Quality of care for pelvic inflammatory disease: room for improvement.

Pelvic inflammatory disease (PID) is a clinical syndrome characterized by upper genital tract inflammation and frequently associated with sexually transmitted or genital tract microorganisms. The Centers for Disease Control and Prevention (CDC) has developed STD treatment guidelines intended to assist health care providers in selecting antibiotic treatment strategies based on objective clinical evidence.1 Several studies over the past 2 decades have demonstrated poor provider adherence to recommended treatment for PID.2–5 The authors of these reports called for aggressive quality improvement measures to improve adherence to PID treatment guidelines. With an ongoing focus on improving quality of care, one would expect to see improvement in provider adherence to PID antibiotic regimens over the past decade. In this issue, Shih and colleagues report on provider adherence to CDC guidelines for outpatient treatment of PID in emergency and urgent care settings.6 The authors used data from a national database (National Hospital Ambulatory Medical Care Survey; NHAMCS) to determine whether provider selection of antibiotics for outpatient PID was consistent with CDC guidelines during the 1999–2006 period. The authors concluded that a guideline-recommended antibiotic regimen was provided only 30% of the time, with partially appropriate antibiotics provided 47% of the time, and completely inconsistent treatment (or no treatment) provided 22% of the time. The authors conclude that provider adherence to PID antibiotic treatment guidelines remains low and has not improved significantly since the previous studies on this topic. Why is adherence to the guidelines so low? Although research has consistently shown poor adherence to PID antibiotic treatment guidelines, little is actually known about why providers are not adherent. It could be that many providers are not aware of the latest guidelines from CDC or other major organizations. Ubiquitous internet access now provides healthcare providers with nearly immediate access to latest treatment guidelines (one needs only to type “PID treatment guidelines” into an internet search engine, and the latest CDC guidelines will be the first link). However, many providers may use other sources of information (eg, review articles or text books) or may continue to practice in a manner that they learned previously. Alternatively, it could be that providers either do not agree with the guidelines, find them impractical, or are aware of new data that has emerged since the latest version of the guidelines were written. For example, the Shih study found that one of the most common reasons for provider nonadherence was the provision of azithromycin, rather than doxycycline, as part of the outpatient antibiotic regimen. Although azithromycin was not included as a recommended treatment for PID in the 2006 CDC guidelines, it appears to be an effective treatment option (although the optional dose and duration remains unknown).1,7 The recommended oral regimens for PID generally require twice-daily medication for 2 weeks, and young women with PID often have significant challenges with adherence to these antibiotic regimens.8 Another possible explanation for the low adherence rates in the Shih study could be measurement error related to misclassification of PID and/or the antibiotic regimens used. In the NHAMCS survey, use of provider-coded diagnosis of PID has not been validated against actual medical records or other means; thus, providers completing the survey could have indicated a PID diagnosis for a variety of pelvic complaints. However, other studies that used chart review or questionnaires to identify PID have found similar rates of nonadherence to CDC guidelines.2,4,5 Some providers in the NHAMCS did not report providing any antibiotics for PID, and while failure to provide any antibiotics is clearly inappropriate, it is possible that either the PID diagnosis was not correct or that the provider did not accurately list all medications provided on the survey form. Should antibiotic treatment that is only partially in agreement with guidelines always be classified as poor quality of care? Recent clinical trials provide evidence of efficacy for several