Treatments for HIV infection
The purpose of this Evidence Map was to see if data indexed from abstracts could be used together with a network tool in two scenarios:
- To create a network of sufficient accuracy from available data to rule out if a network meta-analysis (NMA) is feasible.
- To create a network of sufficient accuracy from available data to assist in planning a phase 3 clinical trial
We found a recently published systematic literature review (SLR) with NMA on a topic of interest that listed the included individual studies:
Kanters S et al. "Comparative efficacy, tolerability and safety of dolutegravir and efavirenz 400mg among antiretroviral therapies for first-line HIV treatment: A systematic literature review and network meta-analysis" EClinical Medicine 2020" DOI: https://doi.org/10.1016/j.eclinm.2020.100573
The data used to create the network in this study had come from full-texts and so could be used as the gold standard when the accuracy of the abstract-predicted and full-text methods were compared.
A list of 206 abstracts was created from the list of studies included in the SLR and then indexed according to the fields below:
Disease: the type of HIV if stated e.g., HIV-1.
Complete HAART regimen with doses: drugs used in the highly active antiretroviral therapy (HAART) regimen with doses (total daily) included when available from the abstract. If more than one regimen was described, these were listed as separate tags.
Comparison of complete HAART regimen with doses: for the comparative studies, the comparison between the HAART regimens was listed as a separate tag e.g., regimen A vs B, regimen A vs C, regimen B vs C. Doses (total daily) were included when available from the abstract. For the majority of tags, the full name of the drugs have been used. However, due to a character limit some tags could not be created for some of the comparisons. In these cases, the drug abbreviations were used (please see drug table).
Drugs within the HAART regimen being compared: the purpose of this indexing field was to highlight which drugs were being compared between the multi-drug regimens. Each comparison was listed as a separate tag.
Third agent drugs only (networker tool): this indexing field was created for the Networker Tool. The SLR compared a type of anti-retroviral therapy called a ‘third agent’ which is an additional drug to the two or three drugs the person with HIV is already taking (please see drug table). Each third agent drug is listed as a separate tag. The only dose that was included was when the abstract gave details for a 400 mg dose of Efavirenz since this was relevant to the SLR.
Outcomes: the specific outcomes reported with their timeline when given in the abstract e.g., viral load at week 96.
Subpopulations: the particular characteristics of the population being studied e.g., co-infection with tuberculosis or hepatitis, pregnant or breastfeeding women, adults, children or infants.
Trial name: the name of the trial or if this was not stated, the NCT number or first author plus year of publication e.g., Albano_2019.
Study type: RCT, non-randomised comparative study or non-comparative study.
Study duration: how long the study went on for.
Study size: the number of patients included in the study.
Year: the year the study was published.
Location: the geographical location of the study or, if not clear, the authors' locations.
Risk factors: any risk factors for being infected with HIV e.g., fetal exposure.
Network scenarios: the scenario or analysis the study belonged to in the NMA e.g., primary analysis, pregnant women or co-infection with TB.
