Objective To describe changes in pre-exposure prophylaxis (PrEP) use during and following the COVID-19 lockdown in France (March–May 2020) and identify the factors associated with PrEP discontinuation among men who have sex with men (MSM) after the lockdown.
Methods Data from the, an anonymous, cross-sectional internet survey among MSM in July 2020, were analysed. Among respondents who were using PrEP prior to the lockdown, a binary logistic regression model was used to compare participants who were still taking PrEP (current PrEP users) with those who were not taking PrEP at the time of the survey (former PrEP users).
Results Among 8345 respondents, 946 were PrEP users before the lockdown, of whom 58.8% (n=556) reported stopping PrEP during the lockdown and 15.4% (n=146) were not using PrEP at the time of the survey. Among the 556 who stopped PrEP during lockdown, 86.5% (n=481) reported no sexual activity; 76.8% (n=427) restarted PrEP after lockdown. Former PrEP users were more likely to be younger, not living with a stable male sexual partner, report moderate anxiety, report increased psychoactive drug use during the lockdown, and report not having tested for HIV or STI since the end of the lockdown because they did not know where to go, preferred to wait or for another reason. Reporting fewer male sexual partners in the last 6 months was also significantly associated with being a former PrEP user.
Conclusions MSM adapted PrEP use to their sexual activity during and after the French lockdown. After the lockdown, discontinued PrEP occurred more often among MSM who had fewer sexual partners and had mental health vulnerabilities. These factors could also be predictive of PrEP discontinuation in a more general context. PrEP users should be informed on how to safely stop/start PrEP and on the use of other prevention tools to reduce potential risk exposure during PrEP discontinuation.
- pre-exposure prophylaxis
- treatment adherence and compliance
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
This article is made freely available for personal use in accordance with BMJ’s website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.https://bmj.com/coronavirus/usage
Statistics from Altmetric.com
Several countries have put in place restrictions such as lockdowns to contain the spread of COVID-19. In France, exceptional measures have been implemented, such as a state of health emergency and, from 17 March 2020, a lockdown. These measures aim to break the chains of transmission and to avoid saturation of the healthcare system.1 The sometimes extreme reorganisations of healthcare services and resources to manage the COVID-19 crisis2 have led to a serious impairment of normal medical care and potentially to disruptions in access to care and self-management of treatment.3
The impact of these disruptions is of particular concern for populations who already experience barriers to healthcare access. Additionally, disruptions in services may have an important impact on the response to other infectious diseases such as HIV, in which access to prevention and treatment services is an essential step to reducing transmission and supporting people living with HIV. Recent reports have confirmed that HIV services have been disrupted all over the world.4 5 Regarding men who have sex with men (MSM), respondents to a large multicountry cross-sectional study reported perceived interruptions to testing, pre-exposure prophylaxis (PrEP) and condom access.6 Other studies among MSM have also confirmed issues in accessing HIV testing, prevention and treatment services in spring 2020.7 8
Several studies have also explored the impact of COVID-19 health measures such as lockdowns and social distancing on sexual activity and HIV prevention.9–11 One Australian study found that, among PrEP users before the COVID-19 lockdown, 41.8% reported discontinuing PrEP after COVID-19 restrictions were put into place and that discontinuation of PrEP was associated with reduced sexual activity.12 Another study among HIV-negative MSM found that PrEP users were more likely to report engaging in chemsex during the lockdown and to have a higher median number of sexual partners than non-PrEP users.11
PrEP is recommended by the WHO for MSM and other key populations and has been available and reimbursed in France since 2016. From 2016 to 2019, uptake of PrEP in France, primarily concentrated among MSM, steadily increased, reaching 20 478 PrEP users in June 2019. National French data have shown, however, that PrEP use decreased during 2020.13 Various reasons may explain this PrEP stoppage, such as barriers to HIV services, decrease in sexual activities and mental health vulnerabilities due to the COVID-19 crisis.13 Better understanding of the use of PrEP among MSM and the determinants of PrEP discontinuation may provide important information for the adaptation of prevention services in the context of the ongoing pandemic and beyond. The objectives of this analysis are to describe changes in PrEP use during and following the first COVID-19-related lockdown in France (March–May 2020) and to identify the factors associated with PrEP discontinuation among MSM after the lockdown.
