Women engaged in commercial sex living with Human Immunodeficiency Virus and viral non-suppression in Uganda: analysis of multiple respondent-driven sampling surveys, 2021—2023

Abstract

Background

Human Immunodeficiency Virus (HIV) Viral load non-suppression (VNS) among women engaged in commercial sex (WECS) is unknown even with persistent new HIV infections in the general population. The objective of this study was to find out the factors correlated with VNS among WECS living with HIV (WECS-LHIV) in Uganda.

Methods

Cross-sectional data were collected through respondent driven sampling (RDS) in Kampala Capital City and 11 upcountry towns. All consented WECS-LHIV reporting commercial sex 3 months prior to the surveys were tested for HIV and viral load (VL), active syphilis and high-risk human papilloma virus (hr.HPV) which was based on genotypes 16 and 18 that are highly associated with all cervical cancer cases. VNS was defined as any VL measure < 1,000 copies/ml. Analysis of correlates of VNS were restricted to WECS-LHIV with VL results irrespective of their antiretroviral therapy (ART) status. We examined variables including – age, marital status, education, active syphilis, hr.HPV and reported ART use to identify correlates of VNS. Sampling weights to reflect the complex sampling design were applied.

Results

Of 2,674 WECS-LHIV (median age 32 years, interquartile range (IQR) 27–37), 667 (24.9%) had VNS. The findings show that 19.2% and 29.7% of WECS-LHIV with VNS had active syphilis and hr.HPV co-infections respectively. WECS-LHIV had reduced odds of VNS if they were aged 30–49 years ([adjusted odds ratio (aOR): 0.46, 95% confidence interval (CI): 0.30–0.69, p < 0.001)] vs. 18–24 years and if they self-reported ART use [aOR: 0.38, 95% CI: 0.28–0.52, p = < 0.001 vs. non-self-reporting ART].

Conclusions

Young women engaged in commercial sex(YWECS) and none ART disclosure WECS may increase risk of new HIV infections in general population. Increase on frequency of ART adherence and viral load monitoring of YWECS including continuous surveillance of the second and third 95 HIV continuum of care and treatment for hidden populations may aid in reduction of new HIV infections burden in Uganda and other low- and middle-income countries.

Supplementary Information

The online version contains supplementary material available at 10.1186/s44263-026-00308-4.

Background

Global Human Immunodeficiency Virus (HIV) programme goal is to have 95% of persons living with HIV on treatment with suppressed viral load [1]. However, in Africa and elsewhere, low level viremia (LLV) is associated with viral non-suppression (VNS). Similarly, in a multi-country survey among women engaged in commercial sex living with HIV (WECS-LHIV), poverty and never being married are associated with VNS in Africa [2–4]. In another study in a similar population in South Sudan, VNS among women engaged in commercial sex (WECS) was at 38% in Juba and 14% in Nimule with anti-retroviral therapy (ART) self-reporting at 73% and 80% respectively [5]. Relatedly, a study in East Africa, revealed sub-optimal 95-95-95 targets achievements among WECS [6, 7].

High HIV infection risk is prevalent among WECS in Uganda and elsewhere in sub-Saharan Africa, making them a high-risk population (HRP) [6–8]. In Uganda, the proportion of women (15 + years) living with HIV who are virally suppressed are approximately 78.3%, below the UNAIDS target [7, 9]. This is supported by evidence in a study in Kampala Uganda indicating that VLS among WECS on ART was only at 55% [10–12].

Relatedly, VNS is significantly elevated among young persons aged 15–29 years, men, persons reporting inconsistent condom use with non-marital/casual partners, and persons exhibiting hazardous alcohol use in Uganda [13]. This may be as a results of inhibition of engaging in commercial sex (CS) [14] that is often perceived as illegal, and culturally prohibited in many countries including Uganda [14–17]. Relatedly, including WECS in general population surveillance systems may not produce plausible estimates because of methodological biases as a result a need for robust sampling designs specifically focused on hard to reach populations. Similarly, monitoring of 95-95-95 performance for WECS is difficult because most national surveys lump them up with general population and program data is not of the right quality for that purpose. Therefore, utilizing respondent driven sampling (RDS) which is a type of snowball sampling used for analyzing characteristics of socially networked populations without a proper sampling frame [18, 19] may be a solution in culturally closed out societies for such populations. In Africa, many surveys focusing on WECS have used RDS though most have not been scaled up to be representative enough of this population, and in Uganda the surveys have in most cases only focused in Kampala capital city [20–22].

