Effect of an mHealth intervention on retention in HIV pre-exposure prophylaxis services among female sex workers: Preliminary evidence of the use of the Jichunge app in Dar es Salaam,Tanzania

Abstract

Background

Mobile health (mHealth) applications have been reported to be effective in promoting access and adherence to health services. However, knowledge about their effect on retention in HIV preventive services among at-risk populations in sub-Saharan Africa is limited.

Objective

We aimed to evaluate the effect of the Jichunge mHealth application on retention in HIV pre-exposure prophylaxis (PrEP) services among female sex workers in Dar es Salaam, Tanzania.

Methods

We used respondent-driven sampling to recruit female sex workers eligible for PrEP and who owned a smartphone. All study participants were provided with a smartphone application (Jichunge app) aiming to promote PrEP use through medication reminders, easy access to PrEP information, online consultations with a doctor and/or peer educator, and online discussions between PrEP users. The effect of optimal use of the Jichunge app on retention in PrEP services at 1-month was modelled using log-binomial regression.

Results

A total of 470 female sex workers with a median age of 26 (interquartile range: 22–30) years were recruited. Overall, 27.7% of female sex workers were retained in PrEP services at 1 month. Retention was twice as high among optimal users of the app than among sub-optimal users (adjusted risk ratios = 2.00, 95% confidence interval (CI): 1.41–2.83, p < 0.001).

Conclusion

The optimal use of the Jichunge mHealth application was significantly associated with higher retention in PrEP services among female sex workers in Dar es Salaam.

Keywords

Mobile health mHealth retention pre-exposure prophylaxis PrEP female sex workers smartphone Africa

Introduction

The incidence of HIV has decreased by more than 50% since the late 1990s, but the reduction is far from evenly distributed between populations.^1,2 Female sex workers are among the groups that still carry a high burden of new HIV infections.^1,3,4 In Tanzania, a recent bio-behavioral study estimated the HIV prevalence among female sex workers in Dar es Salaam to be 15.3%,⁵ more than twice than among women in general.⁶ HIV interventions targeting female sex workers are therefore called for and could lead to a significant public health impact and contribute to the achievement of the global 2030 goal of ending HIV as a public health challenge.

Pre-exposure prophylaxis (PrEP) is antiretroviral medication taken prophylactically by HIV negative individuals to prevent HIV acquisition. It has been shown to be effective and is recommended for use among people at increased risk of HIV.⁷ However, low retention in PrEP services is a major challenge and negatively affects the effectiveness of PrEP programming.^8–10 In Tanzania, PrEP roll-out is ongoing,¹¹ but like in many other countries, low retention is expected to become a challenge. PrEP interventions are complex and their implementation associated with a range of barriers at different levels.^12–15 Finding ways of supporting retention in care could potentially strengthen the impact of PrEP.

Increased access to mobile phones, expansion of mobile networks and innovations in mobile technologies offer an opportunity for leveraging mobile services for public health benefits.¹⁶ Also, the increase in internet use globally represents an opportunity for internet-based interventions that can help mitigate several health challenges.^17–19 Mobile health (mHealth) applications have increasingly been reported to be useful in promoting the use of different medications.^19–22

The effectiveness of text messaging interventions in improving adherence to PrEP and retention in PrEP services has been demonstrated among men who have sex with men in the United States.^21–24 However, there is limited evidence on the impact of mHealth interventions on PrEP use among people at increased risk of HIV infection in sub-Saharan Africa.¹ Considering this, we aimed to evaluate the effect of a smartphone mHealth application called Jichunge (a Swahili word meaning ‘protect yourself’) that was developed with the aim of supporting the use of PrEP among men who have sex with men and female sex workers. In this paper, we present results of the effect of the use of the Jichunge app on retention in PrEP services at 1 month among female sex workers in Dar es Salaam, Tanzania.

Materials and methods

Study design and settings

This paper is based on the analysis of part of the data emanating from a pragmatic quasi-experimental trial for PrEP roll-out in Tanzania (PREPTA) described in our previous publication.²⁵ In short, PREPTA is a research project with the overall aim of assessing the effectiveness of an mHealth intervention in promoting retention in PrEP services and adherence to PrEP among female sex workers and men who have sex with men in two cites of Tanzania: Dar es Salaam (intervention arm) and Tanga (control arm). The project is implemented by Muhimbili University of Health and Allied Sciences (MUHAS), Tanzania and University of Oslo (UiO), Norway. This paper focuses on the female sex workers only and utilises baseline data and 1-month follow-up data from Dar es Salaam.

