Structured PREreview of Dynamic HIV risk differentiation among youth: Validation of a tool for prioritization of prevention in East Zimbabwe

Does the introduction explain the objective of the research presented in the preprint?

Yes

Yes, the preprint clearly explains the objective of the paper. The preprint starts by providing a general overview of the current situation regarding HIV acquisition and the existing state of risk-differentiation and prevention tools. Moreover, the authors explain the WHO’s questions on whether identifying a person’s HIV risk and providing a specific HIV-prevention package when they are HIV or HSV-2 negative is effective. The objective of their study is clear, which is to test the ability of this behavioral questionnaire to determine HIV or HSV-2 acquisition after 12 months of being provided the appropriate prevention package. Minor Improvement suggestion: Would be helpful to include rationale for using a survey that helped predict BOTH HIV and HSV2 acquisition (why not HIV alone?)

Are the methods well-suited for this research?

Somewhat appropriate

The methods are well-suited for the study given the research concern is whether a risk differentiation tool can be used to identify high risk individuals and provide them with HIV prevention strategies. The questionnaire, as the authors described it, is well worded and their system for ranking risk makes sense given their objective. However, in the introduction, the authors state “Young people in Zimbabwe and across Southern Africa become sexually active in their mid-to late-teens”. Since the authors used household surveys in their methodology, they may not capture the most vulnerable women, for example, those who may be marginally housed or in poverty. To improve: Elaborate on why the authors decided on using household surveys, instead of other types of surveys (e.g. school-based, employment based) and what they felt would be the strengths of using this method and its impact on generalizability Elaborate on why the authors omitted some risk category measurements from the follow up survey such as partner or self STI status. Include a copy of the questionnaire given to participants within the methods sections so the readers have a better understanding of how the surveys were conducted.

Are the conclusions supported by the data?

Somewhat supported

The authors’ main conclusion is that their tool was successful in identifying young people that could benefit from HIV prevention services. The authors address limitations of their data in terms of behavior changes and under-reporting of sexual activity. They recognized that the study missed a large population that might have started off with no sexual activity but later became sexually active during the follow up period. They also comment on how risk detection tools may be useful but the main population that requires prevention are those that are younger and maybe having their first exposures to sex, thus prevention strategies are better suited being more focused on sex education. Although the authors address many of their limitations, their conclusions fail to address the most questionable aspect of their results which is that their sensitivity was 39% and specificity was 78% between both surveys, meaning the tool would have a high number of false negatives. Thus, the claim that their tool is successful at screening and predicting high risk youth in need of HIV prevention services is unfounded. They also cite tools that had sensitivities of up to 90% but still claim their tool is a well suited comparison. Major improvement suggestions: Since education strategies aren’t mentioned in their study, that part of the conclusion should be placed in the discussion section where the authors should speak about the next steps that should be taken due to their results. Address the concerningly low sensitivity and explain why/how these numbers occurred For example, the authors should frame that the questions used in this particular risk questionnaire were NOT effective at identifying those at risk given the low sensitivity, so the authors could discuss what other types of questions would help. Knowing which questions were used in the survey would be helpful for future groups to design a more effective method, so these should be included.

Are the data presentations, including visualizations, well-suited to represent the data?

Highly appropriate and clear

The study contains four tables and one visualization. The first table is clear and appropriate, as it separates the female and male participants into groups that demonstrate the study’s demographic makeup. The second table is also highly appropriate and clear because the table distinguishes the male and female risk-awareness to different contributing factors from the baseline study, also separating risk-awareness by being HIV positive or negative. Table 3 displays the participant’s responses to the follow up risk-behavior questionnaire, which I would rate was somewhat appropriate and clear. Table 4 is highly clear and appropriate because it consists of a summary of the data concerning risk of HIV infection at baseline and after the 12 month follow up study. The visualization is effective in demonstrating that there is movement between different risk groups after 12 months. Minor improvements suggestions: - The formatting of this table is very small which makes it harder to read, so I was not able to read all of the data on the table (although not a very big issue and can be fixed on the actual paper). - Unify format between tables - The purpose of the different types of data measure in Table 3 can be explained more clearly in the Risk Tool and Specificity paragraph. For example, the different PY values.

How clearly do the authors discuss, explain, and interpret their findings and potential next steps for the research?

Neither clearly nor unclearly

The Discussion and Conclusion sections are well-worded and the authors effectively address some of the successes and limits of their study. By alluding to other risk differentiation tools via a systematic review, the authors position their tool and how it fairs next to others which allows the readers to reflect on such comparisons without having to do external research or find other papers. However, as mentioned before, they fail to address why their tool has lower sensitivity measures than others and why their tool is still considered successful among other better performing tools. The authors discuss limitations of their study and address where they would take their findings in terms of public health and health education. However, the authors should discuss their potential next steps in the Discussion section instead of the Conclusions. Changing the section would allow them to further elaborate on how their study led them to these conclusions. Additionally, the authors need to add research-specific next steps as they currently only mention ways in which HIV prevention should evolve in terms of sex education and using the tool to support prevention strategies. Major improvement suggestions: - Address their low sensitivity and how this compares to other risk differentiation tools - Sensitivity is reliant on the prevalence of the disease in a population. The authors should elaborate on the prevalence of HIV within this particular community and how it impacted the sensitivities/specificities of the questionnaires - Speak about potential next steps in the Discussion section instead of the Conclusions - Expand on how their tool would benefit sex education and prevention methods - for example, what are the existing methods/tools in this community? How could a screening questionnaire be used to complement this (e.g. sex ed within schools?) - Include research-specific next steps for the tool or HIV prevention; if they could do the study again, what would they change? How should their findings affect research surrounding future risk differentiation tools? - The authors should provide more background on why a self-reported questionnaire in the context of this community could introduce bias or skew the effectiveness of identifying at risk individuals (e.g. stigma related to HIV and HSV) since they mention that “half of all incident infections occurred among participants who reported not being sexually active at baseline” - this is quite alarming that participants did not feel they were able to report their sexual activity levels honestly which can greatly impact the usefulness of a screening questionnaire in general

Is the preprint likely to advance academic knowledge?

Moderately likely

The preprint is moderately likely to advance existing knowledge. This preprint is testing the ability of questions as part of a WHO risk-management program, which means there is already some precedent behind the questions. Improvements: - While validating this tool is important, the study will not change the use or implementation of this particular questionnaire, especially since the sensitivity in this study was low. However, it does provide some evidence that the questions used in this survey, or the way it was implemented, was unhelpful, and knowledge of negative results can help those who are developing new questionnaires to design more sensitive surveys.

Would it benefit from language editing?

No

Overall, the preprint is easy to read and the authors provided adequate explanations to their study findings. There were no significant grammatical or language errors encountered.

Would you recommend this preprint to others?

Yes, but it needs to be improved

The preprint should be updated to correctly represent their findings and provide relevant implications as stated above. Once the preprint is updated, it could be recommended to others who are interested in this type of risk-management plan and HIV risk-management plans in general.

Is it ready for attention from an editor, publisher or broader audience?

Yes, after minor changes

The authors need to revise their conclusions to better align with their study and its results. The paper would only be ready for attention from a publisher or broader audience after major concerns with the study are addressed within the paper. By addressing the extremely low sensitivity, how their tool fared compared to others, discussing their study’s importance to future research studies; the authors can better situate the paper to garner more attention.

Competing interests

The authors declare that they have no competing interests.