The objective of this study was therefore to find out the magnitude of VNS and factors associated among WECS living with HIV (WECS-LHIV) who had their viral load results using a robust pooled RDS database from Kampala capital city and other 11 survey sites. We examined variables including – age, marital status, education, active syphilis, hr.HPV and reported ART use to identify correlates of VNS. These findings may be utilized to guide HIV programmes and policy focusing on WECS on ART to mitigate new HIV infections in general population and among WECS in Uganda and other low- and medium-income countries (LMIC).

Methods

Survey design, population and sample size

Our data analysis was based on data collected from April 2021-September 2023 in Kampala capital city and 11 other cities and towns namely Arua, Busia, Buvuma Island, Fort Portal, Gulu, Tororo, Jinja, Mbarara, Mbale, Masaka and Lira which were based on sites with high HIV and WECS prevalence. Dubbed “Crane Survey”, these surveys were part of the Uganda Ministry of Health (MoH)’s HRP surveillance system, supported by the United States of America Centers for Disease Control and Prevention (U.S CDC), and implemented by the Makerere University School of Public Health Crane survey project (Supplementary Material 1: Survey Questionnaire Instrument).

The inclusion criteria for this study was women engaged in commercial sex with men 3 months before the survey, having consistently resided in the survey site 6 months before the survey, aged 18–49 years, having presented a valid RDS coupon (seeds excluded because they were not part of the sample), living with HIV and having received their VL irrespective of their ART status. Exclusion criteria was being under 18 years and not having a 200 convergence for HIV prevalence and VL results. In order to collect robust data, through community-based organizations providing services to WECS, HIV implementing partners and WECS in HIV hotspots were approached by the survey team to randomly select one WECS from the respective hotspots. The selected WECS became seeds (WECS that were used to start the sampling process) and were trained on how to invite others with similar characteristics with minimum bias to the survey sites. We only included data of sites that reached a convergence of 200 HIV positives and viral load results irrespective of their ART status because the convergence provided enough sample size power to enable us estimate the 95-95-95 targets for this group.

Data source and management

These were routinely collected data at community surveillance sites of participants choice. Sites started sampling by enrolling a minimum of two seeds selected by WECS from hotspots. Survey participants were trained to refer peers using unique and numbered coupons (paper-based or electronic). In the first visit to the single survey office in each location, presented coupons were verified, and fingerprints were scanned and converted into anonymized alphanumeric codes that were disposed after data collection. Candidate participants were provided electronic information that was in the computer tablets that were used for the survey, including specific interview terms, biological tests offered, and how to recruit peers. Survey participants were then screened for eligibility.

Biological specimen collection included a venous blood draw and self-collected cervical-vaginal swabs. Interviews were then conducted through an electronic interview guide through audio-computer assisted self-interviewing (ACASI) offered in English and other WECS preferred languages from April 2021 to September 2023. In the interviews, participants were asked questions to find out about their HIV and sexually transmitted infections (STI), behavioural risks, socio-demographics and HIV services health seeking behaviours like age, marital and economic status, HIV testing and service uptake, sexual and physical violence, alcohol and drug use, depression and disclosing WECS behaviour to health providers these were selected based on the previous HIV surveys and programme interest in Uganda. The objective of the interviews was measure factors that may be associated with the 95-95-95 targets and viremia of interest for the HIV programme among WECS in Uganda.

Laboratory measures and procedure

HIV serologic rapid tests were conducted onsite by trained nurse counsellors according to the serial algorithm prescribed by Uganda Ministry of Health (MoH) guidelines as follows: Determine rapid test (screening) (Abbott Diagnostics Scarborough, Inc., USA), HIV 1/2 STAT-PAK rapid test (confirmatory), and SD Bioline (tiebreaker test) [23]. For WECS who tested HIV positive, Samples of WECS testing positive ethylenedinitrilotetraacetic(EDTA) blood specimens were stored at 2–8 degrees Celsius until viral load (VL) testing could be run on standard systems such as Cobas Ampliprep or Cobas Taqman that were utilized for the quantification of HIV [24].

