Recruitment, experience, and retention among women with HIV and hazardous drinking enrolled in the WHAT-IF? Study

Background: Despite efforts by the NIH to enhance the participation of women and minorities in clinical research, women with or without HIV continue to remain underrepresented. The purpose of this study is to better understand the reasons why women with HIV and hazardous drinking participated in the WHAT-IF? Study and to discuss their experience (positive or negative) in the study. Methods: Convenience and theoretical sampling were used to recruit women with HIV and hazardous drinking to complete qualitative interviews. These women had previously completed a clinical alcohol intervention trial and had consented to be contacted in the future for study-related purposes. The biopsychosocial model was used to frame the interview questions that assessed multiple determinants of drinking behavior and helped explain linkages to broader health constructs. Results: A total of 20 women with HIV and hazardous drinking completed the qualitative interview. Several factors were identified by the women as influential in their decision to participate in the WHAT-IF? study, such as the ability to quit or reduce their drinking to nonhazardous levels (biological), the ability to gain knowledge or a greater understanding of the negative effects of hazardous drinking on HIV disease progression (psychological), and peer pressure and monetary compensation (social). Also, the women identified factors (positive or negative) associated with their clinical trial experience, such as the effects of the study medication on the woman's body (biological), thoughts and feelings toward study procedures (i.e. medication, lab work, study assessments) and the length of the study (psychological), and the interactions with the WHAT-IF? Study staff (social). Conclusion: Recruiting and retaining women with or without HIV in clinical research remains a challenge. Findings from this study suggest that women with HIV who are hazardous drinkers may benefit from participating in research studies that could help them to reduce or quit their drinking, increase their knowledge about specific behavior changes, and earn monetary compensation. Also, positive staff interactions may be instrumental in retaining minority women in clinical research.

. In substance use research, researchers have failed to adequately recruit women, and frequently made the mistake of generalizing results to women without analyzing sex-specific differences ( Dawson, 1996;Weisner, 1993). Likewise, women with HIV may demonstrate certain attitudes toward treatment-seeking practices, such as perceiving a need for substance use treatment, having less knowledge about the benefits of treatment options, or negative beliefs about treatment (Hu et al., 2016). On the contrary, well-documented barriers to treatment-seeking practices among women include pregnancy, lack of services for pregnant women, and preferring one form of treatment of intervention over the other (Grosso et  Moreover, little is known about the clinical trial experiences of women with or without HIV who participate in alcohol research. Our ability to understand these processes, from women's perspectives could enhance effective treatment approaches (Slade et al., 2010). Women participating in a qualitative study examining their experiences of managing depressive symptoms reported the personal qualities of healthcare professionals were associated with their willingness to discuss sensitive topics (Slade et al., 2010). Additional qualitative studies examining study retention reported the interpersonal skills of the healthcare professionals are instrumental in enrolling, retaining, and improving intervention efficacy (Kneipp et al., 2013;Slade et al., 2010;Beal et al., 2009).
The purpose of this study is to better understand the reasons why women with HIV and hazardous drinking participated in the WHAT-IF? Study and to discuss their experience (positive or negative) in the study. Conducting qualitative research (particularly in underserved, disadvantaged groups) could provide critical data on the perceptions, cultural relevancy, acceptability, and salience of specific aspects of intervention approaches of the randomized clinical trial that are acceptable and not acceptable to these populations (Creswell, 2017). The retention rates in the WHAT-IF? study was above satisfactory, with over 89% and 85% completing the 4-month and 7-month follow-ups respectively (Cook et al., 2019). This is notable and does suggest that participation in the study could have influenced study retention among study participants.

