Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Participant Retention in a Randomized Clinical Trial in an Outbreak Setting: Lessons From the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE)

Participant Retention in a Randomized Clinical Trial in an Outbreak Setting: Lessons From the... Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Participant Retention in a Randomized Clinical Trial in an Outbreak Setting: Lessons From the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) Rosalind J Carter, Reynold G B Senesi, Peter Dawson, Ibrahim Gassama, S A S Kargbo, Carey R Petrie, Mohamed Hashim Rogers, Mohamed Samai, Elizabeth T Luman Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 The Journal of Infectious Diseases SUPPLEMENT ARTICLE Participant Retention in a Randomized Clinical Trial in an Outbreak Setting: Lessons From the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) 1 2 3 2 4 3 Rosalind J. Carter, Reynold G. B. Senesi, Peter Dawson, Ibrahim Gassama, S. A. S. Kargbo, Carey R. Petrie, 2 2 1 Mohamed Hashim Rogers, Mohamed Samai, and Elizabeth T. Luman 1 2 3 Centers for Disease Control and Prevention, Atlanta, Georgia; College of Medicine and Allied Health Sciences, University of Sierra Leone, Freetown, Sierra Leone; The Emmes Corporation, Washington, DC; Ministry of Health and Sanitation, Freetown, Sierra Leone XX e S Th ierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE), a phase 2/3 trial of an investigational Ebola vaccine, was implemented in April 2015 in Sierra Leone. Healthcare and Ebola frontline workers were randomized to immediate (within 7 days) XXXX or deferred (18–24 weeks) vaccination and observed from enrollment to 6 months postvaccination for safety and Ebola virus disease. er Th e were concerns that high retention and compliance would be difficult to achieve, particularly for the deferred group. Trial sta ff conducted monthly calls to participants and home visits if needed. Retention was defined as completion of the final assessment at 6 months postenrollment and postvaccination. Full compliance was defined as completion of all monthly assessments before and 6 months aer vaccin ft ation and vaccination per protocol. Logistic regression was used to identify demographic characteristics asso - ciated with these outcomes. Of 8626 participants enrolled, 7979 (92.5%) were retained postenrollment (95.2% in immediate vaccina- tion arm, 89.8% in deferred arm). Overall, 68.8% were fully compliant postenrollment (73.4% in immediate arm, 64.2% in deferred arm). Among 7979 vaccinated participants, 95.9% were retained 6 months postvaccination, with no significant difference between study arms. Frontline workers and younger participants were least likely to be retained and had a lower likelihood of full compliance. High retention of participants in a vaccine clinical trial in a low-resource setting, even among those assigned to deferred vaccina- tion, was achievable. Younger participants and frontline workers may require additional follow-up strategies and resources. Clinical Trials Registration. ClinicalTrials.gov [NCT02378753] and Pan African Clinical Trials Registry [PACTR201502001037220]. Keywords. Ebola; Ebola vaccine; retention; losses to follow-up; randomized clinical trial. In randomized clinical trials, poor participant retention can participant follow-up. First, there was a lack of clinical trial introduce bias and reduce study power, ae ff cting the general - experience and research infrastructure in Sierra Leone. Second, izability, validity, and reliability of results [1]. Factors associ- available resources focused on immediate outbreak response, ated with retention can include the following: study complexity; and it was critical that those resources not be diverted for the GOVERNMENT stigma associated with the disease being studied; not wanting sake of the trial. Third, STRIVE trial participants were health - to disclose trial participation; cost to participate, such as trans- care and frontline Ebola response workers who worked long portation costs and time off from work; health of participants; hours and provided essential clinical or response-related ser- educational level; and occupation [2]. The Sierra Leone Trial vices at the height of the outbreak, and many were displaced to Introduce a Vaccine Against Ebola (STRIVE) is a phase 2/3 due to job loss when the outbreak was waning. In this paper, unblinded, randomized trial of an investigational Ebola vac- we describe the methods used to reduce losses to follow-up and cine, rVSVΔG-ZEBOV-GP (Merck), to evaluate the safety and examine factors associated with retention and compliance with efficacy of the vaccine and was conducted in Sierra Leone. The the study protocol. This trial was registered under ClinicalTrials. study enrolled more than 8000 healthcare and frontline Ebola gov (NCT02378753) and Pan African Clinical Trials Registry workers as study participants [3, 4]. Conducting a clinical trial (PACTR201502001037220). in Sierra Leone during an Ebola outbreak posed challenges to METHODS Description of Sierra Leone Trial to Introduce a Vaccine Against Ebola Participants were enrolled at 7 STRIVE trial sites in 5 districts Received 14 December 2017; editorial decision 14 February 2018; accepted 23 February 2018. highly aeff cted by Ebola virus disease (Western Area Urban Presented in part: IDWeek, October 26–30, 2016; Abstract 749; New Orleans, LA. [Freetown], Western Area Rural, Bombali, Tonkolili, and 3 sites Correspondence: R.  