A comparison of public health lab-based surveillance data vs. clinic visit data to assess HIV patient engagement-in-care

Background:

HIV incidence in the United States failed to decline between 2006 and 2009, due in part to some patients’ poor engagement-in-care.  The Department of Health and Human Services (DHHS) uses clinic visit data to measure HIV patient engagement.  While laboratory surveillance data may be used to gauge engagement, no previous reports have compared surveillance lab vs. clinic visit-based measures of assessing HIV patient engagement-in-care.

Objectives:

Our analysis aims to compare surveillance lab vs. clinic visit-based approaches of identifying non-engaged people living with HIV/AIDS (PLWHA).

Methods: 

We examined 2011 patient visit data from the Ruth M. Rothstein CORE Center, Cook County’s publically funded ambulatory HIV clinic caring for 6000+ PLWHA annually.  We defined engaged patients as those with a visit every 6 months over 2 years.  We used this definition to classify patients as engaged vs. non-engaged and matched patients classified via visit data against HIV viral load data reported to the Chicago Department of Health (CDPH) as part of routine surveillance.  We determined the sensitivity, specificity and receiver operator characteristics (ROC) of surveillance lab vs. clinic visit measures of engagement.          

Results:  

Defining non-engagement as having had < 2 HIV viral load values reported at least 3 months apart from the same facility per year and comparing this definition with the clinic-based measure of non-engagement achieved the best balance of sensitivity (78.0%± 2.0%) and specificity (90.8% ± 1.4%).  The ROC curve comparing performance characteristics of several surveillance lab measures of engagement had an area under the curve of 0.913, indicating that lab-based measures represent a reasonable way to evaluate for non-engagement.

Conclusions:

Our findings demonstrate that surveillance laboratory data can be used to assess level of engagement-in-care for PLWHA.   These data suggest that bi-directional data sharing between public health entities and clinical care providers could improve efforts to re-engage out-of-care PLWHA.