Intervention

The Jichunge app is a smartphone-based mHealth application designed to provide support during start-up and use of PrEP. Details about the app development process have been described in our previous publication.²⁶ The app offers user information about PrEP, reminds them to take their daily PrEP pill, provides an opportunity to consult a doctor and/or a peer educator online, and operates an online forum for anonymous discussions between PrEP users. The overall aim of the app is to improve adherence to PrEP and retention in the PrEP programme.

Study population

We included women aged 18 years and above who had sold sex during the past 3 months. Participants were residents of the city of Dar es Salaam (they had an address in the city and had lived there for the past 6 months) who owned a smartphone and were starting PrEP.

Sample size estimation

The sample size was estimated using a formula for cohort studies.²⁷ Since the retention in PrEP services among female sex workers was not known during the design of the study, we assumed that it was between 10% and 90%. Thus, an estimated 50% of retention rate was used in sample calculation because it would provide the optimum sample size.²⁸ A precision of 5% and a design effect of 2 were used in estimating the sample size to achieve the minimum statistical power of 80%. After adjusting for a 20% potential loss to followup, the minimum sample size was estimated to be 423 female sex workers.

Study procedures

Recruitment

Participants were recruited between March and June 2021 using respondent-driven sampling.^29,30 All participants were screened for PrEP eligibility as per national guidelines and those who qualified were provided with PrEP pills and thereafter invited to take part in the study. Those who consented to participate received the Jichunge app, attended an app onboarding session and thereafter sat for a face-to-face baseline interview with a trained interviewer and thereafter during the 28 days follow-up study visit.

Data collection

Information on socio-demographics and other structural factors at baseline were collected using a questionnaire administered by trained research assistants. Participants were given an appointment to come back to the clinic for PrEP refill 28 days following enrolment, at which time they were interviewed about PrEP experiences and PrEP adherence. From the app, we also continuously collected data on the use of its different functionalities (opening the app, medicine registration, reading editorial contents, accessing online consultations or entering the online discussion forum). Data from the app were collected from all participants regardless of whether they came back to the 1-month follow-up clinic visit or not.

Research ethics and consent

The study was conducted in accordance with the Declaration of Helsinki. Details on how we worked to protect participants’ privacy have been described in a previous article.²⁵ Briefly, all questionnaire data were stored and processed on a secure server developed particularly for the handling of sensitive research data (’Services for sensitive data’ or TSD³¹) owned by the University of Oslo. The Jichunge app does not store, transfer or expose sensitive data. No element in the app makes reference to any information that can identify a participant. The app does not contain any feature to track the location of a participant and does not collect any other information from their smartphones apart from app usage. Written informed consent was obtained from all participants involved in the study. Participants were informed about collection and use of interview and the Jichunge app data.

Study variables

Outcome variable

The primary outcome in this study was retention in PrEP services at 1 month after initiating PrEP. The main reason for choosing to use a 1-month retention as an outcome variable was because it marks the first visit to the clinic after initiation of PrEP and hence provides an opportunity among PrEP users to share their first experience of using PrEP with the healthcare provider. Ability to ensure retention during the first month of PrEP use is critical in informing long-term use. As most complications and access challenges experienced early are more likely to have effect in long-term use of PrEP. Thus, understanding an early retention rate and associated factors would inform early and timely positive modification of services. In addition, the PrEP implementation framework in Tanzania requires that HIV rapid test should be repeated during the first visit to confirm eligibility for PrEP treatment.³² Hence, retention at 1 month was considered one of the meaningful targets in PrEP programming and is likely to affect the long-term cascade of PrEP.³³ Participants were considered to have been retained to services if they visited the clinic within 4 weeks (28 days) after the scheduled date of their follow-up appointment. This definition was informed by Goodman et al.³⁴ who defined retention at 3-month as a visit after 3 months $\pm$ 30 days, and by Hovaguimian et al.³⁵ who defined PrEP programme uptake as ‘the proportion of participants attending first follow-up visit as scheduled $\pm$ 4 weeks’.^34,35

Exposure variables

The exposure variable was what we refer to as ’optimal use’ of the Jichunge app, defined as opening the app and use it at least three out of the four main app functionalities during the first 30 days after enrolment. Access and use of any three of the app functionalities exposed the user to substantial information about PrEP which were sufficient to promote service engagement. The functionalities albeit different, they conveyed similar detailed message about PrEP use to the user. Moreover, we refer to these users as optimal because after installation and introduction to the app, they did not only choose to open the app on their own, but also explored the majority of the app's functionalities. The four functionalities are medicine registration, PrEP editorial contents, the online consultation services and the online discussion forum.

