The association of sociodemographic differences with clinical management following a diagnosis of short cervix.

Introduction: Following diagnosis of a short cervix from 16 through 23 weeks there is considerable variation in clinical practice. We sought to evaluate the association of patient sociodemographic factors with variation in clinical management following diagnosis of a short cervix. Methods: This was a non-prespecified secondary analysis of the randomized Trial of Cervical Pessary for Short Cervix in Singleton Pregnancies (TOPS), in which individuals with singleton pregnancies with a cervical length (CL) ≤20 mm were randomized to either a pessary or usual care. We assessed recommendations for vaginal progesterone, additional CL surveillance, pelvic rest, bed rest, and cerclage. For individuals enrolled in the trial, additional CL assessment and cerclage were discouraged in the absence of significant cervical dilation; activity modification and vaginal progesterone were at provider's discretion. Sociodemographic factors assessed were self-reported race and ethnicity, marital status, employment status, years of education, and type of insurance. Multivariable regression models examined the association between each sociodemographic factor and management, adjusting for other sociodemographic factors, CL at enrollment, clinical center, and treatment arm; odds ratios (ORs) and 95% confidence intervals (CIs) were determined. Results: Of the 446 patients (82% of 544 enrolled in TOPS) with at least one post enrollment visit, 95.5% received vaginal progesterone, 28.0% had at least one additional CL assessment, 55.8% were placed on pelvic rest, 19.5% on activity restriction, and 7.4% underwent cerclage. The high rate of vaginal progesterone usage precluded evaluation of association with sociodemographic factors. Non-Hispanic Black and Hispanic patients had lower odds of repeat CL assessment compared with Non-Hispanic White patients (adjusted OR [aOR], 0.47; 95% CI, 0.24, 0.91; aOR, 0.45; 95% CI, 0.21, 0.99, respectively). No other sociodemographic factors were associated with repeat CL. Patients with private insurance, but no other factor, had higher odds of pelvic rest being recommended (aOR, 2.08; 95% CI, 1.25, 3.45). None of the examined sociodemographic factors were associated with recommendations for bed rest or cerclage. Conclusion: Following diagnosis of a short cervix, vaginal progesterone is prescribed for nearly all patients. Non-Hispanic Black and Hispanic patients were less likely to have a repeat CL assessment following diagnosis of a short cervix and those with private insurance were more likely to have pelvic rest recommended. Variation in the frequency of these interventions may reflect how sociodemographic and socially constructed factors can affect clinical decision making and implementation of interventions in the absence of definitive guidance. Trial registration: ClinicalTrials.gov Identifier: NCT02901626.