Evidence for optimal HIV testing intervals in HIV ‐ negative individuals from various risk groups : a systematic review protocol

Background: Evidence-based recommendations for HIV testing are essential for health care providers. However, it is unclear whether there is sufficient evidence to support recommendations for HIV testing frequencies in a variety of HIV risk groups. Objective: The aim of this document is to outline the methodological protocol of a systematic review that would gather evidence for the optimal frequency of HIV testing among individuals in various HIV risk groups with respect to personal and public health outcomes and cost-effectiveness. Methods: This protocol adheres to the PRISMA-P reporting items, and the review is registered with PROSPERO. The target population includes individuals who may have undiagnosed HIV infection. Different frequencies of HIV testing will be compared and outcomes to do with personal and public health, patient values/preferences and costs will be examined. The search strategy will encompass searches in MEDLINE/Pubmed, Scopus, Embase, Cochrane, PsychINFO, and EconLit, as well as grey literature sources. Articles will be screened by title/ abstract, and subsequently by full-text, in duplicate. Extraction of pertinent data from the screened references will be carried out by one reviewer and verified by a second. Multiple critical appraisal tools will be used to assess individual study quality, and the GRADE approach will be used to appraise the overall quality of the evidence. Data will be synthesized narratively, and the results will be published in a peer-reviewed journal. Discussion: This systematic review, designed with extensive input from content experts, will help to identify key evidence to inform recommendations for HIV testing frequency. Affiliation 1 Centre for Communicable Diseases and Infection Control, Public Health Agency of Canada, Ottawa, ON * Correspondence: karen. timmerman@phac-aspc.gc.ca


Background
HIV testing is a key element of the HIV cascade of care, representing the first "90" of the Joint United Nations Programme on HIV/AIDS (UNAIDS) 90-90-90 global strategy for addressing HIV/AIDS (1).HIV diagnosis is essential for linkage to care and initiation of antiretroviral therapy (ART), which may subsequently lead to decreased viral load, reduced infectivity and improved personal health outcomes (2,3).Low rates of screening and testing are therefore a potential limiting factor for the success of HIV prevention strategies.Up to 50% of new HIV infections may be attributed to those who are unaware of their infection (4)(5)(6)(7).In Canada, it is estimated that individuals who are unaware of their infection represent 21% of all those who are HIV-infected (8).
Certain populations who are at higher risk for HIV infection, such as men who have sex with men (MSM), injection drug users (IDU) or Indigenous peoples, may benefit from more frequent HIV testing.Accordingly, an important question in developing recommendations and strategies for HIV testing is how often should individuals from different risk groups be tested for HIV.Answering this question requires balancing the potential benefits of enhanced screening with the increased costs of the screening, as well as considerations of patient values and preferences.
Clear and specific guidance for how often to test individuals from different risk groups may help health care providers improve their testing behaviours and normalize this practice.A recent systematic review of guidelines found that a number of the guidelines recommend annual screening for groups including MSM and IDU but that others cite a paucity of data frequency of testing for HIV in certain groups, leading to some inconsistencies in guidance (9).At the federal level, the Public Health Agency of Canada's HIV Screening and Testing Guide states that individuals involved in high-risk practices should be screened for HIV at least annually but that there is insufficient evidence to provide recommendations for individual scenarios (10).
Many of the guidelines reviewed in the systematic review of guidelines mentioned above are several years old.In addition, guideline developers do not always describe the basis for their recommendations (e.g.systematic review versus expert opinion).A thorough, up-to-date review of scientific evidence related to HIV testing frequency is warranted and will be useful for developing timely quality guidance in Canada and abroad.The purpose of this article is to describe the protocol for a systematic review aimed at answering a number of questions related to how often to test for HIV.

Objective
The objective of the systematic review is to examine the scientific evidence that supports different frequencies of HIV testing for individuals in various risk groups who may have undiagnosed HIV.
The over-arching research question is: What is the optimal frequency of testing for HIV in individuals from various HIV risk groups with respect to personal and public health outcomes and cost-effectiveness?The sub-questions relevant to this include:  (12).In addition, a health economist was consulted.