Confounder variables

These were socio-demographics, sex work characteristics and other socio-structural factors which were collected at baseline and considered to potentially have an association with both the outcome and exposure variables. They include the age of the respondent, marital status, education level, social support, awareness of PrEP, PrEP stigma, sex work stigma, self-perceived HIV risk, income from sex work, steady partners and condom use. The measurement of scale variables also described in our previous publication²⁵ is provided below:

Social support was measured using an 8-item Likert scale that was adapted from the Duke–UNC Functional Social Support Questionnaire (FSSQ).³⁶ Participants were asked to respond to one of the five responses (1 = Much less than I would like; 2 = Less than I would like; 3 = Some, but would like more; 4 = Almost as much as I like and 5 = As much as I like). We computed the total score for all items and a total score below 32 was considered as ‘inadequate social support’. The scale gave a Cronbach's alpha of 0.88.

Sex work stigma and perceived PrEP stigma were measured using 13 and 10 scale items, respectively, each with five response options (1 = Strongly disagree, 2 = Disagree, 3 = Neither disagree nor agree, 4 = Agree and 5 = Strongly agree), and gave a Cronbach alpha of 0.84 and 0.88, respectively; signifying high reliability. Sex work stigma was thereafter categorised into three groups (‘low’ for scores ≤26, ‘moderate’ for scores between 27 and 38, and ’high’ for scores ≥39). For PrEP stigma, a score above 30 was considered ‘high’.

PrEP awareness was measured using eight true or false questions about PrEP. Participants who answered more than six questions correctly were categorised as being highly aware of PrEP.

Statistical analyses

Using the Chi-square test, we assessed the association between retention in PrEP services at 1 month by the Jichunge app use, socio-demographics, sex work characteristics and other socio-structural factors. Variables with p-values <0.25 in the bivariate analysis were included in the regression modelling. Since the outcome variable was binary and common (p = 27.7%), we used generalised linear model with a log link and binomial distribution (log-binomial regression)³⁷ to model the effect of optimal use of the Jichunge app on retention in PrEP services at 1 month. Log-binomial regression was chosen because it estimates the risk ratios (RR) directly as opposed to logistic regression which gives odds ratios that can overestimate the association if the outcome variable is common (p > 10%).³⁸ We present crude RR and adjusted RR (aRR) with their corresponding 95% confidence intervals. All analyses were performed using STATA version 17 and a p-value <5% was considered statistically significant.

Results

Socio-demographic characteristics at baseline

A total of 470 female sex workers with a median age of 26 (interquartile range: 22–30) years were recruited. Three-quarters (75.5%) of the participants reported to have never been married, and more than half (58.9%) had at least some secondary education.

Optimal use of the Jichunge app

User statistics retrieved from the app administration portal indicated that about half (52.6%) of study participants were optimal users of the app (i.e., they had used at least three app functionalities during the first 30 days after having the app installed on their phones).

Retention in PrEP services

Overall, about one-quarter (27.7%) of the study participants were retained in PrEP services at 1 month following PrEP initiation. Retention was significantly higher among optimal users of the Jichunge app than among sub-optimal users (37.7% for optimal users versus 16.6% for non-optimal users; p < 0.001). Retention increased significantly and linearly with age (23.1%, 28.8% and 42.2%, for those aged 18–24, 25–34 and above 35, respectively; p = 0.012). Retention was high (40.7%) among those reporting low sex work stigma and decreased significantly among those reporting moderate stigma (27.6%) and high stigma (15.6%). Furthermore, there was a significant association between PrEP awareness and retention in PrEP services (32.3% vs. 23.1%; p = 0.026) (Table 1).

Table 1.

Distributions of socio-demographics characteristics and Jichunge app use by retention in PrEP services.

