A comparison of two population-based household surveys in Uganda for assessment of violence against youth.

Violence is associated with health-risk behaviors, potentially contributing to gender-related HIV incidence disparities in sub-Saharan Africa. Previous research has demonstrated that violence, gender, and HIV are linked via complex mechanisms that may be direct, such as through forced sex, or indirect, such as an inability to negotiate safe sex. Accurately estimating violence prevalence and its association with HIV is critical in monitoring programmatic efforts to reduce both violence and HIV. We compared prevalence estimates of violence in youth aged 15-24 years from two Ugandan population-based cross-sectional household surveys (Uganda Violence Against Children Survey 2015 [VACS] and Uganda Population-based HIV Impact Assessment 2016-2017 [UPHIA]), stratified by gender. UPHIA violence estimates were consistently lower than VACS estimates, including lifetime physical violence, recent intimate partner physical violence, and lifetime sexual violence, likely reflecting underestimation of violence in UPHIA. Multiple factors likely contributed to these differences, including the survey objectives, interviewer training, and questionnaire structure. VACS may be better suited to estimate distal determinants of HIV acquisition for youth (including experience of violence) than UPHIA, which is crucial for monitoring progress toward HIV epidemic control.

Introduction

Violence against children (ages 14 years and under) and youth (ages 15–24 years) [1], defined by the World Health Organization as “the intentional use of physical force or power, threatened or actual, against oneself, another person, or a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation” [2], is a significant public health issue: over one billion children are affected each year globally [3]. Individuals who experience violence face wide-ranging consequences, including poor physical and mental health, decreased education and employment opportunities, higher likelihood of drinking alcohol or using drugs in general and before sex, initiating intercourse at an earlier age, having intercourse with multiple partners or strangers, having a sexually transmitted infection, and experiencing suicidal ideation [4-6].

Sub-Saharan Africa has among the highest global prevalence rates of violence against children, youth and women [3,7]. The Sub-Saharan Africa region also has the highest global prevalence and incidence of HIV; 64% of all people living with HIV live in sub-Saharan Africa [8]. Women and girls account for 63% of all new HIV infections in the region, which rises to 86% in the 15–19 year old age group [9]. In Uganda, an estimated 22% of women aged 15 to 49 had experienced some form of sexual violence in their lifetime [10], and an estimated 7.6% of females aged 15 to 64 were living with HIV [11]. Previous studies have suggested an association between violence and HIV risk, through direct mechanisms such as forced sex or indirect mechanisms such as an inability to negotiate safe sex [12-14]. Gender and violence are often intertwined, and the higher incidence of HIV among 15–24 year old females compared to their male peers in some countries likely reflect gender-related drivers of HIV infection, including violence [12,14-16].

National household surveys funded by the US President’s Emergency Plan for AIDS Relief in sub-Saharan Africa have been designed to estimate the prevalence of violence in relation to sexual risk behaviors and HIV status. Violence Against Children and Youth Surveys (VACS; https://www.cdc.gov/violenceprevention/childabuseandneglect/vacs/) are designed to estimate childhood, lifetime, and past-12-month violence prevalence in children and youth aged 13–24 years and include modules assessing sexual behaviors and HIV risk [17]. Population-based HIV Impact Assessments (PHIAs; https://phia.icap.columbia.edu/) are designed to estimate HIV-related outcomes and in most countries included a violence module during the first round of surveys; most countries use an abbreviated version to allow for inclusion of other priority issues.

The Uganda PHIA 2016–2017 (UPHIA) used the full violence module and thus is aligned more closely with the Uganda VACS (2015). UPHIA eligibility criteria were adapted to be comparable with VACS criteria, by restricting violence module eligibility to youth and by modeling the sampling approach after VACS [11]. Therefore, UPHIA served as a “proof of concept”, allowing determination of whether an HIV-focused survey with biomarker data collection resulted in comparable violence prevalence data to VACS. The purpose of this report is to compare violence prevalence estimates generated by VACS and UPHIA. Understanding comparability of violence prevalence estimates in an HIV-focused survey and a violence-focused survey has implications for ongoing surveillance of violence against youth and HIV.

Methods

Detailed descriptions of the VACS [17,18] and UPHIA [11] survey designs have been published previously. Briefly, both were population-based household surveys using a three-stage sampling approach, in which enumeration areas (EAs), limited geographic areas designated by national statistical authorities, were selected within regional strata in Uganda, and households were randomly selected within EAs. A single eligible respondent per household was selected randomly to participate in the survey (for VACS) or the violence module (for UPHIA). For both surveys, a referral mechanism was in place to put respondents in contact with a social welfare officer if further assistance was needed for a violence-related issue. National weights in both surveys were used to generate estimates that accounted for sample selection probabilities and were adjusted for nonresponse and noncoverage. Both surveys were approved by the CDC Institutional Review Board (Protocols #6538 [Uganda VACS] and #6830 [UPHIA]), and the Uganda National Council for Science and Technology.