For the detection of active syphilis, we used the Dual Path Platform (DPP) syphilis screen and confirmation rapid immunodiagnostic test (Chembio Diagnostics, Medford, NY, USA) [25]. Off-site testing was used as a backup for stockouts of Chembio kits using the Rapid Plasma Reagin Syfacard-R Test (Murex Biotech, Dartford, UK; detects current syphilis) and confirmed by anti-syphilis IgG ELISA (Biotec Laboratories, Suffolk, UK) or the TP hemagglutination assay (TPHA) [26]. Cervical-vaginal swabs were tested for 13 high risk strains of human papillomavirus (hr.HPV) at regional government laboratory hubs using the GeneXpert hr-HPV test (Cepheid, Sunnyvale, CA, USA). Genotypes 16 and 18 are associated with 71%, and subtype 45 with 6%, of all cases of cervical cancer in sub-Saharan Africa [27].

Samples collected at sites other than Kampala were transported daily to regional laboratories for storage and then transported weekly for testing at the Uganda Ministry of Health (MoH) Central Public Health Laboratory (CPHL) in Kampala using the national specimen referral network [28]. The samples from the Kampala site were transported daily to the Infectious Diseases Institute (IDI) laboratory (Mulago Hospital, Kampala) for hr-HPV and VL tests using private transport. All testing at offsite laboratories was scheduled to be conducted within 3 days; offsite results were expected to the survey office in < 14 days, in time for the participants scheduled second visit to the survey office. All participants screened positive were either provided start packs (HIV) at site and referred for further monitoring or treatment (active syphilis and hr.HPV) at health facilities of their choice.

Statistical analysis

Data were cleaned and cross-checked for missing observations from participants. Individual weights for analysis were computed using RDS Analyst [29]. RDS-based weights (Gile’s SS) were applied to the data to account for the varying network size of the participants. Data analysis was conducted using STATA version 16.0.

The dependent variable was VNS (defined as ≥ 1,000 copies/mL) among WECS-LHIV as a binary outcome. The independent variables included sociodemographic characteristics (i.e., age group, location, marital status, highest education level, and financial dependents; Table 1). We also examined biomarkers, including hr-HPV and active syphilis, and behavioural variables, self-reported ART use, history of rape, alcohol abuse measured by the WHO Alcohol Use Disorders Identification Test (AUDIT-C) scale [30], illicit drug use, depression-related symptoms, self-reported disclosure of being engaged in commercial sex to health care providers, and self-reported denial of health care because of CS [31].

Table 1.

Characteristics and univariable analysis of 2,674 virally suppressed and non-suppressed WECS-LHIV irrespective of ART status in 12 urban settings in Uganda, 2021–2023. WECS

Characteristic	Total (n = 2,674)	Viral suppressed (2,007)	Viral non- suppression (667)	p value
WECSS living with HIV viral load status (2,674)		2,007(75.1)	667(24.9)
Socio-demographic
Age (median Interquartile range [IQR])	32 (27–37)	32 (28–37)	29 (25–34)	< 0.0001
Age group (n = 2,674)
18–24	379	243 (64.1)	136 (35.9)	< 0.0001
25–29	655	438 (66.9)	217 (33.1)
30–49	1,640	1,326 (80.9)	314 (19.2)
Marital status (n = 2,672)
Never married	906	628 (69.3)	278 (30.7)	< 0.0001
Currently married	227	172 (75.8)	55 (24.2)
Divorced/widowed/separated	1539	1205 (78.3)	334 (21.7)
Highest education level (n = 2,672)
None	666	498 (74.8)	168 (25.3)	0.952
Primary	1,418	1,063 (75.0)	355 (25.0)
Secondary and above	588	444 (75.5)	144 (24.1)
Financial dependents(n = 2,674)
No	626	460 (73.5)	166 (26.5)	0.298
Yes	2,048	1,547 (75.5)	501 (24.5)
Biomarker related factors
hr-HPV (n = 2029)
Negative	1,399	1,071 (76.9)	321 (23.1)	0.001
Positive	630	443 (70.3)	187 (29.7)
Active syphilis (n = 2654)
Negative	2,086	1,533 (73.5)	553 (26.5)	< 0.0001
Positive	568	459 (80.8)	109 (19.2)
Behavioural factors
Known HIV status (n = 2,674)
No	424	298 (70.3)	126 (29.7)	0.013
Yes	2,250	1,709 (76.0)	541 (24.0)
Self-reported ART use (n = 2,674)
No	838	518 (61.8)	320 (38.2)	< 0.0001
Yes	1,836	1,489 (81.1)	347 (18.9)
Ever-raped/-coerced sex (n = 2,672)
No	1,423	1,047 (73.6)	376 (26.4)	0.063
Yes	1,249	958 (76.7)	291 (23.3)
Alcohol abuse (n = 2,674)
No	1,124	857 (76.3)	267 (23.8)	0.063
Yes	1,550	1,150 (74.2)	400 (25.8)
Drug abuse (n = 2,674)
No	1,662	1,242 (74.7)	420 (25.3)	0.617
Yes	1,012	765 (75.6)	247 (24.4)
Ever injected drugs (n = 2,674)
No	2,274	1,718 (75.6)	556 (24.5)	0.160
Yes	400	289 (72.3)	111 (27.8)
Feeling down, depressed, or hopeless (n = 2,672)
No	311	227 (73.0)	84 (27.0)	0.375
Yes	2,361	1,778 (75.3)	583 (24.7)
Denied health care due to WECS status engaged in commercial sex (n = 2,674)
No	1,662	1,247 (75.0)	415 (25.0)	0.968
Yes	1,012	760 (75.1)	252 (24.9)
Disclosed WECS status to health care provider (n = 2,674)
No	843	668 (79.2)	175 (20.8)	0.001
Yes	1,831	1,339 (73.1)	492 (26.9)

