BACKGROUND: Pelvic Inflammatory Disease (PID), an upper genital tract inflammation, is often caused by sexually transmitted infections (STIs) such as chlamydia and gonorrhea and may lead to infertility and ectopic pregnancy. PID is not nationally notifiable and data available to monitor PID rates are limited. We explored the utility of using discharge datasets for surveillance of PID in American Indian/Alaska Native (AI/AN) women, a population disproportionately affected by STIs.
METHODS: We combined Indian Health Service inpatient hospitalization and outpatient visit discharge datasets to analyze acute PID in female AI/ANs ages 15-44 years during 2001-2014 using ICD-9-CM codes. We calculated overall rates of PID and stratified by age group, region and year. Rates were calculated per 100,000 persons, and Poisson regression was used to identify trends. Among patients with more than one PID-associated visit, we calculated the median time in days between visits. We calculated the proportion of concurrent STI diagnoses during the same visit as a PID diagnosis.
RESULTS: A total of 41,789 PID-associated visits were observed from 2001-2014. Ninety-three percent of visits were outpatient (N=38,946), and seven percent (N=2,843) were inpatient. The highest rate of infection was among 25–34 year-old females (Rate=954; N=15,273) and among individuals from the Alaskan region (Rate=1585; N=6,766). Rates significantly decreased for PID overall (2001 Rate=1,090; 2014 Rate=595; p-value=<.0001) and by all age groups. Most regions showed a significant decreasing trend with the exception of the Alaskan and Northern Plains West regions which remained stable. The Alaskan region showed a high rate of PID cases that ranged from 959 in 2001 to 2,370 in 2010. Among patients with multiple PID-coded visits, the median time between visits was 21 days (Q1=4; Q3=391). Seven percent (N=3,012) of PID cases had a concurrent chlamydia diagnosis, and 2% (N=731) had a concurrent gonorrhea diagnosis. Less than 0.1% of patients had HIV or syphilis diagnosed concurrently with PID.
CONCLUSIONS: Decreasing trends of acute PID were observed among female AI/ANs ages 15-44 years during 2001-2014, although there were exceptions in the Alaskan and Northern Plains West regions. Women 25-34 years of age and women from the Alaska region were mostly affected. Our analysis demonstrates that discharge datasets can be useful in the surveillance and examination of PID-associated trends and defining targeted interventions. Additional research is needed to better determine the number of unique PID cases, as patients may have multiple visits for the same episode.