Tables 1 and 2 summarize key differences between the two surveys. While the overarching strategy was similar, there were differences in primary objectives, survey methods and sampling (Table 1), and in questionnaire structure and content (Table 2). Both surveys were designed to generate nationally representative estimates of primary outcomes, and separated EAs such that only males or only females were eligible for either the violence module (UPHIA) or the whole questionnaire (Uganda VACS) within an EA (Table 1). Both surveys were completed via face-to-face interviews with trained interviewers but given the difference in the objectives of the two surveys, there were some differences in interviewer training. There were differences in how experience of violence questions were asked, with UPHIA generally using more parent questions and skip patterns, and Uganda VACS asking about experiences of different types of violence via different perpetrators separately (Table 2).

Table 1

Methods and sampling structure of Uganda Population-based HIV Impact Assessment (UPHIA 2016–2017) and Uganda Violence Against Children Survey (VACS 2015).

Survey Design Element	VACS	UPHIA
Data Collection Timeframe	September–November 2015	August 2016–March 2017
Primary Objective	Estimate national prevalence of violence among children and youth	Estimating national-level annual HIV incidence among adults and national and subnational prevalence of HIV and HIV viral load suppression among HIV-positive adults
Violence Questionnaire Eligibility	Males and females aged 13–24 years	Males and females aged 15–24 years†
Enumeration Area (EA) Sampling	Splits EAs into female or male (only one sex is sampled within each EA)	Splits EAs into female or male (only one sex is sampled within each EA†)
Eligibility within Households	Interviewed only one participant per household	Only one violence module completed per household; the rest of the interview completed with all eligible household members
Survey Administration Method	In-person, face-to-face	In-person, face-to-face
Consent Process	Multi-tiered consent process	Multi-tiered consent process; no additional consent for violence module
Interviewer Training	Interviewers trained in building rapport with adolescents and young adults and violence-specific data collection	Interviewers trained in building rapport among participants and service referral when necessary, less emphasis on violence data collection specifically or interviewing adolescents and young adults
Interviewer Sex	Interviewers of participant’s same sex	Interviewers of participant’s same sex, when possible
Response Plan	Detailed response plan for participants who needed and wanted help. When possible, on-call social workers were contacted by the interviewer while still in the home for immediate counseling and coordination. Otherwise the social worker would make contact with the participant within 72 hours for counseling services and additional referrals.	Response plan outlined in an SOP; interviewers were instructed to provide referrals with follow-up for any participant who met criteria and consent to referral or who requested services
Sample Size †	5,804 (males, 2,645;females, 3,159)	4,069 (males, 1,762;females, 2,307)

Abbreviations: SOP, standard operating procedure.

For UPHIA 2016–2017, refers specifically to the violence module and not to the larger PHIA survey.

Table 2

Wording and response options of comparable violence questionnaire items between Uganda Violence Against Children Survey (VACS 2015) and Uganda Population-based HIV Impact Assessment (UPHIA 2016–2017)*.