The Enquête Rapport au Sexe special COVID-19 edition (ERAS-COVID-19) is a large French national, voluntary, cross-sectional online survey of MSM conducted between 30 June 2020 (ie, 1 month after the end of France’s first lockdown) and 15 July 2020 under the scientific responsibility of Santé Publique France (French national public health agency). The main aim of the ERAS-COVID-19 study was to characterise the impact of the COVID-19 epidemic on living conditions, sexual behaviours and use of specific healthcare services by MSM during the lockdown. Invitations to participate were posted via banners through social media (Facebook), gay online magazines and dating websites and applications. Programmatic marketing was also used to display banner ads to men whose internet browsing suggested an interest in same-sex sexuality. The majority (75%) of participants accessed the survey via social networks. Clicking on the banners took participants to the survey website, where information about the survey objectives, conditions for participation and data confidentiality were presented.
Participants were asked to validate their consent to access the online questionnaire. The questionnaire was only available in French. It took approximately 15 min to complete the 77 questions, which were derived from previous ERAS editions,14 European studies among MSM15 as well as French studies on the COVID-19 pandemic.16 17 The questionnaire was pilot-tested within the research team and among members from community organisations. No financial incentives were given to participants. No personal identifying information (including internet protocol (IP) addresses) was collected. The only inclusion criteria for participation were being male and being 18 years of age or older.
Among PrEP users before the lockdown, we identified people who were still taking PrEP (current PrEP users) and those who were not using PrEP (former PrEP users) after the lockdown with the following question: ‘Are you currently using PrEP?’ PrEP users who indicated stopping PrEP (not currently taking PrEP at the time of the survey) are thus considered to be former PrEP users and to have ‘discontinued’ PrEP after the lockdown for the purpose of this analysis. The authors acknowledge, however, that this may be a transient change.18
The ERAS-COVID-19 questionnaire (77 items) collected information across several themes, including sociodemographic characteristics and living conditions (age, place of birth, education level, size of city of residence, relationship status at the time of the study, employment status before the lockdown, changes in working conditions, financial situation and living with a stable partner during the lockdown) and mental health information during the lockdown (suffering from loneliness, anxiety assessed with the General Anxiety Disorder-7 (GAD-7 scale)19 in four classes ‘no/mild/moderate/severe’,20 alcohol consumption and psychoactive drug consumption), in addition to psychological stress in the past month (assessed with a cut-off of 5621 on the score built according to five questions on feelings experienced in the past month: very nervous, so discouraged that nothing could cheer you up, calm and relaxed, sad and dejected, and happy). Loneliness was evaluated with the question ‘During the 8 weeks of lockdown, did you experience loneliness?’ The questionnaire also collected information about healthcare access (HIV or STI tests since the end of the lockdown) and sexual behaviours during the previous 6 months, during the lockdown and the last intercourse (number and type of male sexual partners (stable or occasional) and HIV prevention tools used). Concerning their last sexual intercourse with a man, respondents also specified the sex acts performed. Information regarding PrEP use before the lockdown and the reasons for discontinuing PrEP during the lockdown (no longer having sexual encounters, no longer wanting to take PrEP, not having any more medication, PrEP visit was cancelled or postponed, or other reasons) were also collected.
Variables of interest were compared between respondents who were still taking PrEP after the lockdown (current PrEP users at the time of the study) and those who discontinued PrEP after the lockdown (former PrEP users at the time of the study) using χ2 test for categorical variables and Wilcoxon-Mann-Whitney test for continuous variables. Logistic regression models that assess the odds of PrEP discontinuation after the lockdown compared with current PrEP use were used to identify factors associated with ‘PrEP discontinuation after the lockdown’. Variables with a p value <0.20 in the univariable analysis were considered eligible to enter the multivariable model. A backward procedure based on the likelihood ratio χ2 test was used to select significant variables for the final model (p<0.05). Stata/SE V.14.0 software was used for all analyses. Only variables with p<0.05 in the multivariable model are shown in the respective table.
