icon-folder.gif   Conference Reports for NATAP  
  17th European AIDS Conference
November 6-9
2019, Basel
Back grey_arrow_rt.gif
More Antimicrobial Prescriptions May Be Clue to Undiagnosed HIV
  17th European AIDS Conference, November 6-9, 2019, Basel
Mark Mascolini
Greater primary care prescription of antimicrobials proved a consistent signal of undiagnosed HIV infection in a nationwide case-control study of HIV-diagnosed and HIV-negative people in Denmark [1]. Researchers suggested that simple monitoring of antimicrobial prescription rates could be a key to unlocking many missed HIV diagnoses.
Researchers at Odense University Hospital and collaborators at other centers observed that half of new HIV cases still get diagnosed at a late stage of infection, with harmful consequences for both the individual and society. They wondered whether antimicrobial prescription rates could provide a marker of undiagnosed HIV infection because these drugs treat HIV indicator conditions like bacterial infections (including recurrent pneumonia and sexually transmitted infections), viral infections (including herpes simplex and herpes zoster), and fungal infections (like candidiasis).
To explore this possibility, the investigators created a case-control study in which cases were adults in the Danish HIV Cohort Study and controls were HIV-negative people in the Danish Civil Registration System matched by age and sex in a 1-to-13 ratio to HIV-positive people. People with HIV were at least 18 years old, got their HIV diagnosis between 1998 and 2016, and lived in Denmark in the 3 years before their diagnosis. The researchers used the Danish National Prescription Register to determine which people had prescriptions for oral antimicrobials in the 3 years before entering the study (for people with HIV, the 3 years before HIV diagnosis). Conditional logistic regression computed odds ratios for factors associated with an HIV diagnosis.
The analysis involved 2784 people with HIV and 36,192 controls without HIV. Men made up 80.4% of each group, and median age stood at 39 years. A lower proportion of cases than controls was born in Denmark (74.5% versus 85.8%). A much higher proportion of people with than without HIV had an antimicrobial prescription in the study period (72.4% versus 46.3%). In the preceding 3 years, the HIV group had higher proportions of prescriptions for every antimicrobial class analyzed--beta-lactams, macrolides, acyclovir, azoles, quinolones, doxycycline, nystatin, and urinary tract infection drugs. Among people with HIV, prescription rates were almost always higher for women than men.
Regression analysis indicated that people with HIV typically had doubled or greater odds of getting more prescriptions for each drug class analyzed in the preceding 3 years, whether analyzed as more than 0 versus 0, more than 3 versus 3 or fewer, more than 10 versus 10 or fewer, more than 20 versus 20 or fewer, and other cut points. The association between prescription number and HIV risk was greatest for acyclovir (all odds ratios between 5 and 10), doxycycline (all odds ratios between 5 to 10), and nystatin (all odds ratios between 10 and 20).
The researchers proposed that "prescription of antimicrobials in primary health care could be regarded as a surrogate marker of occult HIV infection and an opportunity to consider proactive HIV testing." They noted that electronic healthcare systems allow easy monitoring of antimicrobial prescription rates.
1. Martin-Iguacel R, Pedersen C, Llibre JM, et al. Prescription of antimicrobials in primary health care as a marker to identify people living with undiagnosed HIV infection. 17th European AIDS Conference, November 6-9, 2019, Basel. Abstract BPD3/3.