Document Type

Dissertation

Degree

Doctor of Philosophy (PhD)

Major/Program

Public Health

First Advisor's Name

Mary Jo Trepka

First Advisor's Committee Title

Committee chair

Second Advisor's Name

Diana M. Sheehan

Second Advisor's Committee Title

Committee member

Third Advisor's Name

Kristopher P. Fennie

Third Advisor's Committee Title

Committee member

Fourth Advisor's Name

Boubakari Ibrahimou

Fourth Advisor's Committee Title

Committee member

Fifth Advisor's Name

Sandra Gracia Jones

Fifth Advisor's Committee Title

Committee member

Keywords

non-retention, viral suppression, HIV, AIDS, risk prediction tool, risk score, chlamydia, gonorrhea, screening

Date of Defense

3-9-2020

Abstract

This study developed risk prediction tools for non-retention in HIV care and non-viral suppression, and identified factors associated with self-reported chlamydia and gonorrhea diagnosis among people with HIV (PHIV) in the Miami-Dade County Ryan White Program (RWP). Using retrospective cohort study data, we used stepwise logistic regression to develop score-based risk prediction tools for non-retention in care and non-viral suppression. We then used bootstrapping to internally validate the risk prediction tools. We also assessed the prevalence of self-reported chlamydia and gonorrhea diagnoses and factors associated with the diagnoses cross-sectionally using multivariate logistic regression.

Among the 7439 people meeting the inclusion criteria for the retention analysis, we found that non-retention in care in the next year could be predicted using current age, race, poverty level, homelessness, problematic alcohol/drug use and viral suppression status. The risk prediction tool had low discrimination (c-statistic=0.65), and the total score ranged from 0 to 17. Among the 6492 people meeting the inclusion criteria for the viral suppression analysis, non-viral suppression in the next year could be predicted using current age, race, poverty level, AIDS status, homelessness, problematic alcohol/drug use and current viral suppression status. The risk prediction tool for non-viral suppression had good discrimination (c-statistic=0.77), and the total score ranged from 0 to 26.

Of the 7,419 adult PHIV in active Ryan White care during 2017, about half (n= 3528) reported being screened for chlamydia or gonorrhea during 2017. Of these, 2.3% reported having been diagnosed with chlamydia or gonorrhea or both in 2017. Having a chlamydia or gonorrhea diagnosis was associated with being in the age group 18–39 and having multiple sexual partners during the previous 12 months.

In conclusion, using routinely available variables, we developed risk prediction tools for non-retention in care and non-viral suppression that can assist healthcare providers in identifying high-risk individuals to target for intervention. Both risk prediction tools need external validation. The risk prediction tool for non-retention in care additionally needs to include more prognostic factors in order to increase the discrimination. In order to prevent chlamydia or gonorrhea, targeted behavioral risk reduction techniques are highly recommended among those 18–39 years of age and those who have multiple sexual partners.

Identifier

FIDC008897

ORCID

https://orcid.org/0000-0003-4669-2930

Previously Published In

Gebrezgi M.T., Fennie K.P., Sheehan D.M., … & Trepka MJ. (2020). Developing a triage tool for use in identifying people living with HIV who are at risk for non-retention in HIV care. Int J STD AIDS. In press.

Gebrezgi M.T., Fennie K.P., Sheehan D.M., … & Trepka MJ. (2020). Development and validation of a risk prediction tool to identify people living with HIV likely not to achieve viral suppression. AIDS Pt Care and STDs. In press.

Included in

Epidemiology Commons

Share

COinS
 

Rights Statement

Rights Statement

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).