*High risk HPV genotypes highly associated with cervical cancer 16 & 18 DNA testing ** active syphilis test using DPP rapid test by Chembio that accurately diagnose both long-term and active syphilis

We estimated the burden of VNS and identified factors associated using univariable, bivariable and multivariable statistical methods. At the univariable level, descriptive statistics, that is, summary statistics, and frequencies and percentages were used to summarize continuous and categorical variables. Bivariable statistics, namely, Pearson chi-square tests, were used to test for correlations between survey participants’ characteristics and VNS.

The multivariable logistic regression model was fitted with exposure variables—selected based on prior research on risk factors, clinical correlations, and biological plausibility—and built using likelihood ratio tests (LRT), was used to determine factors associated with VNS. Using likelihood ratio test (LRT), each potential independent variable was assessed for its contribution to the overall model adequacy by comparing a model with a given exposure with a model without the variable at a given time. Only a priori confounders such as age and the exposure variables with significant values > 0.2 from the LRT were included in the model. Effect measures were summarized using 95% confidence intervals (95% CIs). All variables with P values < 0.05 and CI that did not include 1 were considered statistically significant and included in the final model.

Results

Out of the 2,674 WECS-LHIV with viral load results included in this analysis, 667(24.9%) were VNS while the remaining 2,007(75.1%) had VLS (Table 1). The median age of the participants was 32 years [interquartile range (IQR)27–37] and did not report any adverse events during the study period that was beyond mitigation.

The majority of participants with VNS were in the ages between 25 and 29 years and 30–49 years, had primary education (25.0%), were divorced/widowed/separated (30.7%), were self-reporting not being on ART (38.2%), had hr.HPV (29.7%), did not have active syphilis (19.2%), had unknown HIV status (29.7%) and had disclosed their WECS status to health workers (26.9%) (Table 1).

In univariable analysis, age, marital status, financial dependents, hr-HPV, active syphilis, known HIV status, self-reported ART use, ever raped/coerced sex, alcohol abuse, ever injected drugs and disclosed WECS status to health workers were associated with VNS(p ≤ 0.05) or (p ≤ 0.2) and were included in the multivariable analysis (Table 2).

Table 2.

