Game Changers for CC Prevention (GC-CCP).

GC-CCP consists of six sessions to empower CC screened women to act as change agents for CC screening in their social networks. Session 1 addresses fears and concerns related to CC risk and use of self-compassion and peer support to overcome these fears. Session 2 focuses on decision making for sharing one’s personal CC screening experience, and how to initiate disclosure and conversations about CC. Session 3 focuses on CC facts and myths, and helping participants address misconceptions about CC during discussions with others. Participants also think about the women in their social network and identify women who may be strategic targets for CC advocacy. Sessions 4 to 6 build CC advocacy skills such as communication skills (e.g., reflective listening, paraphrasing, open ended questions) and how to start and sustain conversations about CC. These sessions aim to inspire commitment to ongoing CC advocacy through peer solidarity and support. The 2-hour, weekly sessions will be conducted in Luganda using a structured manual, and group format to facilitate: sharing of experiences to build solidarity, support, and motivation among participants; group problem solving and role playing to build skills and self-efficacy; and homework between sessions to practice new skills and generate experiences to be processed in the sessions.

Facilitator training, supervision, and fidelity monitoring. Two bilingual (Luganda, English), literate female clients of the clinic will be recruited by the clinic staff and trained as intervention peer facilitators. Criteria for facilitators include having in-depth knowledge of the local community and personal experience of being screened for CC; they may be counselors or community health workers (trained lay persons) whose role often includes community outreach and education, and facilitating client engagement and use of health services. The training, conducted by the study’s senior investigators, will include reviewing objectives for each session, step-by-step scripts, role playing and mock implementation of core exercises. Training will cover group facilitation, building rapport, reflective listening, and dealing with group conflict. The supervisors of the initial two groups at each site (i.e., the groups administered to the research index participants) will attend each session and complete fidelity rating forms; each facilitator will also complete fidelity forms after each session. These forms will include ratings of fidelity to the manual, participant engagement, and quality of facilitation, and will be used to facilitate weekly supervision between sessions.

RCT evaluation of CC-CCP.

At each of the four sites, 40 women screened for CC in the past year will enroll as index participants (n = 160), with 20 randomly assigned to the intervention, and 20 to a wait-list usual care control. The intervention arm will be divided into two groups of 10 (balanced on ages 18–35 and 36+) for receipt of the intervention. The control arm receives the intervention after the 12-month follow-up assessments have been completed. There are no specific criteria for discontinuing or modifying allocated intervention condition for individual participants. A non-intervention control was used, as opposed to an attention control that would focus on the intervention participants, as this attention would not be comparable to the control for attention needed for effects on social network members (who provide data for the primary hypothesized outcomes). Participants in both groups receive care as usual from the study site, which does not involve any training related to advocacy. Each index participant will recruit up to three alters (maximum total = ~440 alters; referred to as first degree alters) with no history of CC screening. All index and first degree alter participants will be followed up at months 6 and 12. At month 6, first degree alters will recruit up to two of their own alters (second degree alters; total = ~800); second degree alters will be administered a single, brief phone interview. Fig 2 outlines the schedule of trial activities, including assessments and intervention sessions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. SPIRIT guideline checklist.

https://doi.org/10.1371/journal.pone.0317491.g002

Eligibility criteria. Index participants: female, age 18 and older, screened for CC in past year (verified by clinic records), and no advanced stage disease (12-month follow-up is feasible). First degree alters: female, age 18 and older, and report no CC screening history. Second degree alters: female and age 18 and older.

Recruitment. Clinic staff will contact eligible women who have been screened for CC in the past year to inform them of the study; if interested, the site coordinator will perform consent processes and formal eligibility screening. Eligible women who enroll will complete the baseline survey, followed by randomization as an index participant. First degree alters: The site coordinator will use the social network assessment within the baseline survey of each index to randomly select (using a random number table) five alters who know the participant’s CC screening experience (or as many as there are if < 5) and ask the index if she is comfortable recruiting three of these alters. The index will be asked to call each selected alter at the end of their baseline interview to describe the study in the presence of the coordinator (or later in private, if she prefers), who will then schedule a study visit for the alter. If an alter refuses or cannot be reached, a replacement will be randomly chosen from the original list of five alters. Second degree alters: The site coordinator will use the social network assessment in the month 6 survey of each first degree alter to randomly select four alters (or as many as there are if < 4) who the alter reports targeting with CC screening advocacy and thereafter was screened for CC. A similar process as described above for first degree alters will be used to recruit and enroll second degree alters. The coordinator will obtain verbal consent from the second degree alter over the phone and then conduct a brief phone survey.

