Developing a dynamic HIV transmission model for 6 U.S. cities: An evidence synthesis

• DOI: 10.1371/journal.pone.0217559
• Publication Date: 2019-05-30T17:28:45Z
• Authors: Emanuel Krebs; Benjamin Enns; Linwei Wang; Xiao Zang; Dimitra Panagioto...; Carlos Del Rio; Julia Dombrowski; Daniel J. Feaster; Matthew Golden; Reuben Granich; Brandon Marshall; Shruti H. Mehta; Lisa Metsch; Bruce R. Schackman; Steffanie A. Stra...; Bohdan Nosyk
• Keywords: Acquired Immunodeficiency Syndrome; Science; Q; R; HIV; HIV Infections; United States; 3. Good health; Epidemiologic Studies; Sexual and Gender Minorities; 03 medical and health sciences; 0302 clinical medicine; Risk Factors; Antiretroviral Therapy, Highly Active; Medicine; Humans; Pre-Exposure Prophylaxis; Cities; Research Article

Dynamic HIV transmission models can provide evidence-based guidance on optimal combination implementation strategies to treat and prevent HIV/AIDS. However, these be extremely data intensive, the availability of good-quality characterizing regional microepidemics varies substantially within across countries. We aim a comprehensive transparent description an evidence synthesis process reporting framework employed populate calibrate dynamic, compartmental model for six US cities. executed mixed-method strategy parameters in categories: (i) initial HIV-negative HIV-infected populations; (ii) used calculate probability transmission; (iii) screening, diagnosis, treatment disease progression; (iv) prevention programs; (v) costs medical care; (vi) health utility weights each stage progression. identified that required city-specific stratification by gender, risk group race/ethnicity priori sought out databases primary analysis augment our synthesis. ranked quality parameter using context- domain-specific criteria verified sources assumptions with scientific advisory committee. To inform 1,667 needed model, we synthesized from 59 peer-reviewed publications 24 public surveillance reports analyses 11 sets. Of parameters, 1,517 (91%) were 150 (9%) common all Notably, 1,074 (64%), 201 (12%) 312 (19%) corresponded categories (i), (iii), respectively. Parameters as best- moderate-quality comprised 39% ranged 56%-60% cities parameters. variation values well race/ethnic groups. Better integration modelling decision making achieved systematically is models, explicitly assessing entered into model. The effective communication this help prioritize collection most informative components local care services order reduce uncertainty strengthen conclusions.