Assay Results vs Selfreported Chlamydial Infections Does Measurement

Reasons for Focusing on Chlamydia Most common bacterial STI Has clear recommendations for screening

Sources of Chlamydia Estimates of chlamydial infections often based on reported cases, clinic-based studies

Self-reported Infections Limitations of self-reports – Miss asymptomatic infections when screening inadequate Screening recommended

Study Objectives Use National Longitudinal Study of Adolescent Health (Add Health) – Nationally representative

Data Add Health http: //www. cpc. unc. edu/addhealth/ Sample for present analyses: – Wave

Chlamydia Measures Tested for Chlamydia (self-report) – past 12 months Self-report Chlamydia Diagnosis– past

Ratio of Test/Self-report Ratio = biological test prevalence/self-report prevalence – If all infections diagnosed

Measure of Risk Behavior Patterns In previous work, created a measure of risk behavior

Analyses Bivariate analyses by – Sex and race-ethnicity – Risk behavior patterns Weighted percents

Results Prevalence of Sexual Activity Biological Sex % Had vaginal intercourse in past 12

% Prior Chlamydia Testing past 12 months Entire Sample Had Vaginal Intercourse Past 12

Chlamydia Prevalence – By Sex Total Female Male Ratio Tested Pos. / Self-reported %

Chlamydia Prevalence By Race-ethnicity Among Females NH White NH Black NH Asian/PI H/NH Nat

Chlamydia Prevalence By Race-ethnicity Among Males NH White NH Black NH Asian. PI H/NH

Prevalence by Sexual & Substance Use Risk Behavior Pattern % Selfreported % Tested Pos.

Conclusions Prevalence of chlamydia testing is low – Screening guidelines for females not achieved

Infections Missing from Self-reports 4% were infected on interview day, but only 3% self-reported

Possible Reasons Some respondents may not be answering accurately Many infections are undiagnosed –

Acknowledgments This research was supported by grant R 01 -DA 14496 -4 from the

Slides: 24

Download presentation

Assay Results vs. Self-reported Chlamydial Infections: Does Measurement Discrepancy Vary by Level of Risk Behavior? Bonita Iritani, 1 Denise Hallfors, 1 Carol A. Ford, 2 Carolyn Tucker Halpern, 2 William C. Miller 2 1 Pacific Institute for Research and Evaluation 2 UNC - Chapel Hill Funded by NIH-NIDA, Denise Hallfors, PI

Reasons for Focusing on Chlamydia Most common bacterial STI Has clear recommendations for screening among females

Sources of Chlamydia Estimates of chlamydial infections often based on reported cases, clinic-based studies Limitations – Not representative of general population – Miss asymptomatic infections Population-based studies that are conducted typically rely on self-reports

Self-reported Infections Limitations of self-reports – Miss asymptomatic infections when screening inadequate Screening recommended for all sexually active females ≤ 25 years – Miss people with poor access to health care – Some respondents may not report accurately (Harrington et al, 2001) By comparing self-reports with biological test results, could look for infections missing from self-reports

Study Objectives Use National Longitudinal Study of Adolescent Health (Add Health) – Nationally representative sample of young adults To assess prevalence of prior chlamydia testing To compare the prevalence of chlamydial infections identified by self-reports vs. assay results

Data Add Health http: //www. cpc. unc. edu/addhealth/ Sample for present analyses: – Wave 3 2001 – 2002 18 -26 years old – Nonmissing data for chlamydia self-reports and assay results (N=12, 359)

Chlamydia Measures Tested for Chlamydia (self-report) – past 12 months Self-report Chlamydia Diagnosis– past 12 months Biological Test of Chlamydia – Urine samples collected at interview

Ratio of Test/Self-report Ratio = biological test prevalence/self-report prevalence – If all infections diagnosed and reported, people having an infection over 12 months > people with infection on one day – Ratio > 1 indicates infections missing in selfreports

Measure of Risk Behavior Patterns In previous work, created a measure of risk behavior patterns Used cluster analysis Grouped participants into 16 categories based on patterns of substance use and sexual behavior Each person assigned to one pattern only

Analyses Bivariate analyses by – Sex and race-ethnicity – Risk behavior patterns Weighted percents Account for the complex sampling design

Results Prevalence of Sexual Activity Biological Sex % Had vaginal intercourse in past 12 mo. s Female 82. 9 Male 78. 2

% Prior Chlamydia Testing past 12 months Entire Sample Had Vaginal Intercourse Past 12 mo. s Female 25. 2 29. 0 Male 8. 0 9. 3