The consent process will include information about how a version of the study data in which all identifying information is removed may be made available for use by other scientists for the purpose of further research. Given the nature of the intervention, there is no anticipated harm. Informed consent materials are available, on request, from the corresponding author. To promote study retention, we will collect tracking information (phone numbers, mapped addresses, contacts for family/friends with whom they have frequent contact).

Randomization. A blocked 1:1 randomization design with stratification by age (age 18-35/36+ years) and screening result (positive/negative) will be used with randomly alternating blocks of 2, 4, and 6 to prevent anticipation of assignment to ensure balance across arms. The study statistician will use a random number generator to devise a sequential randomization code list that will be used by the study coordinator to assign index participants to the intervention or control arms. The coordinator and data collectors are blind to the assignment until after the baseline assessments are completed; after the random assignment is revealed, the only blind party is the data analyst.

Assessment schedule. Surveys will be administered at baseline, month 6 and month 12 for index and first degree alters, while second degree alters will be surveyed only once (at month 6 assessment of first degree alters). Participants will receive 30,000 Ush (~$8) per assessment, except second degree alters who will receive 10,000 Ush (via mobile money). To limit attrition, we will collect tracking information, including contacts for family/friends with whom they have frequent contact. Clinic staff will help us track clients we have difficulty reaching.

Measures. Survey assessments for index and first degree alter participants will be in-person and interviewer-administered in Luganda using Network Canvas software. These interviews will last approximately 75 and 30 minutes, respectively and include a social network assessment. Second degree alters will receive one 15-minute phone interview. CC screening and treatment utilization will be verified with medical chart data. All measures will be assessed at each of three time points.

Primary outcome: CC screening: Participants will self-report utilization of all CC-related services including screening (VIA, HPV, pap smear), treatment (thermal or cryo-therapy; radiation, chemotherapy, surgery) and biopsies. Any reported utilization will be verified with abstraction of medical chart data.

Secondary outcome: engagement in CC prevention advocacy: This will be assessed with participant self-report, both in terms of frequency of general advocacy with respect to specific CC-prevention topics, and more specifically with CC screening advocacy targeted at alters named in their social network assessment (described below).

Other constructs that will be assessed include targets of the intervention that are potential mediators of the intervention effect. These include internalized CC stigma, sharing of CC experience, CC knowledge, and advocacy self-efficacy, all of which will be assessed with participant self-report using measures we developed and validated in our prior research [30–32].

Social network assessment. The participant will list 10 female alters with whom they interact most. For each listed alter, we will gather information to assess network composition (e.g., age, relation to respondent, history of CC screening and treatment; knowledge of respondent’s CC screening experience). In the index survey only, we will ask how if each alter knows each other alter to assess network structure (e.g., density); this section is time consuming, so given the large number of alter participants, alters will not complete this survey component. To assess CC advocacy and perceived effects, we will ask the participants if they have engaged in CC screening advocacy (separate items for discussed, encouraged, gave info, gave direct support, and frequency) with the specific alter in the past 6 months, and any perceived resulting action (e.g., alter was screened). At follow-up, we will determine whether listed alters are the same or unique from those listed at baseline, to facilitate longitudinal analyses of alter data. Enrolled alters of the index participant will be asked about advocacy they received from the index to gain their perspective on the characteristics and impact of the advocacy.

Data Analysis. Power: The power analysis was calculated based on the primary outcome. Assuming 220 first degree alters per arm enroll, 10% attrition, intraclass correlation (ICC) between .01 to .05 to control for clustering within the alters of each index participant, and a 15% CC screening rate in control alters (based on our pilot data): the effective sample size will be 200 (ICC = .05) or 216 (ICC = .01) per arm. If ICC = .01, we will be able to detect a small effect size (11 percentage point difference) in alter CC screening at month 12 (80% power; alpha = .05). If ICC = .05, the detectable effect size is 12 percentage point difference (or odds ratio of 2.0).