Variable	N (%)	Not retained: n (%)	Retained: n (%)	p-Value
Jichunge app use				<0.001
Optimal users	247 (52.6)	154 (62.3)	93 (37.7)
Non-optimal users	223 (47.4)	186 (83.4)	37 (16.6)
Age (years)				0.012^b
18–24	199 (42.3)	153 (76.9)	46 (23.1)
25–34	226 (48.1)	161 (71.2)	65 (28.8)
35+	45 (9.6)	26 (57.8)	19 (42.2)
Marital status				0.094
Never married	361 (76.8)	268 (74.2)	93 (25.8)
Married or previously married	109 (23.2)	72 (66.1)	37 (33.9)
Education level				0.086
No formal	28 (6.0)	25 (89.3)	3 (10.7)
Primary	165 (35.1)	114 (69.1)	51 (30.9)
Secondary+	277 (58.9)	201 (72.6)	76 (27.4)
Social support^a				0.108
Inadequate	283 (60.2)	213 (75.3)	70 (24.7)
Adequate	184 (39.1)	126 (68.5)	58 (31.5)
PrEP awareness				0.026
Low	238 (50.6)	183 (76.9)	55 (23.1)
High	232 (49.4)	157 (67.7)	75 (32.3)
Sex work stigma score^a				0.032^b
Low	27 (5.7)	16 (59.3)	11 (40.7)
Moderate	398 (84.7)	288 (72.4)	110 (27.6)
High	32 (6.8)	27 (84.4)	5 (15.6)
Self-perceived HIV risk^a				0.640
High	334 (71.1)	239 (71.6)	95 (28.4)
Medium	41 (8.7)	28 (68.3)	13 (31.7)
Low/none	60 (12.8)	46 (76.7)	14 (23.3)
Don’t know	33 (7.0)	26 (78.8)	7 (21.2)
Perceived PrEP stigma^a				0.532
Low	355 (75.5)	254 (71.6)	101 (28.4)
High	114 (24.3)	85 (74.6)	29 (25.4)
Reported monthly earnings from sex work (TZS)				0.291^b
$\leq$ 150,000	116 (24.7)	83 (71.6)	33 (28.4)
150,001–299,999	108 (23.0)	83 (76.9)	25 (23.1)
300,000–444,999	144 (30.6)	106 (73.6)	38 (26.4)
$\geq$ 450,000	90 (19.1)	58 (64.4)	32 (35.6)
Having a steady partner				0.346
Yes	305 (64.9)	225 (73.8)	80 (26.2)
No	165 (35.1)	115 (69.7)	50 (30.3)
Used condom during last sex with paying partner^a				0.154
Yes	215 (45.7)	149 (69.3)	66 (30.7)
No	254 (54.0)	191 (75.2)	63 (24.8)
Accepting condom-less sex for more pay^a				0.216
Yes	249 (53.0)	174 (69.9)	75 (30.1)
No	220 (46.8)	165 (75.0)	55 (25.0)

N is < 470 due to missing observations (3 for social support, 13 for sex work stigma, 2 for self-perceived HIV risk, 1 for condom use and condom-less sex variables).

The p-value based on the Chi-square trend test.

PrEP: pre-exposure prophylaxis; TZS: Tanzanian Shillings.

Independent effect of the use of the Jichunge app on retention in PrEP services

The results of log-binomial regression on the effect of the use of the Jichunge app on retention in the PrEP service at 1 month are presented in Table 2. Retention in PrEP services were two times higher for optimal app users when compared to sub-optimal users (aRR = 2.00, 95% CI: 1.41–2.83, p < 0.001). Other variables that were significantly associated with higher retention in the multivariable model were low sex work stigma (aRR = 2.66, 95% CI: 1.10–6.45, p = 0.030) and having used a condom during last sex with a paying client (aRR = 1.38, 95% CI: 1.01–1.91, p = 0.046).

Table 2.

Log-binomial regression analysis of independent effect of Jichunge app on retention in PrEP services.