Of the 9488 fully completed ERAS-COVID-19 study questionnaires, 8345 respondents lived in France, reported having sex with men or identified themselves as homosexual or bisexual. Among these French MSM respondents, 946 were PrEP users before the lockdown. Among them, 15.4% (n=146) indicated that they were not currently using PrEP (former PrEP users) after the lockdown vs 84.6% (n=800) who were currently taking PrEP (current PrEP users) (figure 1). Furthermore, among the 946 PrEP users before the lockdown, 58.8% (n=556) discontinued PrEP during the lockdown, including 76.8% (n=427) who declared restarting PrEP after the lockdown.
Characteristics of the study population
Compared with current PrEP users (see table 1), former PrEP users were younger (median (IQR)=33 (26–38) years vs 36 (30–44) years, p<0.001), less likely to be employed before the lockdown (71.9% vs 85.4%, p<0.001) and more often declared a change in their working conditions during the lockdown (55.5% vs 66.1%, p=0.013).
Concerning mental health during the lockdown, former PrEP users were more likely to report suffering from loneliness (26.7% vs 19.1%, p=0.036) and anxiety (70.6% vs 58.6%, p=0.036) and had higher consumption of psychoactive drugs (11.7% vs 9.4%, p=0.002). They also reported more psychological distress in the past month (41.8% vs 31.9%, p=0.020). Regarding healthcare access, they were less likely to be tested at least once for HIV or STI since the end of the lockdown (32.3% vs 70.2%, p<0.001) and more likely to have used PrEP on demand before the lockdown (68.5% vs 44.8%, p<0.001).
Concerning sexual behaviours, former PrEP users were less likely to report having 10 or more partners in the previous 6 months (29.4% vs 63.9%, p<0.001), were less likely to report having limited the number of sexual partners during the lockdown (22.7% vs 33.5%, p=0.011), had fewer sexual partners during the lockdown (31.4% vs 45.6%, p=0.006), were more likely to report the last sexual intercourse was with a stable male sexual partner (45.2% vs 23.4%, p<0.001) and had not used any HIV prevention method during the last anal intercourse (33.8% vs 7.0%, p<0.001).
No significant association was observed between place of birth, education level, size of city of residence, relationship status at the time of the study, decline in financial situation during the lockdown, living with a stable partner during the lockdown, alcohol consumption during the lockdown and discontinuing PrEP after the lockdown.
Reasons for stopping PrEP during the lockdown (n=556)
Among the 556 respondents who stopped PrEP during lockdown (table 2), 86.5% discontinued because they no longer had any sexual activity, 8.5% no longer wanted to take PrEP, 3.6% did not have any more medication, 5.9% declared a cancelled or postponed PrEP visit and 9.4% discontinued for other reasons (not specified). Among this group, current PrEP users (n=427) were more likely to declare not using it because they no longer had any sexual activity during lockdown than former PrEP users (n=129) (90.2% vs 74.4%, p<0.001).
The number of sexual partners in the last 6 months was higher for current PrEP users compared with former PrEP users (median (IQR)=10 (5; 20) vs 4 (1; 10) p<0.001). Additionally, current PrEP users often had the same or more casual male sex partners since the end of the lockdown than former PrEP users (30.5% vs 18.7%, p<0.001, between 11 May and 1 June 2020; 43.8% vs 24.2%, p<0.001, after 2 June 2020, when further restrictions were lifted).
