COST AND EFFECTIVENESS IMPLICATIONS OF DIAGNOSING HIV INFECTION IN HOSPITALS: A COMPARISON OF ROUTINE SCREENING IN EMERGENCY DEPARTMENTS AND DIAGNOSTIC TESTING IN INPATIENT UNITS

Purpose:

Historically, human immunodeficiency virus (HIV) testing in hospital inpatient (IP) facilities has been for diagnostic purposes. Newer guidelines call for routine screening in medical care settings, including hospital emergency departments (ED).  ED screening may result in diagnoses at an earlier stage of illness, with more timely initiation of treatment and prevention services, but may have a lower yield (proportion of positives identified to total persons tested). We therefore compare the cost and effectiveness of routine ED screening to IP diagnostic testing using published data on median CD4 counts of HIV-infected persons diagnosed in both settings.  

Methods:

We develop an individual-based Markov state transition model, Progression and Transmission of HIV/AIDS (PATH), to simulate HIV disease progression and treatment for 10,000 persons diagnosed in IP and ED settings. We use quarterly changes in CD4 count and viral RNA level to map disease progression. We additionally include published survival estimates based on the age of the patient and CD4 count at the initiation of antiretroviral therapy. We use program characteristics from each setting such as testing costs, median CD4 count at diagnosis, and prevalence of HIV infection, as well as the costs of treating HIV infection, to estimate the incremental cost per quality-adjusted life year (QALY) saved for a person diagnosed with HIV.

Results:

Diagnosing HIV infection in the ED versus IP setting led to an additional 2.3 QALYs at a cost of $16,549 per QALY. In sensitivity analysis, with an ED prevalence of 0.1%, the incremental cost-effectiveness ratio increased to $22,644 per QALY. Factors with the most effect on quality-adjusted life expectancy include age at infection, duration of antiretroviral therapy, and viral load set-point.

Conclusions:

Routine ED screening results in increased QALYs compared to IP testing and falls  within the standards of cost-effectiveness. Hospitals that have not already done so should consider routine HIV screening of persons in the ED.

Cost effectiveness analysis of testing in different settings*

Setting (Prevalence)	CD4 count at diagnosis (cells/mL)	Discounted costs1 ($)	Discounted life expectancy1	Discounted QALYs1	Incremental Cost ($)	Incremental QALY	ICER ($/QALY)
IP  (14.6%)	24	245,498 (241,530-249,466)	16.3 (16.1-16.5)	14 (13.9-14.2)
ED (1%)	276	283,562 (280,108-287,015)	18.8 (18.7-19.0)	16.3 (16.1-16.4)	38,064	2.3	16,549

* 3% discount rate

¹ Figures in parentheses are 95% confidence intervals