	Crude estimates		Adjusted estimates
Variable	RR (95% CI)	p-Value	aRR (95% CI)	p-Value
Jichunge app use
Optimal users	2.27 (1.62–3.17)	<0.001	2.00 (1.41–2.83)	<0.001
Sub-optimal users	1	Ref.	1	Ref.
Age (years)
18–24	1	Ref.	1	Ref.
25–34	1.24 (0.90–1.72)	0.189	0.93 (0.65–1.32)	0.685
35+	1.83 (1.19–2.79)	0.006	1.17 (0.72–1.88)	0.531
Marital status
Married or previously married	1	Ref.	1	Ref.
Never married	0.76 (0.55–1.04)	0.086	0.86 (0.62–1.20)	0.377
Education level
No formal education	1	Ref.	1	Ref.
Primary	2.88 (0.97–8.61)	0.058	2.30 (0.77–6.89)	0.137
Secondary+	2.56 (0.86–7.59)	0.09	2.07 (0.70–6.14)	0.191
Social support
Adequate	1.27 (0.95–1.71)	0.106	1.27 (0.95–1.70)	0.105
Inadequate	1	Ref.	1	Ref.
PrEP awareness
High	1.40 (1.04–1.88)	0.027	1.15 (0.86–1.54)	0.350
Low	1	Ref.	1	Ref.
Sex work stigma
Low	2.61 (1.03–6.57)	0.042	2.66 (1.10–6.45)	0.030
Moderate	1.77 (0.78–4.02)	0.173	1.66 (0.74–3.71)	0.218
High	1	Ref.	1	Ref.
Used condom during last sex with paying partner
Yes	1.23 (0.92–1.66)	0.155	1.38 (1.01–1.91)	0.046
No	1	Ref.	1	Ref.
Accepting condom-less sex for more pay
Yes	1.20 (0.90–1.62)	0.219	1.34 (0.98–1.83)	0.069
No	1	Ref.	1	Ref.

aRR: adjusted risk ratios; CI: Confidence interval; Ref: reference; PrEP: pre-exposure prophylaxis; RR: risk ratios.

Discussion

Evidence shows that some mHealth applications are effective in promoting the use of different health services. In this study, we aimed to determine the effect of a smartphone mHealth application on early retention in PrEP services (1-month retention) among female sex workers in Dar es Salaam, Tanzania. The findings from this study have the potential to inform different national and global strategies to reduce new HIV infections.

Overall, about a quarter of study participants were retained in the PrEP programme at 1 month. This retention level resembles that found in a recent study in neighbouring Kenya where 24% of female sex workers were retained in PrEP and other HIV prevention services after 3 months.³⁹ However, the proportion differs significantly from that reported by a study in Senegal, where more than three-quarters of female sex workers were retained in PrEP services after 6 months.⁴⁰ One reason for the difference could be that sex work is legal and regulated by policy in Senegal (as opposed to the situation in Tanzania and Kenya),⁴¹ and this could have positive implications for the will to seek and remain in health services. Our findings underscore the considerable challenge of poor retention in PrEP services that has been reported elsewhere^42,43 and calls for more and better strategies to support retention in PrEP services.

mHealth interventions like the Jichunge app can be one such promising innovation. Clearly, however, the impact of mobile phone apps depends on how actively they are used. In this study, objective user statistics demonstrated that half of the participants were optimal users of the app (defined as persons who had used at least three functionalities of the Jichunge app during the first 30 days after enrolment). We interpret this as a sign that the app was quite well received. A similar study that assessed the use of a youth-tailored mobile health intervention (“PrEPmate”) aiming to support PrEP retention and adherence among young people at high risk of HIV in Chicago, Illinois reported use of different app services ranging from 58% (the proportion that accessed PrEP information) to 35% (the proportion that had engaged in discussion with other PrEP users).²³ However, more persuasively designed mobile phone applications are needed to attract more users and maximise the benefit of the intervention.

In the bivariate analysis, we observed a significant higher retention in the PrEP service among women who were optimal users of the Jichunge app. In multivariable regression analysis (after adjusting for potential confounders), optimal use of the Jichunge app independently predicted retention in PrEP services among female sex workers. The retention in the PrEP programme was twice as high among optimal users than among non-optimal users.

The findings from this study provide the preliminary evidence on the effect of the Jichunge mHealth intervention in improving retention in PrEP services. These findings compare with a study by Pitye et al.⁴⁴ which found that two-way text messages improved the return to the first PrEP follow-up visit among women initiating PrEP treatment in Kenya. Our findings are also in line with studies into the effect of mHealth on PrEP use that have been conducted elsewhere in other populations at HIV increased risk, such as men who have sex with men.^21,23,24

While well-designed mHealth applications may be effective in promoting PrEP use, they may not be equally effective in all settings and populations. For example, Colson et al.⁴⁵ found that an enhanced mHealth application to support PrEP adherence did not improve adherence among men who have sex with men and transgender women in New York in the United States. Also, Songtaweesin et al.⁴⁶ found that mobile phone applications did not provide additional PrEP adherence benefits over youth friendly services. Furthermore, the effect of mHealth interventions on PrEP use is likely to differ between individuals, by age, and by other socio-economic characteristics.²⁵ This underlines the need for designers and implementers of mHealth interventions to closely consider the local context and population specific factors.