Search strategy
A comprehensive search strategy was designed in consultation with a research librarian.The search strategy was also peer-reviewed by an external research librarian prior to execution of the search.The full search strategy can be found in Appendix 2.
The following databases will be searched: The following sources of grey literature will be searched: • Open Grey • ClinicalTrials.gov• All relevant sources from the CADTH Grey Matters checklist (13) The search strategies that will be used for Open Grey and ClinicalTrials.govcan be found in Appendix 3.
The CADTH Grey Matters tool is a checklist used to guide online searches for grey literature.It includes national and international health technology assessment websites, drug and device regulatory agencies, clinical trial registries, health economics resources, Canadian health prevalence or incidence databases and drug formulary web sites (13).A total of 40 relevant websites from the checklist were identified by the research team (drug formulary, drug advisory and warning, and surveillance databases were not considered relevant to the research question, for example).Many of the websites in the checklist do not include an advanced search option, so the large majority will be searched using the term "HIV" to provide the widest range of potentially relevant results.Websites with advanced search functions will be searched with combinations of "HIV" and "testing" or "screening" or "test" or "screen." Grey literature searches will be carried out by two members of the research team independently, and all articles deemed potentially relevant will be added to the results of the database searches for further screening.

Data management
All references will be uploaded into the DistillerSR, a secure, internet-based systematic review management software (Evidence Partners).This software platform will be used for screening eligible articles, data extraction and quality assessment.

Eligibility criteria Language
Articles published in English or French will be considered for the review.

Study design
Eligible study designs will vary depending on the specific research sub-question.For all search questions, other systematic reviews will not be explicitly excluded.Rather, the reference lists will be scanned for relevant articles that may have otherwise been missed.
Table 1 outlines the study designs that will be considered for each of the research sub-questions.

Population
The target population includes individuals who may have undiagnosed HIV infection.

Intervention
The intervention of interest is HIV screening/testing at varying intervals.

Comparison
The effects of the intervention will be compared to any the following: • Other interventions • "Normal" or "standard" state of care (as defined in a given study) • Before/after comparisons Setting Studies will be considered if they are held in any setting where HIV testing could be conducted.

Exclusion criteria
Articles will also be excluded if they are: • Published prior to 2000 • Commentaries, editorials, letters to the editor, conference abstracts, poster presentations • Guidelines/ policy papers/ policy documents

Outcomes
Several potential outcomes that could be relevant for each of the research sub-questions have been identified.However, the possibility exists that not all of the outcomes will be represented in the literature, as some may not have been examined.The primary and secondary outcomes of interest for each research question are outlined in Table 2.

Screening/selection of articles
All references identified in the search will be screened based on title and abstract following removal of duplicates.The eligibility criteria above will be used to determine inclusion and exclusion Quantitative and qualitative studies will be considered What is the most cost-effective frequency or interval of re-testing for HIV in people who have an unknown or a previously confirmed negative serostatus?
Costing studies: • Cost-effectiveness studies, cost-benefit, cost-consequence, cost-utility, cost minimization • Modelling studies Abbreviation: HIV, human immunodeficiency virus of articles at the title/abstract-screening stage.Screening of titles and abstracts will be performed in duplicate by two reviewers.Disagreements will be reconciled through discussion with a third reviewer.
Full-text screening will be performed on titles with abstracts that meet the above criteria or for cases in which it is unclear whether this is the case.At the full-text screening stage, the eligibility criteria, as well as the outcomes listed above, will guide exclusion of irrelevant articles.Full-text screening will be performed in duplicate by two reviewers.Disagreements will be reconciled through discussion with a third reviewer.
Reasons for exclusion will be recorded at each step of the screening process.The results of the screening will be presented in a flowchart consistent with PRISMA recommendations (14).

Data extraction
Extraction of pertinent data from the screened references will be carried out by one reviewer and verified by another in order to reduce bias or errors in extraction.Any disagreements will be resolved through discussion with one or two other reviewers.Study authors may be contacted if there are any major uncertainties.All pertinent data (i.e.year of publication, period of data collection, study population, sample size, location, study setting, study design, intervention, comparison, results/ outcomes, quality assessment score, etc.) will be extracted into evidence tables.Duplicate, overlapping or companion studies, if identified during the screening process, will be dealt with by extracting the data into a single collection form or by extracting the data separately and combining these into a single input afterwards, as per the Cochrane Handbook for Systematic Reviews of Interventions (15).

Quality assessment
Quality assessment will be carried out for individual studies included in the review, as well as for the overall body of evidence.Different methods of assessment will be used as appropriate for the study type.
The following assessment tools will be used for individual studies: • Cochrane risk of bias for RCTs (15) • Effective Public Health Practice Project (EPHPP) Quality Assessment Tool for other quantitative studies (16-18) • Critical Appraisal Skills Programme (CASP) Qualitative Checklist for qualitative studies (19) For the overall body of evidence, the quality of evidence will be assessed using the GRADE approach (15).