Study design
Qualitative interviews were conducted among women with hazardous drinking and HIV who participated in the WHAT-IF? (Will Having Alcohol Treatment Improve my Functioning) study. Detailed information about the study is published elsewhere (Cook et al., 2019;Cook et al., 2017). Briefly, the WHAT-IF study was a large, multi-site double-blind randomized clinical trial that evaluated pharmacotherapy for a reduction in drinking among women with HIV in Miami, FL. Detailed information about the study is published elsewhere. Briefly, the primary goal of the study was a reduction in hazardous drinking, defined as consuming <7 drinks per week or <3 drinks in one sitting.
Multiple strategies were employed to reduce barriers to recruitment among women into an alcohol intervention trial. The research team placed brochures in clinics and community settings and encouraged current participants to refer to others. The study required multiple in-person visits, and participants could earn up to (insert number) if they completed the study. In total, the study enrolled 196 women, of whom 83% were African American, 16% were married or in a long-term relationship, 10% were employed, and 43% had less than high school education. The participant's mean age was month follow-ups respectively.

Study population
Convenience and theoretical sampling were used to recruit women with HIV and hazardous drinking to complete the qualitative interviews. These women had previously completed the WHAT-IF study and had consented to be contacted in the future for study-related purposes. The women were recruited at the time of study completion or up to two years after completing the study. The primary goal of the qualitative study was to recruit a diverse range of women with HIV, who reportedly reduced, quit, or had no change in their drinking. While the women were chosen for their drinking status (high or low), the interviewer did not know their drinking status going into each interview.
However, toward the end of the qualitative study, the interviewer was made aware of four participants for whom the number of drinks per week was extremely high. The participant ID numbers were shared with the research coordinator in Miami who managed study recruitment and retention. A total of four to six participants were recruited every two months until theoretical saturation was achieved and the participants provided no new information in the interview.

Ethics, consent, and permissions
The WHAT-IF study and the qualitative study was approved by IRBs at the University of Florida (Gainesville, FL) and the University of Miami Miller School of Medicine (Miami, FL). Written informed consent was obtained at the Miami site. All participants were informed that the purpose of the qualitative study was to better understand why women with or without HIV choose to participate in clinical research.

Study instrument
The biopsychosocial model was used to frame the interview questions that assessed multiple determinants of drinking behavior and helped explain linkages to broader health constructs. The interview questions were developed after extensive discussions with the research study team as well as members of the Qualitative Research Colloquium (QRC) at the University of Florida. The QRC is comprised of experienced qualitative investigators and graduate students who provide support and oversight for qualitative research. The complete list of interview open-ended questions fell into five categories that are listed in Table 1. For this study, the category denoted by an asterisk (*) was analyzed.

Procedure
The qualitative interviews were conducted by the first author (S.C.) who received hands-on informed consent and qualitative research training by an experienced qualitative investigator. On the day of the scheduled interviews, participants were provided with written informed consent and were able to ask questions related to the study and seek clarification on matters they did not understand. The

Data analysis
This study is an analysis of data collected from qualitative interviews among women with HIV and hazardous drink who completed the WHAT-IF? Study. Each interview lasted approximately one hour and was transcribed by a University of Florida approved, Health Insurance and Portability and Accountability Act (HIPAA) compliant transcription service provider. The transcribed interviews were uploaded to NVivo 11.0, a qualitative data analysis program. The data was stored on an encrypted computer with password protection. Moreover, data analysis was continuous and began after the first interview was completed.
While this is not a grounded theory study, the findings are adapted analytical methods from grounded theory (e.g. iterative and simultaneous data collection and analysis, theoretical saturation). Also, thematic analysis was employed (Vaismoradi et al., 2013;Braun and Clarke, 2006). In this process, the following steps were involved: 1) immerse oneself in the data, 2) generate initial codes, 3) search for themes, 4) review the themes, 5) define and name the themes, and 6) produce the reports. In the beginning, the transcripts were read numerous times to immerse oneself in the data. Also, ideas, questions, and comments were recorded for future use. Next, the data was organized into categories At what age did you first start drinking?

2.
When you first started drinking, were you a light, moderate, or heavy drinker?

3.
Why did you decide to start drinking?

5.
What was it about alcohol that made you continue to drink?

6.
Have you tried to quit drinking?

a.
Tell me about your previous attempts to quit b. i.