One potential limitation of this study is that the findings are based on comparisons within a population (all participants received the app). As no standard control group (group not receving the app) was used in this analysis, it is difficult to claim that the effect of the Jichunge app observed represent actual intervention effectiveness measure. Future comparisons between intervention and control groups are recommended for effectiveness results. The current study provides evidence only on short term retention in PrEP services, and we propose further studies on the effect of mHealth and long-term retention. The strength of this analysis is the use of actual app usage data which are free of desirability bias and we recommend future analysis of the impact of long-term use of the app.

Conclusion

While only a quarter of the participants were retained in the PrEP service at 1 month, the proportion of retained participants was significantly higher among optimal users of the Jichunge app than among non-optimal app users. Retention in PrEP services was two times higher for the Jichunge app optimal users when compared to non-optimal users. Our findings suggest that smartphone apps that offer services such as online consultations, pill-taking reminders, information about PrEP and discussion between PrEP users hold a potential to promote retention in PrEP services among HIV at-risk populations in East Africa and beyond. Innovative mHealth applications could therefore contribute to the achievement of the global goal of ending HIV by 2030.

Footnotes

Acknowledgements

The technical development and operation of the Jichunge app was carried out by Sveve AS, Teiealleen 28, 2030 Nannestad, Norway, and we would like to acknowledge Ivar Moen and Ole Habberstad for their technical support throughout the research process. We extend our sincere thanks to Inga Haaland, Willhellmuss Mauka, Patrick Kazonda and Brenda Mashiri for their efforts and support during fieldwork for this study. We also extend our thanks to Ashura Rangi, Frank Pius, Martine Alfred, Zainab Mashi, Lucia Akaro, Richard Senkondo, Happy Mbaga, Hellen Mwita, Angelina Mwimba, Winnie Jeremiah, Epmack Raymond, Nivece Kinemo, Alexandrina Kajuna and Ibrahim Mauki for their assistance during data collection. Last but not least, we extend our thanks to all the study participants for accepting to participate in this study.

Availability of data

The data for this analysis will be available from the principal investigator (PI) of the PREPTA project upon reasonable request.

Contributorship

CHM conceptualised, designed the study, analyzed the data and drafted the manuscript. EJM and MRK conceptualised, designed the study, interpreted the findings and revised the manuscript. HOL, KM, MTL and EM conceptualised and revised the manuscript. All authors read and approved the final version of the manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

This study was approved by the National Health Research Ethics committee in Tanzania (protocol code: NIMR/HQ/R.8a/Vol. IX/3454) as well as the Regional Ethical Committee (REK) in Norway (protocol code: 33675).

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Research Council of Norway through the Global Health and Vaccination Programme (GLOBVAC) [project number 285361]. The project is also part of the European & Developing Countries Clinical Trials Partnership (EDCTP2) programme supported by the European Union.

Guarantor

EJM.

Trial Registration

International Clinical Trials Registry Platform (Registration number: PACTR202003823226570);

ORCID iDs

Christopher H Mbotwa

Kåre Moen

References

Joint United Nations Programme on HIV/AIDS. UNAIDS data 2021. Geneva, https://www.unaids.org/en/resources/documents/2021/2021_unaids_data (2021, accessed 30 May 2022).

World Health Organization. Global progress report on HIV, viral hepatitis and sexually transmitted infections, 2021. Accountability for the global health sector strategies 2016–2021: actions for impact. Geneva, https://www.who.int/publications/i/item/9789240027077 (2021, accessed 31 May 2022).

Papworth

Ceesay

, et al. Epidemiology of HIV among female sex workers, their clients, men who have sex with men and people who inject drugs in West and Central Africa. J Int AIDS Soc 2013; 16: 18751.

Shannon

Crago

Baral

, et al. The global response and unmet actions for HIV and sex workers. The Lancet 2018; 392: 698–710.

Mizinduko

Moen

Likindikoki

, et al. HIV prevalence and associated risk factors among female sex workers in Dar es Salaam, Tanzania: tracking the epidemic. Int J STD AIDS 2020; 31: 950–957.

Tanzania Commission for AIDS (TACAIDS) and Zanzibar AIDS Commission (ZAC). Tanzania HIV Impact Survey (THIS) 2016-2017 , https://www.nbs.go.tz/index.php/en/census-surveys/health-statistics/hiv-and-malaria-survey/382-the-tanzania-hiv-impact-survey-2016-2017-this-final-report (2018, accessed 16 January 2020).