Bivariable and multivariable logistic regression of viral non-suppression in 667 WECS-LHIV irrespective of ART status in 12 urban settings in Uganda, 2021–2023

Characteristic	Viral non-suppression % (n/N)	Unadjusted OR (95% CI)	Adjusted OR (95%CI)	p value
Age group (ref = 18–24)	34.3 (136/379)
30–49	17.6 (314/1640)	0.41 (0.29–0.56)	0.46 (0.30–0.69)	< 0.001
Marital status (ref=Single married)	27.5 (278/906)
Currently married	19.8 (55/227)	0.65 (0.42–1.01)	0.89 (0.54–1.48)	0.662
Divorced/widowed/separated	20.5 (334/1539)	0.68 (0.53–0.88)	0.77 (0.56–1.05)	0.093
Highest education level (ref=None)	22.8 (168/666)
Primary	22.9 (355/1418)	1.01 (0.76–1.34)	0.96 (0.68–1.35)	0.662
Secondary and above	21.7 (144/588)	0.94 (0.67–1.31)	0.91 (0.61–1.37)	0.093
Financial dependents(ref = No)	24.3 (166/626)
Yes	22.0 (501/2048)	0.88 (0.67–1.15)
Biomarker related factors
hr-HPV (n = 508) (ref=Negative)	22.6 (321/1392)
Positive	26.8 (187/630)	1.26 (0.95–1.66)	1.22 (0.91–1.63)	0.168
Syphilis (n = 662) (ref=Negative)	23.3 (553/2086)
Positive	20.2 (109/568)	0.84 (0.60–1.15)	0.69 (0.47–1.01)	0.06
Behavioural factors
Known HIV status(ref = No)	25.0 (126/424)
Yes	22.2 (541/2250)	0.85 (0.64–1.14)	1.41 (0.98–2.03)
Self-reported ART use (ref = No)	33.9 (320/838)
Yes	17.0 (347/1836)	0.40 (0.31–0.51)	0.38 (0.28–0.52)	< 0.001
Ever-raped/-coerced sex (No)	24.3 (376/1423)
Yes	20.9 (291/1249)	0.82 (0.65–1.04)
Alcohol abuse(ref = No)	21.5 (267/1124)
Yes	23.7 (400/1550)	1.13 (0.90–1.43)
Drug abuse (ref = No)	22.6 (420/1662)
Yes	22.6 (247/1012)	1.00 (0.79–1.27)
Feeling down, depressed, or hopeless(ref = No)	24.3 (84/311)
Yes	22.4 (583/2361)	0.90 (0.63, 1.28)	0.96 (0.62, 1.48)	0.839
Denied healthcare because of engaged in commercial sex(ref = No)	22.1 (415/1658)
Yes	23.4 (252/1012)	1.08 (0.85–1.37)	1.18 (0.88, 1.58)	0.259
Disclosed WECS status to health care provider (ref = No)	20.8 (175/843)
Yes	26.9 (492/1831)	1.18 (0.91–1.53)	1.17 (0.86–1.59)	0.331

In multivariable analysis, WECS-LHIV had reduced odds of VNS if they were aged 30–49 years ([adjusted odds ratio (aOR): 0.46, 95% confidence interval (CI): 0.30–0.69, p < 0.001)] vs. 18–24 years and if they self-reported ART use [aOR: 0.38, 95% CI: 0.28–0.52, p = < 0.001 vs. non-self-reporting ART]. Though not statistically significant with VNS, WECS reporting being divorced or widowed or separated, having hr.HPV, active syphilis, aware of their HIV status and disclosing WECS status to health workers had high magnitude of VNS (Table 1).

Discussion

To our knowledge, there has never been a robust detailed analysis of VNS and it’s correlates among WECS in Uganda. Our study confirmed the importance of age and ART adherence monitoring in determining VNS among WECS-LHIV [32–34]. High active syphilis and hr.HPV co-morbidities were also reported in this study; these though not associated with VNS may have influence on ART adherence.

There were high magnitude of VNS among WECS as compared to results from studies in general populations and among other WECS-LHIV elsewhere [35] and even among WECS in an HIV program setting in Uganda [10, 11], suggesting a likelihood of HIV new transmissions from WECS-LHIV to their clients, in a marriage setting and probably male clients being a source of the new transmissions as data reveals.

In addition, our results suggest high active syphilis and hr.HPV co-infections among WECS-LHIV with VNS, although these were not significantly correlated with VNS, they contradict other studies in Africa and elsewhere among WECS that reported a positive association between active syphilis and VNS [36, 37] and hr.HPV [38]. These results imply that HIV co-infections may increase the likelihood of poor ART adherence [39].

