 2003and April 2005 to a maternity centre in Hamilton for their 35- to 37-week prenatal visit were randomly allocated to having vaginal-rectal swabs self-collected, and then collected by a clinician, or to having the swabs clinician-collected, and then self-collected. The main outcomes were prevalence of infection and sensitivity of the two methods. Other analyses compared women who refused participation in the study with those who participated, and preference for sampling method before and after conducting the tests. RESULTS: Of the 386 women approached, 330 (85.5%) agreed to participate. The prevalence of GBS was 17.0% (56/330) and 18.8% (62/330) in the self-obtained and clinician-obtained specimens respectively (difference =1.8%; 95% confidence intervals [CI] -2.0-6.0). Sensitivity was 87.5% (95% CI 77.0-93.8) and 96.9% (95% CI 88.7-99.8) for the self-obtained and clinician-obtained specimens respectively. Women who declined to participate in self-sampling were significantly more likely not to have completed high school and to prefer clinician sampling. Prior to testing, 79% of women preferred self-sampling or had no preference. Preference was unchanged for approximately two-thirds of women after sampling. CONCLUSION: Self-sampling for GBS is an accurate and acceptable alternative for the majority of pregnant women. Less-educated women may be hesitant to self-sample, and clinician sampling should remain an option.