World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendation for the public health approach - 2nd ed. , https://www.who.int/publications/i/item/9789241511124 (2016, accessed 8 February 2020).

Koss

Charlebois

Ayieko

, et al. Uptake, engagement, and adherence to pre-exposure prophylaxis offered after population HIV testing in rural Kenya and Uganda: 72-week interim analysis of observational data from the SEARCH study. Lancet HIV 2020; 7: e249–e261.

Cowan

Delany-Moretlwe

. Promise and pitfalls of pre-exposure prophylaxis for female sex workers. Curr Opin HIV AIDS 2016; 11: 27–34.

10.

Chan

Patel

Mena

, et al. Long-term retention in pre-exposure prophylaxis care among men who have sex with men and transgender women in the United States. J Int AIDS Soc 2019; 22: e25385–e25385.

11.

PrEPWatch. A snapshot of PrEP scale-up, registration and resources for Tanzania., https://www.prepwatch.org/country/tanzania/ (2022, accessed 31 May 2022).

12.

Kadiamada-Ibarra

Hawley

Sosa-Rubí

, et al. Barriers and facilitators to pre-exposure prophylaxis uptake among male sex workers in Mexico: an application of the RE-AIM framework. BMC Public Health 2021; 21: 2174.

13.

Valente

Mimiaga

Chan

, et al. Health service- and provider-level factors influencing engagement in HIV pre-exposure prophylaxis care among male sex workers. AIDS Patient Care STDS 2021; 35: 279–287.

14.

Mayer

Owaraganise

Kabami

, et al. Distance to clinic is a barrier to PrEP uptake and visit attendance in a community in rural Uganda. J Int AIDS Soc 2019; 22: e25276–e25276.

15.

Hillis

Germain

Hope

, et al. Pre-exposure prophylaxis (PrEP) for HIV prevention among men who have sex with men (MSM): a scoping review on PrEP service delivery and programming. AIDS Behav 2020; 24: 3056–3070.

16.

ITU. Measuring digital development: Facts and figures 2021, https://www.itu.int/en/ITU-D/Statistics/Pages/facts/default.aspx (2021, accessed 31 May 2022).

17.

Hampshire

Mwase-Vuma

Alemu

, et al. Informal mhealth at scale in Africa: opportunities and challenges. World Dev 2021; 140: 105257.

18.

van Luenen

Garnefski

Spinhoven

, et al. Guided internet-based intervention for people with HIV and depressive symptoms: a randomised controlled trial in the Netherlands. Lancet HIV 2018; 5: e488–e497.

19.

World Health Organization. Global difusion of eHealth: making universal health coverage achievable. Report of the third global survey on eHealth . Geneva: WHO Document Production Services, http://www.who.int/goe/publications/global_diffusion/en/ (2016, accessed 1 June 2022).

20.

Horvath

Azman

Kennedy

, et al. Mobile phone text messaging for promoting adherence to antiretroviral therapy in patients with HIV infection. Cochrane Database Syst Rev 2012. Epub ahead of print 14 March 2012. DOI: 10.1002/14651858.CD009756.

21.

Weitzman

Zhou

Kogelman

, et al. Mhealth for pre-exposure prophylaxis adherence by young adult men who have sex with men. mHealth 2021; 7: 44.

22.

Fuchs

Stojanovski

Vittinghoff

, et al. A Mobile health strategy to support adherence to antiretroviral preexposure prophylaxis. AIDS Patient Care STDS 2018; 32: 104–111.

23.

Liu

Vittinghoff

von Felten

, et al. Randomized controlled trial of a mobile health intervention to promote retention and adherence to preexposure prophylaxis among young people at risk for human immunodeficiency virus: the EPIC study. Clin Infect Dis 2019; 68: 2010–2017.

24.

Moore

Jain

Dubé

, et al. Randomized controlled trial of daily text messages to support adherence to preexposure prophylaxis in individuals at risk for human immunodeficiency virus: the TAPIR study. Clin Infect Dis 2018; 66: 1566–1572.

25.

Mbotwa

Kazaura

Moen

, et al. Predictors of mHealth use in promoting adherence to pre-exposure prophylaxis among female sex workers: an evaluation of the Jichunge intervention in Dar es Salaam, Tanzania. BMC Health Serv Res 2022; 22: 859.

26.