Our findings indicate likelihood of VNS among young WECS, this is similar to results from other surveys in Uganda where age was seen to affect population based viral non suppression [32, 33].This suggests that challenges still exist among younger WECS-LHIV in knowing their HIV status, initiating ART, and achieving viral suppression [34]. In addition, feeling uncomfortable about seeking HIV services [40] than older WECS and being mixed with older WECS-LHIV at service points because of legal and culturally environment on commercial sex among the young [41, 42].

Our results demonstrate that self-reported ART use may have reduced odds of VNS, meaning WECS-LHIV who did not report ART use were more likely to be VNS. Several studies have found correlation between ART adherence and viral load suppression which is a similar correlation to our study [1, 43, 44]. Relatedly, research has shown that challenges such as mobility, and socioeconomic factors can impede consistent ART use and adherence among WECS-LHIV, potentially contributing to VNS [45]. Harsh cultural and legal environments against WECS have been reported and could lead to reduced uptake of services, such as ART initiation [6–8, 46]. This may suggest that WECS-LHIV self-reporting ART use may be more adhering and have better health seeking behaviour than those not reporting ART use.

It is important to acknowledge the limitations of this survey. The findings are based on cross-sectional data, which makes establishing a causal relationship impossible. Future research should consider longitudinal designs and include large sample sizes powered to be representative of national population to further explore factors associated with VNS among WECS-LHIV especially the young ones. Use of a tiebreaker in HIV diagnosis may have been another limitation of the study; however, we worked with accredited national laboratories to confirm the viral load and HIV results.

Social desirability bias may have occurred because of asking self-reporting sensitive questions. However, we reduced this bias by using ACASI which minimizes stigma and fear of responding to sensitive questions [47]. We also included more towns and cities than the previous biobehavioural surveys in Uganda and elsewhere, providing a sufficient sample size to draw plausible conclusions.

Enhanced ART adherence and viral load monitoring including ramping up interventions that target young women engaged in commercial sex (YWECS) may reduce transmission of new HIV infections. In addition, there is need to integrate HPV and active syphilis screening among WECS-LHIV at maternal and childcare clinics and set up YWECS friendly corners with trained health workers to reduce discrimination. Finally, robust and national wide disease focused surveillance systems for vulnerable hidden populations should be developed to consistently monitor disease programmes in Uganda.

Conclusions

Young women engaged in commercial sex(YWECS) and those who fear ART use disclosure may increase risk of HIV infections in general population. Increase on frequency of ART adherence and viral load monitoring of YWECS including continuous surveillance of the second and third 95 HIV continuum of care and treatment for hidden populations may aid in reduction of new HIV infections burden in Uganda and other low- and middle-income countries.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ART

Anti-retroviral therapy

CS

Commercial Sex

HIV

Human immunodeficiency virus

HRP

High Risk Population

WECS-LHIV

Women engaged commercial sex -living with HIV

WECS

Women engaged in commercial sex

VLS

Viral load suppression

VNS

Viral non-suppression

LMIC

Low- and middle-income countries

Author contributions

All authors substantially contributed to the survey design, conduct, or data analysis and interpretation; wrote or edited parts of the manuscript; and approved the final version for publication.

Funding

This project has been supported by the President’s Emergency Plan for AIDS Relief (PEPFAR) through the U.S. Centers for Disease Control and Prevention under the terms NU2GGH002307. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the funding agencies. The funder was not involved in data collection.

Data availability

The datasets used and/or analyzed during the current survey are available from the corresponding author on reasonable request.

Declarations

Ethical Approval and Consent to Participants

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. The survey protocol was approved by Makerere University School of Public Health’s Higher Degrees Research and Ethical Committee, the Population Council Institutional Review Board, and the US CDC. This activity was reviewed by Makerere University’s School of Public Health Research and Ethics Committee, as well as by the US CDC, where it was deemed not to be research but a public health surveillance activity for program monitoring, and was conducted consistent with applicable federal law and CDC policy 45 C.F.R. part 46.102(1)(2), 21 C.F.R. part 56; 42 U.S.C.§ 241(d); 5 U.S.C. §552a; 44 U.S.C. §3501 et seq. The protocol was also approved by the Uganda National Council of Science and Technology. Verbal informed consent was obtained with a waiver for written consent obtained because of the legal and social environment; participants were compensated for their transport costs, time, and recruitment efforts. All data including finger prints which used as an alphanumeric for unique identification was de-identified after data collection and disposal periods for respective ethical committees respected.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.