J. Carter, PhD, CDC 1600 Clifton Road, MS A04, Atlanta, GA 30329 (RDC6@cdc.gov). in Port Loko) from April 9 through August 15, 2015 [5]. To be The Journal of Infectious Diseases 2018;217(S1):S65–74 eligible for the trial, participants had to be at least 18 years of age Published by Oxford University Press for the Infectious Diseases Society of America 2018. and be actively working as a healthcare provider (in a clinical or This work is written by (a) US Government employee(s) and is in the public domain in the US. DOI: 10.1093/infdis/jiy094 nonclinical role) or frontline Ebola worker (eg, contact tracer, Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S65 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 ambulance crew, burial team). Participants had to assert that Control and Prevention Institutional Review Board, the they were planning to reside in Sierra Leone for 6 months aer ft Pharmacy Board of Sierra Leone, and the US Food and Drug enrollment and agree to be contacted monthly by telephone. Administration. All participants gave written informed consent. Detailed contact information was collected during enrollment, Participant Follow-Up including the following: participant address with important Participants were followed for 6  months postvaccination. nearby landmarks; participant personal cell phone number; and Several strategies were used to maximize participant retention name and cell phone number of an alternative contact, typically and to ensure that participants would be reachable (Table  1). a close relative or friend who would know how to reach the par- Each participant who enrolled was given a new cell phone and ticipant if investigators had difficulty reaching them on their a SIM card that provided free access to a “closed-user-group.” study or personal telephone. This allowed free calls for the study staff to contact partici - At enrollment, participants were individually randomized to pants each month and for participants to call the trial hotline. immediate (<7  days) or deferred (18–24 weeks) vaccination. This hotline was available for 24 hours a day and 7 days a week Starting at approximately 17 weeks aer enr ft ollment, partici - throughout the trial for procedural questions and referral for pants in the deferred group were contacted to return to the trial medical issues. site where they were rescreened for eligibility, and those still eligible were vaccinated. Once deferred participants were vac- Monthly Calls cinated, they were referred to as “crossover vaccinated.” Each month trained study staff conducted a structured tele - Community sensitization to build awareness and support for phone interview with participants that included questions the trial was initiated 4 months before the trial’s launch, followed to identify new or ongoing health events, including Ebola by targeted information sessions conducted at local healthcare virus disease and newly recognized pregnancy for women. facilities, specialized Ebola facilities, and District Ebola Response Participants who reported a new or ongoing health issue were Centers (Ebola coordination centers) to recruit participants. referred to the study nurse for medical care as needed. Two call Senior staff from the Ministry of Health and Sanitation and attempts were made each day (morning and evening) for 3 days the College of Medicine and Allied Health Sciences conducted until the participant was contacted, with staff having discretion approximately 50 stakeholder sensitization meetings with com- of continuing attempts for up to 7 days to maximize the chance munity and government leaders at the national, district, and local of reaching participants. Home visits were conducted if the par- levels [6]. es Th e activities forged important relationships that ticipant was not reachable by telephone aer a ft t least 6 attempts. supported community outreach throughout the trial. Details of At every call or visit, locator and contact information was veri- STRIVE trial procedures are described elsewhere [3, 5]. fied and updated if necessary. e p Th rotocol was approved by the Sierra Leone Ethics and If the participant