Mauka

Mbotwa

Moen

, et al. Development of a mobile health application for hiv prevention among at-risk populations in urban settings in East Africa: a participatory design approach. JMIR Form Res 2021; 5: e23204.

27.

Kirkwood

Sterne

JAC

. Essential medical statistics. 2nd ed. Hoboken: Blackwell Publishing, 2003.

28.

Naing

Nordin

Abdul Rahman

, et al. Sample size calculation for prevalence studies using Scalex and ScalaR calculators. BMC Med Res Methodol 2022; 22: 209.

29.

Heckathorn

. Respondent-driven sampling: a new approach to the study of hidden populations. Soc Probl 1997; 44: 174–199.

30.

World Health Organization (WHO). Regional Office for the Eastern Mediterranean. Introduction to HIV/AIDS and sexually transmitted infection surveillance module 4: introduction to respondent-driven sampling. In: Introduction to HIV/AIDS and sexually transmitted infection surveillance. Cairo: World Health Organization, 2013, pp.1–392.

31.

UiO. About TSD, https://www.uio.no/english/services/it/research/sensitive-data/about/description-of-the-system.html (accessed 26 February 2022).

32.

United Republic of Tanzania. Ministry of Health Community Development Gender Elderly and Children. National AIDS Control Program. Implementation framework for pre-exposure Prophylaxis of HIV in Tanzania Mainland . 2021.

33.

Spinelli

Glidden

Anderson

, et al. Brief report: short-term adherence marker to PrEP predicts future nonretention in a large PrEP demo project: implications for point-of-care adherence testing. JAIDS J Acquir Immune Defic Syndr 2019; 81. Epub ahead of print 2019. DOI: 10.1097/QAI.0000000000002005.

34.

Goodman

Patten

Castillo-Mancilla

, et al. Factors associated with retention and adherence in a comprehensive, diverse HIV pre-exposure prophylaxis clinic. AIDS Res Hum Retroviruses 2022; 38: 327–335.

35.

Hovaguimian

Martin

Reinacher

, et al. Participation, retention and uptake in a multicentre pre-exposure prophylaxis cohort using online, smartphone-compatible data collection. HIV Med 2022; 23: 146–158.

36.

Broadhead

Gehlbach

De Gruy

, et al. The Duke–UNC functional social support questionnaire: measurement of social support in family medicine patients. Med Care 1988; 26: 709–723.

37.

Cummings

. Methods for estimating adjusted risk ratios. Stata J Promot Commun Stat Stata 2009; 9: 175–196.

38.

McNutt

L-A

Xue

, et al. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 2003; 157: 940–943.

39.

Eshikumo

Awuor

Blanco

, et al. Factors associated with retention in HIV prevention and treatment clinical services among female sex workers enrolled in a sex workers’ outreach program (SWOP) in Nairobi, Kenya. AIDS Behav 2022; 26: 2969–2980.

40.

Sarr

Gueye

Mboup

, et al. Uptake, retention, and outcomes in a demonstration project of pre-exposure prophylaxis among female sex workers in public health centers in Senegal. Int J STD AIDS 2020; 31: 1063–1072.

41.

Ito

Lépine

Treibich

. The effect of sex work regulation on health and well-being of sex workers: evidence from Senegal. Health Econ 2018; 27: 1627–1652.

42.

Kagaayi

Batte

Nakawooya

, et al. Uptake and retention on HIV pre-exposure prophylaxis among key and priority populations in South-Central Uganda. J Int AIDS Soc 2020; 23: e25588.

43.

Schumacher

Chandran

, et al. Patterns of PrEP retention among HIV pre-exposure prophylaxis users in Baltimore City, Maryland. JAIDS J Acquir Immune Defic Syndr 2020; 85: 593–600.

44.

Pintye

Rogers

Kinuthia

, et al. Two-way short message service (SMS) communication may increase pre-exposure prophylaxis continuation and adherence among pregnant and postpartum women in Kenya. Glob Heal Sci Pract 2020; 8: 55–67.

45.

Colson

Franks

, et al. Adherence to pre-exposure prophylaxis in Black men who have sex with men and transgender women in a community setting in Harlem, NY. AIDS Behav 2020; 24: 3436–3455.

46.

Songtaweesin

Kawichai

Phanuphak

, et al. Youth-friendly services and a mobile phone application to promote adherence to pre-exposure prophylaxis among adolescent men who have sex with men and transgender women at-risk for HIV in Thailand: a randomized control trial. J Int AIDS Soc 2020; 23: e25564–e25564.