Chlamydia Surveillance in the United States: New Analytic Approaches and Alternate Considerations for Monitoring Trends in Disease Burden Open Access

Satterwhite, Catherine Lindsey (2011)

Permanent URL: https://etd.library.emory.edu/concern/etds/hx11xf840?locale=en
Published

Abstract

Over 1.2 million cases of chlamydia were reported in the U.S. in 2009. Chlamydia may
lead to pelvic inflammatory disease (PID), ectopic pregnancy, and infertility. Annual
screening is recommended for sexually active women aged <25 years. Reported case
rates have increased likely due to increased screening and improved test technology.
Other data suggest prevalence has decreased. Studies in this dissertation explored new
analytic approaches to enhance the utility of U.S. chlamydia surveillance data.

Using data from the Infertility Prevention Project (IPP), a national chlamydia screening
program, chlamydia positivity trends among women aged 15-24 years were assessed in
two settings: family planning clinics (2004-2008) and prenatal clinics (2004-2009).
Women seeking prenatal care are routinely screened and less influenced by health care
seeking behaviors. After evaluating the linearity of year, a multivariate, correlated,
longitudinal data analysis with a random intercept was conducted using the clinic as the
unit of analysis and treating year as continuous. The odds ratio (OR) associated with a
single year change (1.00; 95% confidence interval [CI]: 0.99, 1.00) suggested that
positivity did not change from 2004-2008 in family planning clinics. In prenatal clinics,
positivity decreased from 2004-2009 (OR: 0.93 per year, 95% CI: 0.92, 0.95). A survey
assessing chlamydia screening practices in prenatal clinics was also conducted. Of the
166 clinics completing the survey, 163 (98.2%) had documented chlamydia screening
criteria.

Because the purpose of chlamydia screening is to prevent adverse reproductive outcomes,
data from 2 large health maintenance organizations, Group Health Cooperative (GH) and
Kaiser Permanente Colorado (KPCO), were analyzed to develop a PID case-finding
algorithm. A classification and regression tree analysis identified 2 main predictors
beyond ICD-9 codes: PID-specific treatment and age 15-25 years. When using ICD-9
codes alone to identify PID cases, the positive predictive value (PPV) was 78.8% in GH
and 79.1% in KPCO. Algorithm PPV improved to 86.9% and 84.5%, respectively.

Findings support previous analyses suggesting that chlamydia prevalence is not
increasing and may be decreasing. Maximizing the utility of administrative data by
examining the privately insured prenatal population and monitoring adverse outcomes
can improve surveillance around this important public health issue.

Table of Contents

Table of Contents

1---Overview and Rationale
6---Chapter 1: Background

23---Chapter 2: How can family planning data from the Infertility Prevention
Project, a national chlamydia screening program, be better used
to monitor trends in chlamydial infections?

44---Chapter 3: Is the prenatal population an alternate population for monitoring
chlamydia trends?
69---Chapter 4: How can administrative data be better used to detect PID cases,
an important adverse outcome of untreated chlamydial infections?
89---Chapter 5: Discussion and Summary
96---References
102---Appendices

About this Dissertation

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research field
Keyword
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files