Chlamydia Prevalence – By Sex Total Female Male Ratio Tested Pos. / Self-reported % Selfreported % Tested Pos. (CI) 3. 0 4. 2 1. 4 (2. 5, 3. 4) (3. 5, 4. 9) (1. 2, 1. 7) 4. 1 4. 7 1. 1 (3. 5, 4. 9) (3. 9, 5. 7) (0. 9, 1. 4) 1. 8 3. 7 2. 1* (1. 4, 2. 3) (3. 0, 4. 7) (1. 4, 2. 7) *p<. 05 for difference between ratios of males vs. females (CI)

Chlamydia Prevalence By Race-ethnicity Among Females NH White NH Black NH Asian/PI H/NH Nat Am H White H Black % Selfreported % Tested Pos. Ratio Test/Self 2. 7 2. 5 0. 96 (2. 1, 3. 3) (1. 9, 3. 4) (0. 6, 1. 3) 10. 1 14. 0 1. 38 (8. 2, 12. 4) (11. 2, 17. 4) (1. 0, 1. 8) 2. 3 3. 4 1. 50 (0. 8, 6. 0) (1. 7, 6. 9) (0. 2, 2. 8) 3. 5 7. 9 2. 26 (1. 3, 8. 8) (4. 2, 14. 6) (-. 0, 4. 5) 4. 9 3. 8 0. 78 (2. 7, 8. 6) (2. 5, 5. 8) (0. 3, 1. 3) 4. 7 10. 22 2. 16 (1. 5, 14. 1) (2. 6, 32. 9) (-1. 6, 6. 0)

Chlamydia Prevalence By Race-ethnicity Among Males NH White NH Black NH Asian. PI H/NH Nat Am H White H Black % Selfreported % Tested Pos. Ratio Test/Self 1. 3 1. 4 1. 07 (0. 9, 1. 9) (0. 9, 2. 1) (0. 5, 1. 6) 4. 7 11. 4 2. 40* (3. 3, 6. 7) (8. 7, 14. 9) (1. 3, 3. 5) 1. 2 0. 96 (0. 3, 4. 6) (0. 4, 3. 3) (-. 7, 2. 7) 1. 6 7. 4 4. 68 (0. 6, 4. 0) (4. 2, 12. 7) (0. 0, 9. 3) 0. 8 7. 1 8. 73 (0. 3, 2. 5) (4. 2, 11. 6) (-1. 3, 18. 8) 0. 0 14. 8 -- ( -- ) (5. 9, 32. 5) -- *p<. 05 for difference between ratios of NH Black vs. NH White.

Prevalence by Sexual & Substance Use Risk Behavior Pattern % Selfreported % Tested Pos. Ratio Test/Self IV Drugs 0. 8 5. 3 6. 96 Marijuana and Other Drugs 2. 0 6. 8 3. 49 Light Dabblers 0. 7 2. 1 3. 15 Marijuana 1. 7 3. 6 2. 12 Sexually active, Few partners, Low ATOD 3. 2 6. 5 2. 05 Regret AOD with sex, Moderate AOD 2. 7 4. 4 1. 62 Light Alcohol & Sex 2. 5 4. 0 1. 61 Regret AOD with sex, High AOD 2. 7 3. 4 1. 23 Smokers 4. 1 4. 6 1. 12 Binge Drinkers 2. 5 2. 7 1. 11 Heavy Dabblers 2. 7 2. 8 1. 06 MSM 4. 2 1. 01 High Marijuana & Sex 5. 2 5. 3 1. 00 Sex for Money 14. 5 10. 2 0. 70 Multiple Partners 5. 5 3. 8 0. 70 -- -- -- Abstainers – never had sex or used AOD

Conclusions Prevalence of chlamydia testing is low – Screening guidelines for females not achieved (only 29% of sexually active young women were tested)

Infections Missing from Self-reports 4% were infected on interview day, but only 3% self-reported infection in entire past 12 months Infections particularly missing among – males compared to females – NH black males compared to white males – Some lower risk behavior categories

Possible Reasons Some respondents may not be answering accurately Many infections are undiagnosed – Due to low levels of chlamydia screening

Acknowledgments This research was supported by grant R 01 -DA 14496 -4 from the National Institute on Drug Abuse, Denise Hallfors, PI. We thank Martha W. Waller and Jon M. Hussey for consultation. This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by a grant P 01 -HD 31921 from the National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in obtaining data files from Add Health should contact Add Health, Carolina Population Center, 123 W. Franklin Street, Chapel Hill, NC 27516 -2524 (www. cpc. unc. edu/addhealth/contract. html).