Assess effects of GC-CCP on alter CC screening: We will use an intent-to-treat approach. In addition to comparing the arms at baseline and months 6 and 12, we will apply logistic generalized mixed models to our repeated-measures data to examine intervention effects, using a time by arm interaction to assess differences between the arms over time. We will account for correlation among participants in the same intervention group, and among alters referred by the same index, by adjusting standard errors for statistical inference tests with a sandwich estimator. We will use imputation for item nonresponse and attrition weights to account for non-random dropouts using logistic regression. We will control for and examine interaction effects with index (e.g., age), alter (e.g., knowledge of index’s CC screening), and intervention (sessions completed) variables.

Evaluate the cost and cost-effectiveness of implementing GC-CCP to increase CC screening: We will compare GC-CCP vs. usual care on the marginal cost of increasing alter CC screening. Following standard convention, [49] we will define the incremental cost-effectiveness ratio (ICER) as the difference in per-capita cost of the intervention versus control group divided by the difference in their average effectiveness: (μ_c2 = mean per-capita cost of GC-CCP, μ_c1 = mean per-capita cost of usual care, δ_e2 = % of alters who get screened in GC-CCP, δ_e1 = % of alters who get screened in usual care. We will estimate confidence intervals with bootstrap methods [50]. We will examine economic costs from a societal perspective, incorporating estimates of the frequency and duration of time spent by individuals engaging in receipt of services, as well as transportation costs and forgone employment as relevant. We will use a micro-costing approach, time-driven activity based costing, to track all costs associated with implementing GC-CCP and usual care as estimated from data collected by the sites and study team [51]. The cost per resource will be calculated by multiplying the quantity used by unit cost; total cost will be derived by summating individual costs [52]. Capital costs will be annualized using a discount rate of 3% with an assumed lifespan of 30 (buildings) and 10 (furniture) years [53]. Facilitators will record time spent on the sessions, identifying each activity (e.g., training, preparation, in-session) and session-related materials (e.g., consumable materials). Unit costs for labor will be characterized as capacity cost rates. Intervention training and ongoing costs will be differentiated; ongoing costs will be tracked to determine cost efficiencies over time, and we will differentiate fixed from variable intervention costs to assess the marginal cost of providing care to additional clients.

Evaluation of ongoing GC-CCP implementation and maintenance.

After implementing GC-CCP with research enrolled index participants in the first 6 months of implementation to evaluate effectiveness, clinic staff will take over all aspects of implementation during the Sustainment phase (2.5 years) of implementation. The study team will transition the training and supervision of peer facilitators to clinic providers. This phase is important to (a) fully understand implementation processes, and facilitators of and barriers to real-world implementation of GC-CCP (without research staff); (b) to study sustainability of GC-CCP (as well as CC prevention advocacy and CC screening) when GC-CCP is fully implemented by clinic staff; and (c) to prepare the clinics to maintain GC-CPP into the future, if it is found to be cost-effective.

Identifying and training clinic staff to supervise GC-CPP. The study team will work with the clinics to identify suitable persons to serve as supervisors for the Sustainment phase. The selected staff member will likely vary from site to site based on variability in staff composition and skill sets, but female counselors or volunteer lay persons who assist in the clinic may be suitable, if they have experience supervising others. Nurses are not preferrable in this role given that they are already overburdened in these busy clinics. The peer facilitators will be identified and trained at the start of the Implementation phase for the RCT, but there is likely to be some turnover and need for training of new facilitators during the 3-year implementation period. The person who will train and supervise the peer facilitators will be identified at the start of the RCT; this will enable them to participate in the initial facilitator training, and to shadow the supervisors from the study team as they carry out supervision during the RCT. In the initial supervision sessions conducted by the staff member in the Sustainment phase, one of the study team supervisors will attend these sessions to provide continued mentorship. Supervision will be held after each session for the first group they oversee, followed by after every other session.