could not be contacted within 14 days aer ft Scientific Review Committee, the US Centers for Disease the targeted monthly assessment date, that month’s assessment Table 1. STRIVE Trial Strategies to Promote Participant Retention Category Description of Strategy Considerations and Notes Equipment Cell phone/SIM card provided at enrollment • At time of st udy planning, unclear whether most adults would own a cell phone. Communications Closed-user-group • Participant could make free calls to trial hotline and study staff to report changes in health and request medical triage. • Free calls facilitated contact between participants and staff be- tween scheduled monthly calls. • Study staff could make free calls to participants’ study cell phone in addition to personal telephone (multiple contacts). Community engagement Discussions with community leadership • Ongoing community engagement to identify and address rumors. Leadership at local healthcare and Ebola facilities kept in- • Engagement of hospital leadership important for keeping partic- formed throughout the trial ipants engaged in study over time including returning for cross- over vaccination. Flexible appointments Study staff worked in 2 shifts, which allowed calls and home • Because participants were working at time of enrollment, they had visits to participants outside of working hours as needed to schedule appointments taking into account their work hours. Multiple locator If participant could not be reached by telephone after 6 • Multiple methods to track participants including study, personal information attempts, study staff attempted to reach them through rel- and a relative's cell phone number, home address. ative telephone number or a home visit Staff training and Ongoing staff training to improve communication skills, con- • Staff assigned to specific areas for tracking participants at home. feedback fidentiality during home visits. Progress on percentage of Teams provided feedback on success of making all telephone persons due for follow-up reached discussed at weekly calls; troubleshoot problems together. staff meetings Abbreviations: STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola. S66 • JID 2018:217 (Suppl 1) • Carter et al Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 was recorded as not completed. Study staff attempted to con - including those who were never vaccinated. For postvaccination tact the participant for the subsequent monthly calls using the retention, we compared vaccinated participants from the imme- same procedures described above. For the final assessment diate group to vaccinated participants in the deferred group (ie, call, 6 months postvaccination (or postenrollment if not vacci- crossover participants). For the immediate vaccination group, the nated), contact was attempted for up to 28 days aer t ft he target postenrollment and postvaccination time periods are the same, date before terminating the participant from the trial. and the final assessment occurred 6  months aer enr ft ollment/ Telephone calls and home visits were conducted by approxi- vaccination. For the deferred vaccination group, the final posten - mately 100 recent nursing and pharmacy school graduates who rollment assessment occurred 18–24 weeks aer ft enrollment (the were based at the 3 follow-up coordination and data management assessment immediately before vaccination or at 6 months if not centers (follow-up centers): Western Area Urban (for participants vaccinated), and the final postvaccination assessment occurred from Western Area Urban and Western Area Rural districts), 6 months aer cr ft ossover vaccination. Bombali (for participants from Bombali and Tonkolili districts), Outcome: Full Compliance and Port Loko (for participants from the 3 sites in Port Loko We also evaluated full compliance with the study protocol for District). Aer t ft he main protocol training with all study sta,ff postenrollment and postvaccination activities. Participants in refresher training sessions were conducted at individual follow-up the immediate vaccination group who were vaccinated within sites to cover protocol updates, interpersonal communication 7  days of enrollment and completed all 6  monthly follow-up skills, and maintain confidentiality while tracking participants interviews were defined as being in full compliance postenroll - in their communities. Retention rates were regularly reviewed by ment and postvaccination. Deferred participants were consid- the STRIVE Data Safety and Monitoring Board and presented at ered in full compliance postenrollment if they were vaccinated weekly staff meetings at the 3 participant follow-up centers, where per protocol (approximately  18–24 weeks aer enr ft ollment) staff discussed strategies for contacting hard-to-reach participants. and completed