Indicator	VACS 2015	VACS Responses	UPHIA 2016–2017	PHIA Responses
Physical Violence (Lifetime)	Has (a romantic partner) ever punched, kicked, whipped, or beat you with an object?Has (a person your own age) ever punched, kicked, whipped, or beat you with an object?Has (a parent, adult caregiver, or other adult relative) ever punched, kicked, whipped, or beat you with an object?Has (an adult in the community, such as teachers, police, employers, religious leader, etc.) ever punched, kicked, whipped, or beat you with an object?	Individual item responses:1 –Yes2 –No99 –Don’t Know/DeclinedCollapsed into single item:1 –Yes to any physical violence by any perpetrator2 –No physical violence reported	Has anyone ever done any of these things to you:• Punched, kicked, whipped, or beat you with an object,• Slapped you, threw something at you that could hurt you, pushed you or shoved you• Choked, smothered, tried to drown you, or burned you intentionally• Used or threatened you with a knife, gun, or other weapon?(asked in single question)	1 –Yes2 –No• 8 –Don’t Know• 9 –Refused
	Has (a romantic partner) everstrangled, suffocated, tried to drown you, or burned you intentionally?Has (a person your own age) everstrangled, suffocated, tried to drown you, or burned you intentionally?Has (a parent, adult caregiver, or other adult relative) ever strangled, suffocated, tried to drown you, or burned you intentionally?Has (an adult in the community, such as teachers, police, employers, religious leader, etc.) ever strangled, suffocated, tried to drown you, or burned you intentionally?
	Has (a romantic partner) everused or threatened you with a knife, gun, or other weapon?Has (a person your own age) everused or threatened you with a knife, gun, or other weapon?Has (a parent, adult caregiver, or other adult relative) ever used or threatened you with a knife, gun, or other weapon?Has (an adult in the community, such as teachers, police, employers, religious leader, etc.) ever used or threatened you with a knife, gun, or other weapon?
Physical Intimate Partner Violence (Past Year) †	In the last 12 months, has a romantic partner punched, kicked, whipped, or beat you with an object?	1 –Yes2 –No99 –Don’t Know/Declined	If yes to physical violence (lifetime) question above AND reported somebody doing this to them in the past 12 months:In the last 12 months, did a partner do any of these things to you?	1 –One or more times in past 12 months2 –No physical violence ever OR No physical intimate partner violence ever OR zero times in past 12 months
	In the last 12 months, has a romantic partnerstrangled, suffocated, tried to drown you, or burned you intentionally?	1 –Yes2 –No99 –Don’t Know/Declined
	In the last 12 months, has a romantic partnerused or threatened you with a knife, gun, or other weapon?	1 –Yes2 –No99 –Don’t Know/Declined
Any Sexual Abuse (Lifetime)	Indicator of whether any of the following four survey items were reported (unwanted sexual touching, attempted forced sex, physically forced sex, and pressured into sex)	1 –Yes2 –No	Indicator of whether any of the following four survey items were reported (unwanted sexual touching, attempted forced sex, physically forced sex, and pressured into sex)	1 –Yes2 –No
Unwanted Sexual Touching (Lifetime) *†	Has anyone ever touched you in a sexual way without you wanting to but did not try and force you to have sex? Touching in a sexual way without permission includes fondling, pinching, grabbing, or touching you on or around your sexual body parts.	1 –Yes2 –No99 –Don’t Know/Declined	How many times has anyone ever touched you in a sexual way without your permission but did not try and force you to have sex? Touching in a sexual way without permission includes fondling, pinching, grabbing, or touching you on or around your sexual body parts.	1–1 or more times2 –Zero times• 8 –Don’t Know• 9 –Refused(Recoded)
Attempted Forced Sex (Lifetime) *†	Has anyone ever tried to make you have sex against your will but did not succeed?	1 –Yes2 –No99 –Don’t Know/Declined	How many times in your life has anyone tried to make you have sex against your will but did not succeed?	1–1 or more times2 –Zero times• 8 –Don’t Know• 9 –Refused
Physically Forced Sex (Lifetime) *†	Has anyone ever physically forced you to have sex and did succeed?	1 –Yes2 –No99 –Don’t Know/Declined	How many times in your life have you been physically forced to have sex?	1–1 or more times2 –Zero times• 8 –Don’t Know• 9 –Refused
Pressured into Sex *†	Has anyone ever pressured you to have sex, through harassment, threats, or tricks and did succeed?	1 –Yes2 –No99 –Don’t Know/Declined	How many times in your life has someone pressured you to have sex through harassment, threats, and tricks but without force and did succeed?Being pressured can include being worn down by someone who repeatedly asks for sex, feeling pressured by being lied to, being told promises that were untrue, having someone threaten to end a relationship or spread rumors or sexual pressure due to someone using their influence or authority.	1–1 or more times2 –Zero times• 8 –Don’t Know• 9 –Refused

*Emotional violence comparisons are not made in this report because VACS asked only about emotional violence by a parent or caregiver, whereas UPHIA asked about emotional violence by any person.

Limited to those who have ever been married or partnered in both surveys.

The original sexual violence items in UPHIA had multiple response options with 0 = zero times, 1 = 1 to 5 times, and 2 = 5 or more times. These variables were recoded by collapsing responses 1 and 2 into a single value to create an indicator variable with 1 = at least 1 time and 0 = zero times.

Weighted prevalence estimates and 95% confidence intervals (CI) were calculated stratified by gender, accounting for the complex sampling design of each survey. Comparisons between UPHIA and VACS were restricted to individuals aged 15–24 years so that comparisons included identical age groups. Two proportion z-tests were used to calculate p-values comparing prevalence estimates between VACS and UPHIA, with statistical significance determined at p<0.05. Analyses were completed using SAS 9.4.

Results

Demographic characteristics of the two groups were comparable after weighting (Table 3). Slightly over half of participants were female (UPHIA, 51.3%; VACS, 52.6%), with similar mean age distributions (UPHIA males, 19.1 years; VACS males, 18.9 years; UPHIA females, 19.2 years; VACS females, 19.4 years). Fewer male than female participants had been married or lived with someone like they were married (UPHIA males, 19.3%; VACS males, 19.5%; UPHIA females, 45.2%; VACS females, 46.9%) and had been previously told they were HIV-positive (UPHIA males, 0.1%; VACS males, 0.3%; UPHIA females, 1.6%; VACS females, 1.7%). None of the sex-specific differences between UPHIA and VACS presented in Table 3 were statistically significant (p>0.05).