Successful or Unsuccessful
ii.
Time period that you were able to quit or reduce your drinking iii.
Were there any barriers that stopped you from quitting or reducing drinking? 1.
How did you find out about the study?

1.
What was it about the study that motivated you to participate?

2.
Tell me about your participation in other research studies, if any? B.
What benefits do you get out of participating in research? C.
Tell me about your experience in this study 1.
What were some things that you like about the study?

2.
What were some things that you did not like about the study?

a.
What was difficult? D.
Describe your relationship with the study staff E. 1.
What did you like about the study staff?

2.
What did you not like about the study staff? F.
Changes as a result of participating in the WHAT-IF study G.
What changed in your life as a result of being in the study?

1.
How were you able to quit or reduce drinking? 2. a.

3.
Do you think that support is needed in order to successfully quit or reduce drinking?

a.
Why or why not?
What could be done differently in the study? B.
If you were going to give advice to women about quitting or reducing their drinking, what would you tell them? C.
Would you like to add anything else?

Results
A total of 20 women with HIV and hazardous drinking completed the qualitative interview. The demographic characteristics of the women in the qualitative study were similar to the women in the WHAT-IF? Study. In this study, 85% were African American, 80% were single, 100% of the women were unemployed and 60% had less than high school education. The mean age of women was 49.3 (Table 2). Several factors were identified by the women with HIV as influential in their decision to participate in the WHAT-IF? Study. The women also identified factors (positive or negative) associated with their clinical trial experience. These were categorized into three themes: biological, psychological, and social (Table 3).

Experience and retention in the WHAT-IF? Study Biological
Women in the WHAT-IF? study was randomly assigned to receive either 50mg of Naltrexone or placebo for four months. In the qualitative study, women discussed their experience with taking the study medication, although they were unaware of which study arm they were in. Many of the women discussed the effects of the study medication on their bodies, including any experienced adverse health effects. While enrolled in the WHAT-IF? study, the women were required to complete computer assessments.
During the assessments, women were prompted to answer questions about their past and current drinking. After the assessments, the women received feedback and information they could use to assist with making a behavioral change. Also, many of the women commented on the length of the WHAT-IF? Study which lasted for 7-months.
The majority of the women believed the study length was appropriate and convenient.
"Actually, it was a very convenient study so it wasn't taking me out of nothing else I was really doing.

Social
Many of the women reported the social interactions with the WHAT-IF? Study staff impacted their experience and influenced their retention in the study. The women discussed their ability to talk freely and socialize with study staff. Also, multiple women praised the study staff for their professionalism, compassion, and respect that was displayed throughout the study and follow-up, respectively. However, some women were not as positive about their study interactions. As noted in the comment, even though they described a negative interaction, they did not find the study staff at fault.

Discussion
Women with HIV and hazardous drinking were asked to participate in qualitative interviews to better understand their reasons for participating in the WHAT-IF? Study and to discuss their experience (positive or negative) in the study. In this study, the reasons for participation and their experience in the study were examined within the context of the biopsychosocial model.
For many women, participating in the study could help them quit or reduce their drinking to nonhazardous levels (biological). The women discussed psychological reasons for study participation, such as the ability to gain knowledge or a greater understanding of the negative effects of hazardous drinking on HIV disease progression and the opportunity for self-reflection by examining their lives and recognizing the areas for behavioral change. This finding was consistent with previous research examining reasons for clinical trial participation among women, such as the opportunity for selfreflection or self-empowerment (Green, 2006;Grella and Joshi, 1999). Women in both studies believed their lives were out of control or needed services to help them regain control of their lives (Green et al., 2006;Grella and Joshi, 1999

Conclusion
Recruiting and retaining women with or without HIV in clinical research remains a challenge.
Conducting qualitative research (particularly in underserved, disadvantaged groups) could provide critical data on the perceptions, cultural relevancy, acceptability, and salience of specific aspects of intervention approaches of the randomized clinical trial that are acceptable and not acceptable to

Consent for publication
Not applicable

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding