Adherence to New CDC Chlamydia Treatment Guidelines in Adolescents: Effective Uptake and Equitable Implementation

To the Editor:

Chlamydia trachomatis (CT) is the most common bacterial sexually transmitted infection (STI) in the United States with 1.6 million cases reported in 2021. ¹ Approximately, two-thirds of annual cases are among adolescents and young adults. ² Untreated Chlamydia infections can lead to serious reproductive health morbidity, including pelvic inflammatory disease, ectopic pregnancy, and infertility. Until 2021, the first-line antimicrobial treatment regimens were single-dose azithromycin or a week-long doxycycline course, which both had cure rates >96% for urogenital chlamydia. ³

However, in July 2021, due to growing evidence of inferiority of single-dose azithromycin for extragenital infections, ⁴ the CDC updated chlamydia treatment guidelines recommending the multidose doxycycline (100 mg twice daily for 7 days) as the preferred regimen and classifying azithromycin as an alternative. ⁵

Clinical practice guideline implementation is a challenging process, and success is dependent on intervention at the health system, clinic, practitioner, and patient levels. ⁶ Previous studies demonstrate both slow initial uptake and lack of sustained change for new clinical practice guidelines. ^6,7 In addition, there is well-documented evidence of inequities by race, insurance, and gender in antimicrobial prescribing practices. ^8,9 Thus our research aims were to (1) determine adherence to the 2021 CDC STI treatment guidelines for Chlamydia in a pediatric primary care network in the first 6 months after guideline release and (2) investigate for evidence of inequities in antibiotic delivered by patient or clinic factors.

We analyzed electronic health record (EHR) data from a 31-clinic pediatric primary care network serving urban and suburban areas in the mid-Atlantic region of the United States. Two of the urban clinics receive Title X federal funding and provide confidential sexual health services while the remaining 29 clinics provide standard pediatric primary care. After guideline introduction, the rationale for clinical practice change was disseminated through lectures to adolescent medicine providers in August 2021 and all primary care providers in December 2021.

Our study sample included patients who tested positive for Chlamydia between July 1st, 2019, and January 31st, 2022. To assess trends in prescribing practices, we analyzed data from three periods: July–January 2019–2020, July–January 2020–2021, and July–January 2021–2022. We chose matching noncontiguous periods to account for seasonal variations in care. We excluded emergency department visits and visits without medication orders. Our primary outcome was antimicrobial prescribed, azithromycin or doxycycline, after a positive chlamydia nucleic acid amplification assay.

To assess for inequities in guideline adherence, we utilized data from the post-guideline period only, comparing antimicrobial prescribed by key patient, clinic, and clinician characteristics. Patient characteristics included age in years at encounter and sex assigned at birth. Race and ethnicity correspond to “observed race” recorded by the staff at the time of registration rather than self-identified race as the health system did not capture self-identified race during the study period. Insurance status at encounter was categorized as either Medicaid, private, self-pay, or missing. Clinic setting was categorized as Title X urban, Non-Title X urban, or suburban. Clinicians were categorized as attending physicians, nurse practitioners, nurses, or residents/fellows.

Descriptive statistics summarized patient and clinical characteristics. Analysis of variance (ANOVA) with post hoc test assessed differences in linear doxycycline prescription trends between the three periods. To examine associations between patient and clinic characteristics on antimicrobial received, we used mixed effects logistic regression models and estimated odds ratios accounting for random effects of clinics and patients.

We generated separate models to measure the associations between patient and clinical characteristics on treatment received. A final multivariable model contained both patient and clinic characteristics. A p value <0.05 was considered statistically significant. All statistical analyses were performed using R Statistical Software (v4.1.2; R Core Team 2021). This research was approved by our institutional review board.

A total of 1077 chlamydia-positive encounters were identified during the 6-month period across the three time periods (Fig. 1). After introduction of 2021 CDC STI treatment guidelines, 54% (n = 161) of chlamydia cases were treated with doxycycline and 46% received azithromycin. This 54% doxycycline prescription rate was significantly different from the 6% and 9% rates in 2019 and 2020, respectively [f(2) = 30.69, p < 0.01]. Post-hoc testing revealed significant pairwise differences between 2020 and 2021, with an average difference of 44.9% (p < 0.01) and between 2019 and 2021, with an average difference of 49.8% (p < 0.01).

OPEN IN VIEWER

FIG. 1.

Comparison of percentage of doxycycline prescriptions over 6-month time periods across 3 years 2019–2021.

After guideline release (July 2021 to January 2022), we identified n = 344 chlamydia-positive encounters, 298 (86%) of which met our inclusion criteria. Patients (n = 275) were 86% Black or African American, 98% Non-Hispanic, 70% assigned female at birth, with a median age of 17.5 years (interquartile range:16.6–18.4). Encounter-level demographics and clinical characteristics by type of antimicrobial received are displayed in Table 1. Insurance type, clinic setting, and provider size showed significant associations with antimicrobial received in bivariate analysis.

In the multivariable model, there were no significant associations between any patient-level characteristics and receipt of doxycycline versus azithromycin. In the model assessing associations with clinic characteristics, each additional provider in a practice was associated with a 3% higher odds of doxycycline prescription (adjusted odds ratio 1.03, 95% confidence interval: 1.00–1.06). In the combined model, neither patient nor clinic characteristics demonstrated significant associations with doxycycline versus azithromycin prescription.

In summary, we identified a greater than sixfold absolute increase in doxycycline, compared with azithromycin, treatment of chlamydia after the introduction of the 2021 CDC STD treatment guidelines. Our findings demonstrate rapid early adoption of clinical practice guidelines in the pediatric primary care setting. Notably, we did not identify inequities in antimicrobial prescription by race, ethnicity, age, payor, or clinic and practice characteristics. These data represent a promising early signal for uptake and equitable diffusion of clinical practice guidelines in pediatric settings.

Although this early shift in prescribing practices is promising, 46% of chlamydia episodes after guideline introduction continued to be treated with azithromycin. Future research is needed to determine whether persistent azithromycin prescription represents poor guideline diffusion, or conversely, appropriate clinical judgment. For adolescent patients, for whom medication adherence is typically worse than adults, shared decision making between a clinician and patient may lead to rightly prioritizing single-dose treatment over providing a multidose regimen with challenging tolerability.

Prior systematic review and the implementation science literature demonstrate the need for multilevel and multistrategy efforts to achieve and sustain optimal guideline implementation, including educational efforts, audit and feedback of prescribing practices at the individual clinician and practice level, and shifts in organization culture. ^6,10 In our health system, the only intervention involved clinician education, which is typically insufficient to lead to widescale practice change. In the face of emerging practice guidelines, health systems will need sufficient funding and resources to consider interventions such as EHR clinical decision support to optimize guideline diffusion and equitable application.

This cross-sectional descriptive study cannot assess the causal effects of provider education, and we are unable to assess the impact of other learning efforts undertaken by clinicians that may have influenced practice change. In addition, our sample was overwhelmingly non-Latinx Black individuals, which may have limited our ability to detect inequities by race. This homogeneity is likely reflective of upstream health inequities that lead to higher rates of screening in Black young women.

Lastly, it is important to note that these guidelines were released during the COVID-19 pandemic, ^11,12 during a period of massive disruption to STI services. The decreased chlamydia screening rates during the pandemic, which led to likely artifactual decreases in prevalence, may have also impacted treatment choice. It will be critical to assess trends in treatment as screening volumes continue to normalize.

In summary, we identified rapid and equitable early uptake of the CDC clinical practice guideline change supporting doxycycline therapy as the first-line antimicrobial for treatment of uncomplicated Chlamydia trachomatis. Future research is needed to assess whether these changes continue and are sustained in health systems.