Table 3

Weighted demographic characteristics and self-reported HIV status of Uganda VACS 2015 and UPHIA 2016–17 participants, by gender.

	Uganda VACS 2015		UPHIA 2016–17
	Females	Males	Females	Males
	N = 3,159 (52.6%)*	N = 2,645 (47.4%)*	N = 2,307 (51.3%)	N = 1,762 (48.7%)
Age in years [Mean (95% Confidence interval)]	19.4 years (19.2–19.6 years)	18.9 years (18.7–19.1)	19.2 years (19.1–19.3)	19.1 years (19.0–19.2)
Marital Status*
Ever Married/like married	1259 (46.9%)	411 (19.5%)	1190 (45.2%)	397 (19.3%)
Never Married	1276 (53.1%)	1653 (80.5%)	1116 (54.8%)	1365 (80.7%)
Self-report HIV Status*
`HIV-positive	39 (1.7%)	9 (0.3%)	32 (1.6%)	2 (0.1%)
Not self-report HIV-positive	2438 (98.3%)	1978 (99.7%)	2242 (98.4%)	1739 (99.9%)

*Presented as number of participants (weighted %). Marital status missing for 3 females in Uganda VACS and 1 female for UPHIA. Self-report HIV status missing for 60 females and 77 males in Uganda VACS, and 32 females and 21 males in UPHIA.

Lifetime prevalence of physical violence was significantly higher in VACS than UPHIA for both males (VACS, 77.1%; UPHIA, 32.0%; p<0.0001; Fig 1) and females (VACS, 67.0%; UPHIA, 26.0% p<0.0001; Fig 2). Similarly, past-year prevalence of physical intimate partner violence was significantly higher in VACS for females (VACS, 14.7%; UPHIA, 7.8%; p<0.0001) and males (VACS, 4.4%; UPHIA, 1.6%; p = 0.003).

Fig 1

Lifetime and past-12-month prevalence of sexual and physical violence in males aged 15–24 years in Uganda Population-based HIV Impact Assessment (UPHIA 2016–2017) and Uganda Violence Against Children Survey (VACS 2015).

Violence domains include lifetime physical violence, physical intimate partner violence (IPV) in the past 12 months, and lifetime sexual violence. Asterisks (*) indicate significantly higher prevalence in VACS than UPHIA (p<0.05). Daggers (†) indicate significantly higher prevalence in UHPIA than VACS (p<0.05).

Fig 2

Lifetime and past-12-month prevalence of sexual and physical violence in females aged 15–24 years in Uganda Population-based HIV Impact Assessment (UPHIA 2016–2017) and Uganda Violence Against Children Survey (VACS 2015).

Violence domains include lifetime physical violence, physical intimate partner violence (IPV) in the past 12 months, and lifetime sexual violence. Asterisks (*) indicate significantly higher prevalence in VACS than UPHIA (p<0.05). Daggers (†) indicate significantly higher prevalence in UHPIA than VACS (p<0.05).

Prevalence estimates of sexual violence between UPHIA and VACS were more comparable, but some differences in estimates emerged. For females, lifetime sexual violence was significantly higher in VACS (48.6%) than UPHIA (36.2%; p<0.001), as was unwanted sexual touching (VACS: 35.5%; UPHIA: 26.0%; p<0.0001) and attempted forced sex (VACS: 25.0%; UPHIA: 19.5%; p = 0.01). Meanwhile, being pressured to have sex was significantly higher in UPHIA (10.0%) than in VACS (6.6%; p = 0.006). For males, lifetime sexual violence was significantly higher in VACS (26.3%) than UPHIA (22.6%; p = 0.03), while forced sex was significantly higher in UPHIA (7.1%) than in VACS (3.1%; p<0.0001).

Discussion

Compared to VACS, UPHIA generated significantly lower prevalence estimates of lifetime physical violence, any lifetime sexual violence, and past-12-month intimate partner violence among both males and females. The differences between survey findings were more consistent for physical violence. Findings were mixed in domain-specific sexual abuse analyses; among males, lifetime reports of having been physically forced to have sex were higher in UPHIA, and among females, being pressured into having sex was higher in UPHIA while unwanted sexual touching and attempted forced sex were higher in VACS. Since underreporting is much more likely in violence data collection than overreporting [19], the overall findings suggest that UPHIA likely underestimated the burden of violence against youth, particularly physical violence, in Uganda [19,20].