Data synthesis
Data will be synthesized narratively.If there is sufficient homogeneity among the evidence, a meta-analysis may be considered, although this is unlikely due to the variety of evidence sources being sought.

Subgroup analysis
The initial search will not target any specific subgroups.If evidence emerges for specific subgroups (e.g.MSM, IDU, Indigenous peoples, etc.) then the results will be disaggregated and reported narratively in separate sections for each subgroup.

Assessment of meta-biases
Statistical assessment of meta-biases such as publication bias across studies (e.g.Egger's test) will likely not be possible given the wide variety of evidence being sought and the likelihood of inclusion of a large proportion of studies with observational designs (20).The potential for publication bias will however be reduced by employing a rigorous search of grey literature, and the potential for substantial publication bias in observational studies will be taken into account when assessing the overall body of evidence using the GRADE approach (20).Selective outcome reporting will be assessed in RCTs by comparing the reported results of studies with the outcomes reported in the methods section of the protocol of the study.This will factor into quality assessment with the Cochrane risk of bias tool (15).

Amendments
The research team does not foresee any amendments to the protocol prior to carrying out the systematic review.However, if this is necessary, all amendments will be recorded as they occur and reflected in the PROSPERO record for this review.Amendments will also be documented in the final publication.

Dissemination
A manuscript of the results of the systematic review will be prepared and submitted for publication in a peer-reviewed journal.The results will be presented according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (14).

Discussion
To the best of our knowledge, this will be the first published systematic review examining evidence supporting recommendations for HIV testing frequency.The evidence identified in this review may be useful to update or create new guidance around HIV testing in Canada; groups outside Canada may also find it useful.The development of specific, evidence-based recommendations will help health care providers streamline and improve their HIV testing practices.Such recommendations can also be used by the public to manage their own sexual health.
Enhanced HIV screening and testing will help decrease the substantial portion of individuals with HIV who are unaware of their infection.This group contributes to a substantial proportion of new HIV infections, and evidence suggests that once these individuals are aware of their infection, they will be more likely to take steps to minimize the likelihood of transmission (4,21).HIV diagnosis is also the first step toward obtaining treatment, and thus, enhanced screening and testing yields benefits for personal and public health.
The strengths of this systematic review protocol include extensive input from content experts in the development of HIV and sexually transmitted infection guidelines and health economics.Another strength is the external peer review of the research librarian-designed search strategy.
One limitation of the review could be the potential inclusion of predominantly observational studies, given the difficulty of carrying out experimental studies on the topic.This may raise concerns regarding the quality of evidence; however, the results of the quality assessments will be published, and thus assessment of bias in the evidence will be transparent.
If this systematic review fails to find evidence to answer one or more of the research questions, this will be documented in the results.A "negative" finding will still be of use to guideline developers as it provides validity to those guidelines that cite a lack of evidence for HIV-testing interval recommendations (9) and suggests that such recommendations may need to be made on a jurisdiction-by-jurisdiction basis, based on expert opinion.It would also help highlight gaps in evidence, which could be useful for guiding future research.

Table 2 :
Potential outcomes for each of the research sub-questions

Table 1 :
Study designs eligible for inclusion in the review, by research sub-question All searches run and downloaded onSeptember 16, 2016.time OR timing OR moment OR temps OR often OR souvent* ) W/3 ( screen* OR rescreen* OR test OR tests OR tested OR testing OR retest* OR "diagnostic du VIH" OR depistage ) ) ) OR ( ( TITLE-ABS ( screen* OR rescreen* OR test OR tests OR tested OR testing OR retest* OR "diagnostic du VIH" OR depistage ) ) mult OR arts OR busi OR deci OR econ OR psyc OR soci ) hiv+ or hiv-1 or hiv-2 or hiv1 or hiv2 or hivaids or (human immun* adj2 virus) or virus de l'immunodeficience humaine or vih or vih+ or vih1 or vih2 or vih-1 or vih-2).ti,kw.or rescreen* or test or tests or testing or tested or retest* or depistage or "diagnostic du VIH").ti,kw.608987 15 (screen* or rescreen* or test or tests or testing or tested or retest* or depistage or "diagnostic du VIH").ab.
Abbreviation: Line no(s), Line numbers Abbreviation: ID, Identification number Abbreviation: No., Number