Factors associated with ‘PrEP discontinuation after the lockdown’
Following adjustment on HIV prevention tools used during the most recent anal intercourse (see table 3), participants who were younger (adjusted OR (aOR)=0.96, 95% CI 0.94 to 0.99), did not live with stable male sexual partner during the lockdown (aOR=2.08, 95% CI 1.21 to 3.56), had moderate anxiety (aOR=2.37, 95% CI 1.20 to 4.69), had higher consumption of psychoactive drugs during the lockdown (aOR=2.33, 95% CI 1.01 to 5.40), and did not have HIV or STI test since the end of the lockdown (aOR=3.36, 95% CI 1.80 to 6.28, because they ‘didn’t know where to go’ or ‘preferred to wait’; aOR=3.30, 95% CI 1.99 to 5.48, for ‘other reason’) were more likely to have discontinued PrEP than those who were still taking PrEP (current PrEP users) after the lockdown. Reporting fewer male sexual partners in the previous 6 months (aOR=7.70, 95% CI 3.63 to 16.31, for 0–1 partners; aOR=2.24, 95% CI 1.37 to 3.66, for 2–9 partners) and not using any HIV prevention tools during the last anal intercourse (aOR=15.01, 95% CI 7.86 to 28.68) or exclusively a condom (aOR=7.33, 95% CI 3.40 to 15.81) were also significantly associated with discontinued PrEP after the lockdown.
Among 946 MSM respondents of the ERAS-COVID-19 study who were using PrEP before lockdown, 15.4% (n=146) discontinued PrEP after the lockdown. Factors significantly associated with discontinued PrEP after lockdown were having fewer sexual partners and having mental vulnerabilities compared with current PrEP users. Our results support other evidence that PrEP use is associated with sexual activity.9 11 22 It is likely that sexual activity decreased among MSM in France due to lockdown measures, and those taking PrEP decided to discontinue PrEP for this reason. Interestingly, discontinuation of PrEP among MSM in France was more likely an individual choice and not due to inaccessibility of PrEP services. Overall, this study shows that PrEP is an important prevention tool that can be adapted to an individual’s sexual activity, regardless of the health crisis. It is therefore essential that PrEP users (and providers) are well informed about how to safely stop/start PrEP to reduce the risk of HIV infection during transition periods.
Changes in sexual behaviour have been investigated in various COVID-19 studies, with the reduction of sexual acts often highlighted,10 11 23 and our results further support this phenomenon. In Australia, a survey among MSM PrEP users highlighted that reduction in sexual activity was the main reason for having stopped PrEP.24 Our study showed that, 3 months after the first lockdown, PrEP was restarted by a little more than three-quarters (76.8%) of participants who had stopped it during the lockdown. Although sexual activity was reduced among some MSM, only 24.0% of former PrEP users reported 0 or 1 sexual partner in the previous 6 months (vs 46.6% reporting 2–9 partners and 29.4% reporting more than 10) and 31.4% declared having sexual partners during the lockdown. Therefore HIV risk exposure may remain present for these MSM, mainly due to the level of HIV risk exposure within the MSM community.25 Moreover, our data showed that 33.8% of MSM who discontinued PrEP reported not using an HIV prevention method during the last sexual intercourse. PrEP stoppage has already been identified as a period of high risk of HIV infection before the COVID-19 crisis,26 27 and in the open-label extension of the ANRS-IPERGAY trial the only HIV-seroconversion occurred among MSM who discontinued PrEP.28
Although our results highlighted that a decrease in sexual behaviours associated with HIV exposure led to discontinuing PrEP, for some PrEP users the decision to discontinue may be associated with mental health issues. Former PrEP users were more likely to report anxiety and an increase in psychoactive drug use during the lockdown. Similar results in terms of stress, anxiety and isolation were also identified among MSM in London,11 internationally,8 and among people with pre-existing health, social or structural vulnerabilities.29 Therefore, the COVID-19 crisis and its effects on mental health may decrease the ability of vulnerable individuals and communities to practise self-care, to take an active role in protecting their own well-being and to maintain their health, including HIV prevention.