RE-AIM outcomes will be assessed to evaluate GC-CCP implementation, as operationalized in Table 2. Effectiveness for increasing CC screening (primary RCT outcome) will be based on self-report and medical chart abstracted data from alter participants enrolled in the RCT. Reach in terms of how many women engage in GC-CCP and CC prevention advocacy will be assessed using chart data from CC screened clients, and survey data from enrolled index and alter participants, respectively. At the beginning of the RCT (Implementation Phase), we will work with the clinic to add a field to the client visit form to record referral source for CC screening [e.g., healthcare provider, peer (family, friend or other client)]; and if relevant, RCT enrollment (and date) and participation in GC-CCP (dates of sessions attended). Follow-up survey data from enrolled index and alter participants in the RCT will inform Maintenance in terms of sustained engagement in CC prevention advocacy. Semi-structured interviews and surveys with clinic staff and clients, supervisor ratings of facilitators (fidelity), and clinic budget ledgers (cost), will be used to assess EPIS constructs [33, 34].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. RE-AIM outcomes.

https://doi.org/10.1371/journal.pone.0317491.t002

Semi-structured interviews with clinic staff. We will assess implementation qualitatively with semi-structured interviews with clinic staff who are involved in CC screening and GC-CCP (e.g., clinic manager, GC-CCP peer facilitator, CC screener/treater). We will interview 20 staff (5 per clinic) at the beginning of Sustainment, followed by 15 and 30 months later (ideally, all with the same staff), to complement the quantitative outcomes in helping to understand the implementation process. Staff will be assured of confidentiality and that their responses will not be shared with others at the workplace. We will use the interview guide developed for the Preparation phase, revising questions as needed to ask about current practices and adding probes about specific barriers that emerged in prior interviews, to assess changes in barriers over time. We also will ask what resources are needed to sustain CC screening and GC-CCP.

Client focus groups. Using chart data on GC-CCP participation and CC screening referrals, a random sample of 10 clients per site who participate in GC-CCP, and a random sample of 10 CC screened alter clients per site who were referred by a peer, will be asked to participate in one of two focus groups per site. Selected women will be stratified by age (age 18–35 vs. 36+). Client participants will be referred by clinic staff to the study team for consent processes. Examples of questions to be asked of GC-CCP participants include: How easy or challenging was it for you to engage with the training sessions? How useful was the content of the training? How influential was the training on your engagement in advocacy with other women? Examples of questions for the alters include: How much, if at all, did the advocacy you receive address the questions and/or concerns you had about CC screening? How influential was the advocacy on your decision to get screened? How do you feel about advocating to other women?

Survey data (clinic staff). To complement the qualitative data on implementation and sustainability factors, we will collect brief survey data from clinic staff once at the beginning and once at end of the 30-month Sustainment phase. These data will include adapted survey items [53] from three companion measures focused on the fit of GC-CCP with the clinic operations (e.g., “I welcome GC-CCP into the clinic,” “GC-CCP is a good fit with clinic practices,” and “It is possible to offer GC-CCP,” from 1 = completely disagree to 5 = completely agree). The questions will be tailored to the Ugandan clinic contexts during the Preparation phase of the project. In addition, to assess adoption, clinic staff will be given a quarterly anonymous brief survey to assess if they helped to implement GC-CCP, and if so, the extent to which they performed key activities (e.g., identification and recruitment of CC screened clients to participate in GC-CCP; training or supervision of facilitators). A secure anonymous survey link will be emailed to providers, or they will be given a hard copy to complete and insert into an envelope. We will convene clinic providers and leadership quarterly to assess implementation challenges (re. CC screening and GC-CCP) that are raised in the surveys or during the meeting, and how to address them.

Data analysis. The qualitative data will be analyzed using methods similar to the Preparation phase. A longitudinal qualitative analysis will compare themes within and across timepoints to determine change in barriers over time (e.g., using summative analytic memos about themes at each timepoint) [54]. We will triangulate data collected from medical charts, semi-structured interviews and brief clinic surveys on CC screening uptake using a mixed methods approach [55].