Methodological differences between the two surveys likely contributed to these differences. VACS was designed specifically to estimate violence prevalence, while UPHIA was designed primarily to estimate HIV-related outcomes with an additional module to measure violence as a secondary aim. Adding a violence module to a larger survey focused on other health issues may result in underestimation of violence [21]. This can be attributed to confidentiality concerns, survey fatigue, and discomfort with the nature of questions related to violence when compared to other survey items [19]. Additionally, interviewer selection and training related to violence data collection are crucial; respondent willingness to disclose violence may be affected by the rapport established between the interviewer and respondent [21-23]. Interviewers may exhibit implicit biases, either perceived or real, that could result in underreporting. Although substantial interviewer training was conducted for data collectors in both UPHIA and VACS, the VACS interviewer selection and training focused more specifically on strategies to facilitate disclosure and assess violence in the targeted age group [24].

The content and structure of the questionnaire also likely contributed to differences in prevalence across the surveys. For physical violence, the VACS questionnaire repeats stem questions for each of the four categories of perpetrators (intimate partners, peers, parents/caregivers/other adult relatives, and adults in the community) to anchor the responses to different types of relationships and facilitate recall. UPHIA asked one question about physical violence from any perpetrator and followed affirmative answers with a question about the perpetrator. Given the normalization of many forms of violence, without orienting respondents to the possible perpetrator, they may not consider certain acts to be violent (e.g., violence perpetrated by a peer or a caregiver) [25]. The respondent may think of only the most salient relationships in their life rather than thinking about all of the types of relationships addressed in the VACS questionnaire. Conversely, UPHIA provided additional clarification to the respondent regarding examples of being pressured into sex, which was the only violence domain more frequently reported in UPHIA than VACS among females. Previous research has demonstrated that using multiple behaviorally specific questions about violence generates higher prevalence estimates than broader, aggregate questions [25]. Our findings suggest that the structure of violence questionnaires, specificity of questions, and design of parent questions and skip patterns may be critical in avoiding under-reporting of violence. Table 2 demonstrates that the structure of the UPHIA questionnaire for physical violence was much different than the VACS, while they were more similar for sexual violence; underestimates for physical violence in UPHIA were also more pronounced than for sexual violence.

PHIAs provide valid and reliable information on important indicators of HIV epidemic control, including incidence, prevalence, and progress toward Joint United Nations Programme on HIV/AIDS 95-95-95 targets (95% of HIV-positive individuals are aware of their HIV status; of these, 95% are receiving antiretroviral therapy; and of these, 95% have achieved viral load suppression) in countries with generalized HIV epidemics. However, our findings suggest that enhancing the VACS to include key laboratory outcomes relating to HIV (e.g. HIV testing, incidence testing, and viral load testing) rather than incorporating additional violence data collection into HIV-focused PHIAs, may be a better approach to understand the prevalence of violence in youth and the relationship between violence and HIV. This approach has been implemented to varying degrees in several recent VACS, which also included rapid HIV testing (in Botswana, Côte d’Ivoire, Lesotho, Kenya, Namibia, Zimbabwe, and Mozambique); the violence module of PHIA has been removed for all countries as of 2019. UPHIA’s violence module represented an attempt to incorporate lessons learned from VACS surveys to estimate violence prevalence in an HIV-focused survey; however, many violence-related outcomes were underestimated in UPHIA. Generating reliable violence estimates in an HIV-focused survey may be possible, given the similar findings in some of the similarly worded sexual violence questions between VACS and UPHIA. However, this would likely require further changes to questionnaire structure (further increasing interview length), as well as additional specialized training for data collectors in facilitating violence disclosure. VACS and PHIA have different primary objectives, and VACS may be better suited to estimate distal determinants of HIV acquisition for youth (including experience of violence), since they can focus more of the interview time as well as the recruiting and training of the interviewers on these topics.

12 Oct 2021

PONE-D-21-25230A comparison of two population-based household surveys in Uganda for assessment of violence against young adultsPLOS ONE

Dear Dr. Currie,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Nov 26 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Lindsay Stark

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dear Editor,

Thank you for the opportunity to review the manuscript titled “A comparison of two population-based household surveys in Uganda for assessment of violence against young adults”. This paper does an important need – advancing our understanding of the comparability of household datasets that serve as the primary source of data that is the basis of many programmatic and policy decisions, particularly on issues such as violence against children that are vastly under researched. The manuscript is well written; the research objectives are clearly articulated; and the utilizes sound methodology.

I have one comment about the manuscript. The authors’ attempt to provide guidance on how best to address the discrepancies identified is laudable but I am confused by the suggestion that “our findings suggest that enhancing the HIV module of 178 VACS with biomarker data (e.g., incorporating HIV testing), rather than incorporating additional 179 violence data collection into HIV-focused PHIAs, may be a better approach to understand the 180 prevalence of violence in youth and the relationship between violence and HIV.” How would including biomarker data on HIV testing advance this goal? This would be understandable if the authors were suggesting biomarkers of chronic stress? Also, I understand that household surveys have several constraints – the budget, survey lengths, country priorities etc., should we not consider perhaps adapting strategies that have been effective for the VACS? Given that VACS does not include HIV related data, I would think that strengthening this component within a HIV focused survey would most likely advance our goals of understanding the consequences of violence in this population.

Reviewer #2: The is manuscript covers an important topic and the findings can provide a useful contribution to the measurement of violence against children and young adults. Between two studies, VACS and UPHIA, the authors find significant differences in violence prevalence for both females and males aged 15-24 in Uganda. The discussion provides practical implications for measuring violence against children and young adults within HIV studies.

Abstract

- The authors mention that the association of violence to health-risk behaviors may contribute to gender-related HIV disparities. While this is an important point, it is also important to consider that sexual violence may directly contribute to HIV transmission. The abstract would be more compelling if it started by acknowledging the complex ways in which violence, gender, and HIV are linked.

Introduction

- Lines 40-42: The source of the definition for violence against children and youth is unclear. The citation provided is relevant for the annual estimation that over one billion children are affected by violence globally; however, the reference does not provide the quoted definition.

- Lines 44-46: It would strengthen the authors' introduction to amend this sentence to describe the wide-ranging consequences faced by individuals who experience violence during childhood per the Berenson, Wiemann, and McCombs article and other VAC-specific articles.

Lines 46-47: The sentence that begins with "In addition to having the highest HIV prevalence..." starts quite abruptly. The authors may want to consider including a separate paragraph to describe the violence and HIV risks in sub-Saharan Africa or Uganda specifically.

- It is not always clear what differences exist when the authors mention children, youth, young adults, or adolescents. It would strengthen the readability if the authors used consistent terminology and provided corresponding definitions when utilizing various age-specific terminology.

Methodology

- The authors provide a clear description of the similarities between the VACS and UPHIA implementation and the authors guide readers to Tables 1 and 2 to better understand differences. It is important that the authors add a paragraph describing key differences as well.

- When discussing the similarities between the VACS and UPHIA sampling strategies, it would be helpful to mention the split-stage design utilized by both.

- Table 2: There seem to be a few formatting inconsistencies or errors that the authors should review. For example, the narrative regarding repeated questioning per perpetrator is described differently in the first cell than in the second and third cell. As another example, the first example of "Physical Intimate Partner Violence (Past Year)" is an example of lifetime experience: "Has a romantic partner ever punched, kicked, whipped, or beat you with an object."

Results

- Lines 106-112: The content in the first paragraph would benefit from a complementary Table.

- Lines 106-113: It is unclear where there are statistically significant differences in findings provided in this paragraph, which is important since the authors describe differences and similarities between the two studies.

- Lines 108-109: It would help if the authors provided CIs rather than the IQRs

Discussion

- Lines 141-143: The authors describe "significantly lower prevalence estimates of both lifetime physical and sexual violence for both males and females"; however, the findings related to sexual violence are mixed. For some questions, the UPHIA resulted in higher prevalence.

- Lines 142-143: The authors describe that past 12-month IPV was lower in the UPHIA than the VACS for females. It is unclear why the authors did not mention that past 12-month physical IPV was lower in the UPHIA than the VACS for males, as described in lines 116-117.

- Lines 143: Suggest that the authors use the term "more consistent" rather than "largest"

- Lines 153-158: Sexual violence is often more sensitive to disclose; yet, there was not an overall lower reporting in the UPHIA than the VACS. The authors should clarify how interviewer selection and training may have influenced reporting for physical violence but not sexual violence.

- Lines 159-172: The discussion would benefit from the authors further describing why they think the VACS consistently measured higher prevalence of physical violence but not sexual violence. Lines 168-172 indicate that the specificity of questions, including forms of violence and specific perpetrators, may be critical to increased reporting.

- When similar questions were asked in the VACS and UPHIA, such as sexual violence questions, results were more consistent between the two studies. Thus, the authors opinion that the PHIAs should not incorporate additional violence questions seems inconsistent with the evidence. However, the suggestion for the VACS to include biomarker data is compelling.

- Lines 184-185: The evidence provided in this manuscript suggests that the inclusion of perpetrators within the physical violence questions of the VACS enables it to better measure violence as a distal determinant of HIV acquisition, but the authors should clarify if/what other factors contribute to this conclusion.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

18 Nov 2021

Reviewer #1: Dear Editor,

Thank you for the opportunity to review the manuscript titled “A comparison of two population-based household surveys in Uganda for assessment of violence against young adults”. This paper does an important need – advancing our understanding of the comparability of household datasets that serve as the primary source of data that is the basis of many programmatic and policy decisions, particularly on issues such as violence against children that are vastly under researched. The manuscript is well written; the research objectives are clearly articulated; and the utilizes sound methodology.

We would like to thank the reviewer for their thoughtful review of our manuscript.

I have one comment about the manuscript. The authors’ attempt to provide guidance on how best to address the discrepancies identified is laudable but I am confused by the suggestion that “our findings suggest that enhancing the HIV module of 178 VACS with biomarker data (e.g., incorporating HIV testing), rather than incorporating additional 179 violence data collection into HIV-focused PHIAs, may be a better approach to understand the 180 prevalence of violence in youth and the relationship between violence and HIV.” How would including biomarker data on HIV testing advance this goal? This would be understandable if the authors were suggesting biomarkers of chronic stress?

We understand the confusion with the biomarker language – this is terminology that we use broadly in PHIAs as an umbrella term for laboratory related HIV outcomes (e.g., antibody testing, incidence testing, viral load, ARV metabolite detection, etc.). We have revised the language to more clearly reflect this in the manuscript.

Also, I understand that household surveys have several constraints – the budget, survey lengths, country priorities etc., should we not consider perhaps adapting strategies that have been effective for the VACS? Given that VACS does not include HIV related data, I would think that strengthening this component within a HIV focused survey would most likely advance our goals of understanding the consequences of violence in this population.

Thank you for raising this important point. We agree with the reviewer that household surveys, particularly those focused on HIV, have the constraints acknowledged (budget, survey length and corresponding participant response rates and willingness to participate, country priorities, etc.). VACS is also a household survey, but has the benefit of focusing specifically on violence, which appears to elicit more reliable violence prevalence data. UPHIA was an attempt to do just what the reviewer is suggesting, strengthening an HIV focused survey to collect violence data using lessons learned from VACS. Unfortunately, UPHIA resulted in underestimates, as described in the study. Therefore, we think the approach of including HIV related laboratory data in VACS (which is already underway in some surveys) may be more beneficial. We’ve attempted to make this point more explicitly in lines 218-227 of the revised manuscript. We’ve also acknowledged in the revised manuscript that it may yet be possible to collect reliable violence estimates in an HIV-focused household survey, but that further changes to the approach would be required. These further changes may jeopardize the ability of PHIAs to achieve their primary objectives, and so the violence module was removed in PHIAs in favor of continuing to complete VACS surveys for the purposes of violence estimation (in adolescent girls and boys and young women and men).

Reviewer #2: The is manuscript covers an important topic and the findings can provide a useful contribution to the measurement of violence against children and young adults. Between two studies, VACS and UPHIA, the authors find significant differences in violence prevalence for both females and males aged 15-24 in Uganda. The discussion provides practical implications for measuring violence against children and young adults within HIV studies.

Thank you very much for your thoughtful review.

Abstract

- The authors mention that the association of violence to health-risk behaviors may contribute to gender-related HIV disparities. While this is an important point, it is also important to consider that sexual violence may directly contribute to HIV transmission. The abstract would be more compelling if it started by acknowledging the complex ways in which violence, gender, and HIV are linked.

Thank you for this suggestion, we have revised the abstract accordingly.

Introduction

- Lines 40-42: The source of the definition for violence against children and youth is unclear. The citation provided is relevant for the annual estimation that over one billion children are affected by violence globally; however, the reference does not provide the quoted definition.

Thank you, we have provided an additional reference for the direct quote defining violence and have clarified that this is a World Health Organization definition. We’ve also added a citation with the definition of youth.

- Lines 44-46: It would strengthen the authors' introduction to amend this sentence to describe the wide-ranging consequences faced by individuals who experience violence during childhood per the Berenson, Wiemann, and McCombs article and other VAC-specific articles.

The sentence as currently written lays out many of the wide-ranging consequences described by Berenson et al. and some of the other cited work, e.g. poor physical and mental health, decreased education and employment opportunities, and increased health risk behaviors. We have further elaborated on the increased health risk behaviors (see track change manuscript lines 48-50) to highlight the negative effects of experiencing violence in youth.

Lines 46-47: The sentence that begins with "In addition to having the highest HIV prevalence..." starts quite abruptly. The authors may want to consider including a separate paragraph to describe the violence and HIV risks in sub-Saharan Africa or Uganda specifically.

Thank you, we have further elaborated on the risks of both HIV and violence in sub-Saharan Africa, and also a sentence on the prevalence on HIV and lifetime sexual violence in Uganda among females. Relevant citations have also been added.

- It is not always clear what differences exist when the authors mention children, youth, young adults, or adolescents. It would strengthen the readability if the authors used consistent terminology and provided corresponding definitions when utilizing various age-specific terminology.

Thank you, we have made an effort to use children and youth more consistently throughout the manuscript, with youth referring specifically to the 15-24 year old age group (see new UN citation for definition of youth provided - https://www.un.org/esa/socdev/documents/youth/fact-sheets/youth-definition.pdf).

Methodology

- The authors provide a clear description of the similarities between the VACS and UPHIA implementation and the authors guide readers to Tables 1 and 2 to better understand differences. It is important that the authors add a paragraph describing key differences as well.

Thank you, we have further elaborated on differences between the two surveys within a new paragraph in the methods section (lines 96-103 in track change version).

- When discussing the similarities between the VACS and UPHIA sampling strategies, it would be helpful to mention the split-stage design utilized by both.

Thank you, the revised paragraph explaining the differences between the surveys mentions this design used in both surveys.

- Table 2: There seem to be a few formatting inconsistencies or errors that the authors should review. For example, the narrative regarding repeated questioning per perpetrator is described differently in the first cell than in the second and third cell.

Thank you, we have modified formatting for consistency in these cells.

As another example, the first example of "Physical Intimate Partner Violence (Past Year)" is an example of lifetime experience: "Has a romantic partner ever punched, kicked, whipped, or beat you with an object."

Thanks very much for pointing this one out, this was indeed an error that we have corrected.

Results

- Lines 106-112: The content in the first paragraph would benefit from a complementary Table.

We agree that a Table helps presents the data in a more digestible way, Table 3 has been added in the revised manuscript (see line 135 in track changed version).

- Lines 106-113: It is unclear where there are statistically significant differences in findings provided in this paragraph, which is important since the authors describe differences and similarities between the two studies.

None of the differences presented in newly created Table 3 between UPHIA and VACS were statistically significant, either when comparing males or females. We have stated this in lines 134-135.

- Lines 108-109: It would help if the authors provided CIs rather than the IQRs

In Table 3, which was created based on the reviewers suggestion, we have added the 95% CI for ages.

Discussion

- Lines 141-143: The authors describe "significantly lower prevalence estimates of both lifetime physical and sexual violence for both males and females"; however, the findings related to sexual violence are mixed. For some questions, the UPHIA resulted in higher prevalence.

Thanks for pointing out some of the inconsistencies in the reporting here; we’ve modified the opening paragraph of the discussion to address both this comment and the following comment.

- Lines 142-143: The authors describe that past 12-month IPV was lower in the UPHIA than the VACS for females. It is unclear why the authors did not mention that past 12-month physical IPV was lower in the UPHIA than the VACS for males, as described in lines 116-117.

See comment above.

- Lines 143: Suggest that the authors use the term "more consistent" rather than "largest"

Thank you, modified as suggested.

- Lines 153-158: Sexual violence is often more sensitive to disclose; yet, there was not an overall lower reporting in the UPHIA than the VACS. The authors should clarify how interviewer selection and training may have influenced reporting for physical violence but not sexual violence.

Our results found that lower prevalence of sexual violence than physical violence in UPHIA than VACS, particularly in females; the difference was just not as great as the physical violence findings and were less consistent when looking at specific violence domains.

- Lines 159-172: The discussion would benefit from the authors further describing why they think the VACS consistently measured higher prevalence of physical violence but not sexual violence. Lines 168-172 indicate that the specificity of questions, including forms of violence and specific perpetrators, may be critical to increased reporting.

We agree that this is an interesting finding, and think that the specificity, skip patterns, and wording of the questions likely contributed to the larger difference between surveys in physical violence than sexual violence. We’ve elaborated on this point in line with the reviewer’s suggestion (lines 202-206 in track changed manuscript).

- When similar questions were asked in the VACS and UPHIA, such as sexual violence questions, results were more consistent between the two studies. Thus, the authors opinion that the PHIAs should not incorporate additional violence questions seems inconsistent with the evidence. However, the suggestion for the VACS to include biomarker data is compelling.

Thank you for making this important point; we have modified this paragraph in response to this comment as well as a comment made by another viewer. We agree with the reviewer that the more consistent estimates for questions asked in similar ways suggests that it may be possible to generate valid and reliable estimates of violence within an HIV-focused survey. However, this would likely involve expanding the questionnaire to be even longer, and given the differences in the primary objectives of the survey, the approach of having specialized surveys to attempt to estimate violence prevalence in children and youth along with HIV laboratory data makes the most sense. That way the survey can do a deeper dive into violence by asking each of the questions as intended.

- Lines 184-185: The evidence provided in this manuscript suggests that the inclusion of perpetrators within the physical violence questions of the VACS enables it to better measure violence as a distal determinant of HIV acquisition, but the authors should clarify if/what other factors contribute to this conclusion.

Thanks for mentioning this – in addressing the previous comment and the other reviewer’s comments, we have reframed our concluding suggestions – see lines 218-227 in the revised track changed version of the manuscript.

Submitted filename: Response to Reviewer Comments_PLoS One.docx

Click here for additional data file.

22 Nov 2021

A comparison of two population-based household surveys in Uganda for assessment of violence against youth

PONE-D-21-25230R1

Dear Dr. Currie,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Lindsay Stark

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

2 Dec 2021

PONE-D-21-25230R1

A comparison of two population-based household surveys in Uganda for assessment of violence against youth

Dear Dr. Currie:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Lindsay Stark

Academic Editor

PLOS ONE