Several studies prior to COVID-19 have highlighted PrEP discontinuation among MSM with lower self-perception of HIV risk exposure, fewer sexual partners or mental health difficulties.27 Therefore, behavioural and psychosocial factors also play a role in continuing PrEP in a more general context and should be explored further. Additionally, we found that MSM respondents who discontinued PrEP were younger. Young age has already been identified as a risk factor for discontinued PrEP in previous studies,27 further supporting our hypothesis that discontinued PrEP observed in our study is similar to observations in other settings pre-COVID-19. HIV prevention programmes should therefore pay special attention to behavioural and psychosocial factors, specifically regarding mental health support, among MSM. PrEP discontinuation (definitively or temporarily) should be accompanied by personalised counselling on HIV prevention strategies and HIV exposure assessment to decrease the risk of infection after stopping PrEP. Moreover, PrEP use and HIV prevention should be integrated into a comprehensive healthcare package which includes mental healthcare and support.
Finally, access to PrEP services did not seem to be a barrier to PrEP use during the lockdown in France. Among the various reasons proposed for PrEP discontinuation during lockdown, only 9.5% of respondents reported discontinuing PrEP during lockdown due to access issues (no more medication and cancelled or postponed PrEP visit). However, PrEP access in France seems to contrast with other available data where MSM reported postponed PrEP appointments,23 difficulties accessing PrEP and testing10 and other sexual healthcare services when needed.11
This study has some limitations. Data were collected a few months after the first lockdown in France; retrospective bias cannot be excluded. We have no qualitative data to provide indepth information about the reason(s) for discontinuing PrEP during or after the lockdown. While our results suggest that MSM who discontinued PrEP during the lockdown did so because they wanted to, we lack information regarding the reason for discontinued PrEP after lockdown. Due to the cross-sectional nature of the study, we are unable to determine if PrEP discontinuation is a temporary or permanent decision. For this analysis, respondents who took PrEP before lockdown but who indicated not currently using PrEP at the time of the survey were considered ‘former PrEP users’ after lockdown. It is possible, however, that these respondents have discontinued PrEP temporarily. More longitudinal studies are needed to better understand the use of PrEP over time. Finally, MSM respondents in this study may not be representative of the MSM community in France. Subpopulations, such as migrant MSM or MSM with precarious situations for example, could be underestimated in our study.
In the months following the first national COVID-19 lockdown in France, discontinuation of PrEP occurred more often among MSM who were younger, had fewer sexual partners and had mental health vulnerabilities. This study contributes to other PrEP studies, within and outside of the COVID-19 context, highlighting the relationship between PrEP use and sexual activity. More research is needed regarding the use of PrEP over time, and particularly discontinuing PrEP, so that sexual health and prevention providers can better support and educate PrEP users on how to safely stop/start PrEP to reduce the risk of HIV infection during transition periods.
Decrease in sexual behaviours associated with HIV exposure, mental health vulnerabilities and younger age was associated with discontinued pre-exposure prophylaxis (PrEP) use after the lockdown.
Factors associated with discontinued PrEP after the lockdown were similar to those identified before the COVID-19 crisis in other studies.
Discontinued PrEP after lockdown in France was not associated with problems in access to health services, in contrast to other countries.
When PrEP is discontinued (temporarily or definitively), it is necessary to focus on comprehensive HIV prevention services which include mental health services and support.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Patient consent for publication
This study involves human participants but there is no ethics committee. Participants gave informed consent to participate in the study before taking part.
The authors would like to thank Nathalie Lydié, Nicolas Etien (Santé Publique France), Bérangère Gall and Julien Vivant (BVA Institute) for their involvement in the implementation of the survey, our associative partners for their support and relay of the survey in their network, and all gay men and other men who have sex with men for taking the time to complete the survey.
Handling editor Jamie Scott Frankis
Contributors AV designed and supervised the data collection for ERAS-COVID-19 study and supervised the analysis. MDC and VV conducted the analysis and wrote the paper under the supervision and advice of DRC, RMD, SM and DM. AV is responsible for the overall content.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.