How sensitive and acceptable is self-administered testing for group B streptococcus (GBS) in pregnant women?
There is no difference in the sensitivity of swabs for detecting GBS when collected by patients vs. physicians; therefore, either method can be used. (Strength of Recommendation: C, based on cohort and randomized studies.) Patients do not express a consistent preference for self-collection of samples vs. collection by a physician.
Four trials compared self- and physician-collected swabs to detect GBS in pregnant patients.1–4 Each study recruited women at 35 to 37 weeks' gestation. The reference standard in each study was a positive GBS culture by either collection method. Agreement between the two methods was high. GBS prevalence ranged from 12% to 19%. In two of the four trials, cultures were not incubated in an enrichment broth, which may have resulted in a lower GBS prevalence.2,4
In a 2006 prospective trial, 330 women (average age = 27.2 years) at a family medicine prenatal care clinic in Canada were randomly assigned to clinician-collected vaginorectal swabs followed by self-collected swabs, or vice versa.1 The women received written instructions and a diagram of the self-collection procedure. Clinicians were not blinded to the testing sequence. Both collection methods yielded comparable sensitivity (97% clinician-collected vs. 87% self-collected; Cohen's kappa agreement = 0.9; 95% confidence interval [CI], 0.8 to 1). Patients were slightly more likely to prefer self-collection. There were 56 patients who declined to participate. Participants were four times more likely than nonparticipants to have completed high school (odds ratio [OR] = 4; 95% CI, 1.2 to 13.4), and those who declined were more likely to prefer collection by a clinician (OR = 6.9; 95% CI, 2.8 to 17.1).
A 2008 nonrandomized, prospective study of 600 women (average age = 31 years) compared self-collected GBS culture results with those obtained from a subsequent physician-collected sample.2 Patients were recruited from public and private prenatal clinics in Ireland, and clinic staff explained the self-collection procedure to the patients. As in the previous study, both collection methods had high sensitivity (94% clinician-collected vs. 84% self-collected; Cohen's kappa agreement = 0.9; 95% CI, 0.8 to 0.9). The women preferred collection by a physician over self-collection (43% vs. 28%; 28% reported no preference). The most common reason given was concern about collecting the sample correctly.
A 2000 nonrandomized, prospective study at a military hospital in Colorado alternately assigned 250 women to physician-collected swabs followed by self-collection, or vice versa.3 Patients received written instructions and a diagram of the self-collection procedure. There was no difference in first vs. second sample sensitivities in either group (93% clinician-collected vs. 84% self-collected; Cohen's kappa agreement = 0.8; 95% CI, 0.7 to 0.9). The women slightly preferred self-collection (58% vs. 42%).
A 2009 retrospective analysis compared GBS culture results from 293 patients who performed self-collection with those of 507 patients in whom physicians collected the samples.4 The patients attended two community clinics in Colorado; 75% were Hispanic and 85% were below 200% of the federal poverty level. GBS prevalence was slightly higher in the patient-collected samples (13% vs. 11% in clinician-collected samples), but the difference was not statistically significant. The patients did not receive uniform instructions for self-collection. The authors did not report patient preferences.