every monthly assessment from enrollment to Duration of Follow-Up vaccination. Deferred participants who were never vaccinated Participants in the immediate vaccination group were vaccinated but completed every monthly interview through month 6 aer ft within 7  days of enrollment and called monthly for 6  months enrollment were also considered fully compliant postenroll- postvaccination (Figure 1). Deferred vaccination group partici- ment. Full compliance postvaccination in crossover participants pants were observed from enrollment until they were vaccinated were those who were vaccinated per protocol and completed all and then for 6 months postvaccination (Figure  1). Participants 6 monthly interviews aer vaccin ft ation (regardless of interview in both groups who were never vaccinated (eg, refused vaccina- completion between enrollment and vaccination). tion, or for the deferred group, ineligible at the time of vaccina- Statistical Methods tion) were observed for 6 months postenrollment. Participants were analyzed as randomized to either immediate or deferred groups. Participants who died during follow-up Data Analysis Outcome: Retention (n = 25) are excluded from these analyses of retention (although We defined retention as completion of the final monthly assess - they are included in the reports of the trial results [3]). ment, and we measured it at 2 time points: postenrollment We calculated the proportion of participants retained and and postvaccination (Figure  1). For postenrollment retention, in full compliance, and we identified factors associated with we compared results for the immediate and deferred groups, retention and full compliance using univariate and multivariate Vaccinated Immediate Group Post-vaccination follow-up N = 4311 Unvaccinated Enrollment and Immediate vaccination Deferred Group a Post-enrollment follow-up Post-vaccination follow-up N=4315 Enrollment Cross-over vaccination  0123456789 10 11 12 Months Includes 12 participants randomized to the deferred group who were vaccinated in error with the immediate group Figure 1. Postenrollment and postvaccination follow-up of the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) participants. Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S67 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 stepwise logistic regression models adjusted for randomization participants were vaccinated during the crossover period, and assignment (immediate vs deferred), age, sex, education, occu- 12 (<1%) of the deferred participants were vaccinated immedi- pation, facility type, and enrollment site. Retention models also ately in error (Figure 2). These last 12 participants were included included vaccination status. The Western Area Rural site was in all retention and full compliance analyses as vaccinated and selected as the referent group for site comparisons because it deferred randomization arm participants. For analyses of full had the highest enrollment of the nonurban sites. A  stepwise compliance, they were considered noncompliant because they model selection approach was used to determine the best subset were not vaccinated per protocol. of predictors. e Th criterion used to enter the model was P = .10, Most participants were male (60.6%) and the median age was with P = .01 as the criterion to stay in the model once additional 31 years (range, 18–79 years) (Table 2). The majority of partici - predictors were added. Odds ratios (ORs) and 95% confidence pants were nurses (33.3%) or frontline workers (48.2%); 44.2% intervals (CIs) are presented for each level of predictor com- completed secondary school, and 41.6% had tertiary education. pared with its reference groups. Education was related to occupation, with all (100%) doctors and a very high proportion of pharmacists (87%) and nurses RESULTS (76%) having tertiary education. Frontline workers came from diverse educational backgrounds: the majority (59%) completed Of 8626 participants included in the analysis, 4311 were secondary school, whereas 25% had a primary school education randomized to immediate vaccination and 4315 to deferred or less. This diversity reflects the variety of technical roles front - vaccination. Overall, 7979 (92.5%) of 8626 randomized line workers performed during the Ebola response, including participants were vaccinated; 4157 (96.4%) of the immedi- contact tracing, ambulance driving, cleaning Ebola facilities, ate participants were vaccinated, 3810 (88.9%) of the deferred Ineligible: 127 Assessed for Eligibility: 8815 Declined Participation: 15 Excluded from Retention Study: 47 Randomized: 8673 Invalid Consent: 22 Died: 25 Retained Post-Enrollment: 7979 (93%) Retention Study Participants: 8626 Fully Compliant Post-Enrollment: 5934 (69%) Retained Post-Enrollment: 3875 (90%) Retained Post-Enrollment: 4104 (95%) Randomized Immediate: 4311 Randomized Deferred: 4315 Fully Compliant Post-Enrollment: 2769 (64%) Fully Compliant Post-Enrollment: 3165 (73%) Crossover Vaccinated: 3810 (89%) Vaccinated: 4157 (96%) Retained Post-Vaccination: 3628 (95%) Retained Post-Vaccination: 4012 (97%) Fully Compliant Post-Vaccination: 2455 (64%) Fully Compliant Post-Vaccination: 3165 (73%) Immediately Vaccinated (in Error) : 12 (<1%) Not Vaccinated (Declined): 154 (4%) Not vaccinated: 493 (11%) Declined: 352 No Longer Eligible: 100 Lost to Follow-up: 32 Other: 9 All Vaccinated: 7979 Retained Post-Vaccination: 7649 (96%) Fully Compliant Post-Vaccination: 5620 (70%) Figure 2. Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) participant retention and compliance CONSORT diagram. Included in the vaccinated group; not eligible for crossover vaccination. S68 • JID 2018:217 (Suppl 1) • Carter et al Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Table 2. Retention, Compliance, and Participant Characteristics by Randomization Arm and Vaccination Status: STRIVE (n = 8626) Immediate Vaccination Deferred Vaccination Crossover All Vaccinated All Randomized (N = 4311) (N = 4315) (N = 3810) (N = 7979) (N = 8626) Study Variable n (%) n (%) n (%) n (%) n (%) c d d Retained 4104 (95.2) 3875 (89.8) 3628 (95.2) 7649 (95.9) 7979 (92.5) Fully Compliant 3165 (73.4) 2769 (64.2) 2455 (64.4) 5620 (70.4) 5934 (68.8) Male Gender 2611 (60.6) 2617 (60.6) 2454 (64.4) 5032 (63.1) 5228 (60.6) Age (Tertiled) (years) 18–27 1423 (33.0) 1452 (33.7) 1283 (33.7) 2658 (33.3) 2875 (33.3) 28–35 1470 (34.1) 1406 (32.6) 1213 (31.8) 2619 (32.8) 2876 (33.3) >35 1418 (32.9) 1457 (33.8) 1314 (34.5) 2702 (33.9) 2875 (33.3) Primary Occupation Nurse 1443 (33.5) 1428 (33.1) 988 (25.9) 2336 (29.3) 2871 (33.3) Doctor 13 (0.3) 11 (0.3) 7 (0.2) 20 (0.3) 24 (0.3) Pharmacist 20 (0.5) 19 (0.4) 19 (0.5) 39 (0.5) 39 (0.5) Allied health professional 77 (1.8) 72 (1.7) 58 (1.5) 130 (1.6) 149 (1.7) Community health worker 85 (2.0) 93 (2.2) 68 (1.8) 153 (1.9) 178 (2.1) Laboratory worker 133 (3.1) 139 (3.2) 139 (3.6) 269 (3.4) 272 (3.2) Frontline worker 2065 (47.9) 2089 (48.4) 1340 (35.2) 3377 (42.3) 4154 (48.2) Surveillance worker 246 (5.7) 250 (5.8) 170 (4.5) 410 (5.1) 496 (5.8) Other/Not reported 229 (5.3) 214 (5.0) 128 (3.4) 352 (4.4) 443 (5.1) Facility Type Clinic setting 796 (18.5) 794 (18.4) 837 (22.0) 1626 (20.4) 1590 (18.4) Ebola Facility 1491 (34.6) 1493 (34.6) 686 (18.0) 2119 (26.6) 2984 (34.6) Hospital 1722 (39.9) 1711 (39.7) 1365 (35.8) 3020 (37.8) 3433 (39.8) Other/Not reported 302 (7.0) 317 (7.3) 922 (24.2) 1214 (15.2) 619 (7.2) Education Tertiary 1810 (42.0) 1778 (41.2) 1500 (39.4) 3209 (40.2) 3588 (41.6) Secondary 1874 (43.5) 1937 (44.9) 1773 (46.5) 3614 (45.3) 3811 (44.2) Primary 233 (5.4) 216 (5.0) 191 (5.0) 420 (5.3) 449 (5.2) None 374 (8.7) 363 (8.4) 338 (8.9) 709 (8.9) 737 (8.5) Other/Not reported 20 (0.5) 21 (0.5) 8 (0.2) 27 (0.3) 41 (0.5) Enrollment Site District (Town) Western Area Rural 941 (21.8) 949 (22.0) 849 (22.3) 1746 (21.9) 1890 (21.9) Western Area Urban 1624 (37.7) 1620 (37.5) 1370 (36.0) 2915 (36.5) 3244 (37.6) Bombali 599 (13.9) 603 (14.0) 544 (14.3) 1137 (14.2) 1202 (13.9) Tonkolili 366 (8.5) 365 (8.5) 338 (8.9) 703 (8.8) 731 (8.5) Port Loko (Port Loko Town) 433 (10.0) 429 (9.9) 398 (10.4) 823 (10.3) 862 (10.0) Port Loko (Lunsar) 197 (4.6) 199 (4.6) 172 (4.5) 365 (4.6) 396 (4.6) Port Loko (Kaffu Bullom) 151 (3.5) 150 (3.5) 139 (3.6) 290 (3.6) 301 (3.5) Abbreviation: STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola.  There were no significant differences in demographic characteristics between the participants randomized to the immediate and deferred vaccination arms. Participants randomized to deferred vaccination arm who were vaccinated 18–24 weeks after enrollment. Participants randomized to immediate vaccination who were retained 6 months after enrollment. Participants who were retained 6 months after vaccination. Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student, and vaccinator. Includes dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabber (persons taking postmortem skin/mucosal swabs for Ebola testing). burial duties, and social mobilizing. There were no significant participants were retained 6 months postvaccination: 96.5% of differences in demographic characteristics between immediate immediate vaccinated participants and 95.2% of crossover vac- and deferred randomized groups. cinated participants. In the analysis of full protocol compliance, 5934 (68.8%) of the 8626 participants were in full compliance Retention and Full Compliance With Protocol postenrollment: 73.4% of participants in the immediate vaccina- Overall, 7979 (92.5%) of the 8626 participants randomized tion group and 64.2% of those assigned to deferred vaccination. were retained 6  months postenrollment: 95.2% of participants Of the vaccinated participants, 70.4% were in full compliance assigned to immediate vaccination and 89.8% of those assigned postvaccination: 73.4% of the immediate vaccinated and 64.4% to deferred vaccination (Table 2). Almost all (95.9%) vaccinated of the crossover vaccinated participants. Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S69 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Factors Associated With Retention nurses the second highest (99%), and doctors the lowest (90%). In multivariate analysis, 6-month postenrollment retention was Compared with nurses, doctors (aOR = 0.1; 95% CI, 0.02–0.5), higher among participants >27 years (adjusted OR [aOR] = 1.3, frontline workers (aOR  =  0.3; 95% CI, 0.2–0.4), community 95% CI, 1.1–1.6 for those 28–35; and aOR = 1.9, 95% CI, 1.6–2.4 health workers (aOR = 0.3; 95% CI, 0.1–0.8), and those no lon- for those >35) and those who were vaccinated (aOR: 3.5; 95% ger employed at the time of crossover vaccination (aOR = 0.3; CI: 2.9–4.3) (Table 3). Significantly lower retention was associ - 95% CI, 0.2–0.4) had significantly lower retention. ated with (1) frontline workers (aOR = 0.6; 95% CI, 0.5–0.7) and Factors Associated With Full Protocol Compliance surveillance workers (aOR  =  0.6; 95% CI, 0.4–0.9) compared Factors associated with full compliance postenrollment and with nurses and (2) enrollment at the Port Loko Town site com- postvaccination were similar. Randomization to immediate pared with the Western Area Rural site (aOR  =  0.5; 95% CI, vaccination arm, age >27  years, sites in Bombali and Tonkoli 0.4–0.7). Districts, and occupation as a laboratory worker were all associ- Retention 6  months postvaccination was associated with ated with a higher likelihood of full protocol compliance during older age (aOR  =  1.5, 95% CI, 1.2–2.0 for those 28–35; and both postenrollment and postvaccination periods in multivar- aOR  =  2.2, 95% CI, 1.7–3.0 for those >35) and enrollment at iate analyses (Table  5). Bombali and Port Loko (Lunsar, Kaffu sites in Bombali (aOR  =  2.2; 95% CI, 1.3–3.6) and Tonkoli Bullom) District sites generally had higher rates of full compli- (aOR = 3.0; 95% CI, 1.5–6.3) Districts compared with Western ance compared with Western Area Rural. Frontline and com- Area Rural in adjusted analyses (Table  4). Among occupa- munity health workers and those working at Ebola facilities tional groups, pharmacists had the highest retention (100%), Table 3. Factors Associated With Retention 6 Months Postenrollment Among Randomized Participants in the STRIVE Trial (n = 8626) Retained Univariable Multivariable Factor-Level Factor-Level N (%) Odds Ratio (95% CI) P Value Odds Ratio (95% CI) P Value Randomization Arm Deferred Vaccination 3875 (89.8) ref <.0001 Immediate Vaccination 4104 (95.2) 2.3 (1.9–2.7) Vaccination Status Unvaccinated 3958 (88.8) ref <.0001 ref <.0001 Vaccinated 4021 (96.4) 3.4 (2.8–4.1) 3.5 (2.9–4.3) Gender Male 4792 (91.7) ref .0003 Female 3187 (93.8) 1.4 (1.2–1.6) Age (tertiled) (years) 18–27 2587 (90.0) ref <.0001 ref <.0001 28–35 2668 (92.8) 1.4 (1.2–1.7) 1.3 (1.1–1.6) >35 2724 (94.7) 2.0 (1.6–2.5) 1.9 (1.6–2.4) Occupation Nurse 2723 (94.8) ref <.0001 ref .0002 Doctor 22 (91.7) 0.6 (0.1–2.6) 0.5 (0.1–2.3) Pharmacist 37 (94.9) 1.0 (0.2–4.2) 0.9 (0.2–3.9) Allied health professional 143 (96.0) 1.3 (0.6–3.0) 1.2 (0.5–2.7) Community health worker 163 (91.6) 0.6 (0.3–1.0) 0.7 (0.4–1.2) Laboratory worker 260 (95.6) 1.2 (0.7–2.2) 1.1 (0.6–2.1) Frontline worker 3777 (90.9) 0.5 (0.5–0.7) 0.6 (0.5–0.7) Surveillance worker 446 (89.9) 0.5 (0.4–0.7) 0.6 (0.4–0.9) Other/Not reported 408 (92.1) 0.6 (0.4–0.9) 0.6 (0.4–0.9) Facility Type Clinic setting 1468 (92.3) ref .0276 Ebola Facility 2744 (92.0) 1.0 (0.8–1.2) Hospital 3207 (93.4) 1.2 (0.9–1.5) Other/Not reported 560 (90.5) 0.8 (0.6–1.1) Enrollment Site Western Area Rural 1741 (92.1) ref <.0001 ref <.0001 Western Area Urban 3061 (94.4) 1.4 (1.1–1.8) 1.2 (0.9–1.5) Bombali 1117 (92.9) 1.1 (0.9–1.5) 0.9 (0.7–1.3) Tonkolili 670 (91.7) 0.9 (0.7–1.3) 0.7 (0.5–1.0) Port Loko (Port Loko Town) 747 (86.7) 0.6 (0.4–0.7) 0.5 (0.4–0.7) Port Loko (Lunsar) 360 (90.9) 0.9 (0.6–1.3) 0.7 (0.5–1.0) Port Loko (Kaffu Bullom) 283 (94.0) 1.4 (0.8–2.2) 1.1 (0.7–1.8) Abbreviations: CI, confidence interval; ref, reference category; STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola.  Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student and vaccinator. Includes dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabbers (persons taking post mortem skin/mucosal swabs for Ebola testing). S70 • JID 2018:217 (Suppl 1) • Carter et al Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Table 4. Factors Associated With Postvaccination Retention Among Vaccinated Participants, STRIVE (n = 7979) Retained Univariable Multivariable Factor-Level Factor-Level N (%) Odds Ratio (95% CI) P Value Odds Ratio (95% CI) P Value Randomization Arm Crossover Vaccination 3628 (95.2) ref .0039 Immediate Vaccination 4012 (96.5) 1.4 (1.1–1.7) Gender Male 4748 (94.6) ref <.0001 ref .0056 Female 2892 (98.1) 3.0 (2.3–4.0) 1.6 (1.2–2.3) Age (tertiled) (years) 18–27 2491 (93.8) ref <.0001 ref <.0001 >27–35 2539 (96.3) 1.7 (1.3–2.2) 1.5 (1.2–1.9) >35 2610 (97.6) 2.6 (2.0–3.5) 2.2 (1.7–3.0) Occupation Nurse 2312 (99.0) ref <.0001 ref <.0001 Doctor 18 (90.0) 0.1 (0.02–0.4) 0.1 (0.02–0.5) Pharmacist 39 (100) >99.9 (<0.01 to >99.9) >99.9 (<0.01 to >99.9) Allied health professional 124 (95.4) 0.2 (0.1–0.5) 0.2 (0.1–0.5) Community health worker 145 (96.0) 0.2 (0.1–0.6) 0.3 (0.1–0.7) Laboratory worker 261 (97.0) 0.3 (0.1–0.7) 0.5 (0.2–1.1) Frontline worker 3176 (94.3) 0.2 (0.1–0.3) 0.2 (0.2–0.4) Surveillance worker 393 (95.9) 0.2 (0.1–0.4) 0.4 (0.2–0.8) Not currently working 836 (93.6) 0.2 (0.1–0.2) 0.2 (0.1–0.4) Other/Not reported 336 (95.7) 0.2 (0.1–0.4) 0.3 (0.2–0.6) Facility Type Clinic setting 1536 (94.6) ref .0002 Ebola Facility 2037 (96.2) 1.4 (1.1–2.0) Hospital 2921 (96.9) 1.8 (1.3–2.4) Other/Not reported 1146 (94.6) 1.0 (0.7–1.4) Enrollment Site Western Rural 1655 (95.0) ref <.0001 ref <.0001 Western Urban 2783 (95.7) 1.2 (0.9–1.6) 0.8 (0.6–1.1) Bombali 1117 (98.2) 3.0 (1.8–4.9) 2.2 (1.3–3.6) Tonkolili 695 (98.9) 4.6 (2.2–9.6) 3.0 (1.5–6.3) Port Loko (Port Loko Town) 768 (93.4) 0.8 (0.5–1.1) 0.7 (0.5–1.0) Port Loko (Lunsar) 345 (94.4) 1.0 (0.6–1.6) 0.6 (0.3–1.0) Port Loko (Kaffu Bullom) 277 (95.5) 1.1 (0.6–2.1) 0.9 (0.5–1.6) Abbreviations: CI, confidence interval; ref, reference category; STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola.  Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student, and vaccinator. Includes pharmacist, dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabbers (persons taking post mortem skin/mucosal swabs for Ebola testing). Current occupation information was updated for deferred crossover participants during the crossover period. Many were no longer working as the Ebola outbreak subsided, and Ebola response jobs ended. Crossover participants were offered vaccination regardless of work status. had lower likelihood of completing all 6 assessments compared closed and response activities slowed towards the end of the with nurses during both postenrollment and postvaccination outbreak was a sizeable challenge to follow-up. Nonetheless, the follow-up. strategies used in the STRIVE trial, including collecting exten- sive locating information, using closed-user-group cell phone DISCUSSION technology and hotlines to facilitate communication, and hir- e S Th TRIVE trial achieved high participant retention through ing adequate staff who were familiar with the participant pop - 6  months of follow-up despite the challenges of implement- ulation and local areas resulted in overall high retention and ing a vaccine trial during an ongoing Ebola outbreak in a should be considered for future trials in low-resource and other resource-limited setting. The trial, which enrolled and vac - emergency settings. cinated almost 8000 participants, retained more than 92.5% Evaluating and reporting retention in randomized clinical of participants 6  months postenrollment and 97.4% postvac- trials is important because high rates of loss to follow-up (in cination. Participation in every monthly interview was less ranges >20%) can compromise the validity of the trial results complete, with 68.8% and 70.4% achieving full compliance in [7], particularly if the retention rate is different between the postenrollment and postvaccination interviews, respectively. study arms. We observed somewhat lower retention among par- We observed lower retention among Ebola frontline workers. ticipants randomized to the deferred vaccination group com- es Th e workers generally held temporary employment posi - pared with the immediate arm (89.8% vs 95.2%, respectively) in tions during the outbreak, and displacement as Ebola facilities the first 6 months postenrollment. The unblinded design of the Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S71 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 S72 • JID 2018:217 (Suppl 1) • Carter et al Table 5. Predictors of Full Compliance Among Randomized Participants Postenrollment and Postvaccination 6 Months Postenrollment (All Participants, n = 8626) 6 Months Postvaccination (All Vaccinated, n = 7979) Compliant Univariable Multivariable Compliant Univariable Multivariable Odds Ratio Factor- Level Factor- Level Odds Ratio Factor Level Odds Ratio Factor- Level N (%) (95% CI) P Value N (%) P Value N (%) (95% CI) P Value (95% CI) P Value Randomization Arm Deferred Vaccination 2769 (64.2) ref <.0001 ref <.0001 2455 (64.4) ref <.0001 ref <.0001 Immediate Vaccination 3165 (73.4) 1.5 (1.4–1.7) 1.6 (1.4–1.7) 3165 (76.1) 1.8 (1.6–1.9) 1.9 (1.7–2.1) Enrollment Site Western Rural 1189 (62.9) ref <.0001 ref <.0001 1118 (64.1) ref <.0001 ref <.0001 Western Urban 2176 (67.1) 1.2 (1.1–1.4) 0.9 (0.8–1.1) 2021 (69.5) 1.3 (1.1–1.4) 1.1 (0.9–1.2) Bombali 989 (82.3) 2.7 (2.3–3.3) 2.1 (1.8–2.6) 861 (75.7) 1.7 (1.5–2.1) 1.5 (1.2–1.8) Tonkolili 523 (71.5) 1.5 (1.2–1.8) 1.2 (1.0–1.4) 595 (84.6) 3.1 (2.5–3.9) 2.6 (2.1–3.3) Port Loko (Port Loko) 523 (60.7) 0.9 (0.8–1.1) 0.8 (0.6–0.9) 508 (61.8) 0.9 (0.8–1.1) 0.8 (0.7–1.0) Port Loko (Lunsar) 296 (74.7) 1.8 (1.4–2.2) 1.4 (1.1–1.9) 284 (78.0) 2.0 (1.5–2.6) 1.7 (1.3–2.3) Port Loko (Kaffu Bullom) 238 (79.1) 2.2 (1.7–3.0) 1.7 (1.3–2.3) 233 (80.3) 2.3 (1.7–3.1) 2.0 (1.4–2.7) Age (tertiled) years 18.0–27.5 1820 (63.3) ref <.0001 ref <.0001 1747 (65.8) ref <.0001 ref <.0001 27.5–35.4 2017 (70.1) 1.4 (1.2–1.5) 1.3 (1.2–1.5) 1896 (71.9) 1.3 (1.2–1.5) 1.3 (1.1–1.4) 35.4–79.5 2097 (72.9) 1.6 (1.4–1.8) 1.4 (1.3–1.6) 1977 (73.9) 1.5 (1.3–1.7) 1.3 (1.2–1.5) Gender Male 3568 (68.2) ref .1761 3466 (69.0) ref .0001 Female 2366 (69.6) 1.1 (1.0–1.2) 2154 (73.1) 1.2 (1.1–1.4) Occupation Nurse 2071 (72.1) ref <.0001 ref <.0001 1769 (75.8) ref <.0001 ref <.0001 Doctor 19 (79.2) 1.5 (0.6–3.9) 1.3 (0.5–3.4) 16 (80) 1.3 (0.4–3.8) 1.0 (0.3–3.0) Pharmacist 25 (64.1) 0.7 (0.4–1.3) 0.6 (0.3–1.2) 25 (64.1) 0.6 (0.3–1.1) 0.6 (0.3–1.1) Allied Health Professional 101 (67.8) 0.8 (0.6–1.2) 0.8 (0.5–1.1) 93 (71.5) 0.8 (0.5–1.2) 0.8 (0.5–1.2) Community health worker 110 (61.8) 0.6 (0.5–0.9) 0.7 (0.5–0.9) 97 (64.2) 0.6 (0.4–0.8) 0.6 (0.4–0.8) Laboratory worker 218 (80.1) 1.6 (1.2–2.1) 1.6 (1.2–2.2) 216 (80.3) 1.3 (0.9–1.8) 1.5 (1.1–2.0) Frontline worker 2716 (65.4) 0.7 (0.7–0.8) 0.8 (0.8–0.9) 2281 (67.7) 0.7 (0.6–0.8) 0.8 (0.7–0.9) Surveillance worker 361 (72.8) 1.0 (0.8–1.3) 1.2 (1.0–1.5) 312 (76.1) 1.0 (0.8–1.3) 1.2 (0.9–1.6) Other/Not reported 313 (70.7) 0.9 (0.8–1.2) 1.0 (0.8–1.2) 256 (72.9) 0.9 (0.7–1.1) 0.9 (0.7–1.1) Not currently working NA NA NA 555 (62.2) 0.5 (0.5–0.6) 0.9 (0.6–1.2) Facility Type Clinic setting 1140 (71.7) ref <.0001 ref <.0001 1154 (71.1) ref <.0001 ref <.0001 Ebola Facility 1855 (62.2) 0.7 (0.6–0.7) 0.8 (0.7–0.9) 1393 (65.8) 0.8 (0.7–0.9) 0.8 (0.7–0.9) Hospital 2524 (73.5) 1.1 (1.0–1.3) 1.2 (1.0–1.3) 2279 (75.6) 1.3 (1.1–1.4) 1.1 (0.9–1.3) Other/Not reported 415 (67) 0.8 (0.7–1.0) 1.0 (0.8–1.2) 794 (65.5) 0.8 (0.7–0.9) 1.0 (0.7–1.3) Vaccination Status Unvaccinated 2769 (62.1) ref <.0001 Vaccinated 3165 (75.9) 1.9 (1.8–2.1) Abbreviations: CI, confidence interval; ref, reference category; NA, not applicable. Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student, and vaccinator. Includes pharmacist, dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabbers (persons taking post mortem skin/mucosal swabs for Ebola testing). Current occupation information was updated for deferred crossover participants during the crossover period. Many were no longer working as the Ebola outbreak subsided, and Ebola response jobs ended. Crossover participants were offered vaccination regardless of work status. Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 STRIVE trial may have contributed to higher postenrollment outside of Freetown oen h ft ad poor cell phone coverage, and retention in the immediate vaccination arm because trial sta ff there were periods when cellular service was down nationwide. may have prioritized contacting vaccinated participants—espe- Furthermore, because not all parts of the country are on the cially early in the study—because of concerns about possible national electric grid, participants frequently lacked access to vaccine-associated adverse events. However, we believe the electricity to charge their study cell phone. Home visits were modest difference in retention across trial arms is unlikely to oen t ft ime-, labor-, and resource-intensive due to poor quality ae ff ct the validity of our trial results [ 3]. of roads and limited transportation options, especially in the Participants vaccinated immediately aer ft enrollment and districts. Approximately 1  day per week, trial staff had use of those vaccinated aer a de ft lay of 18–24 weeks had similar high a study car to make field visits, but even then they oen h ft ad rates of retention, 96.5% and 95.2%, respectively, 6 months post- to use motorbikes from the main road to the villages where vaccination, suggesting that use of a delayed intervention arm participants lived due to lack of paved roads. Street names and is a feasible option for randomized clinical trials in emergency house numbers were frequently missing in both rural areas and settings where a placebo control arm would not be feasible. The densely populated urban areas, making it difficult to physically assurance that the intervention would be oer ff ed to everyone locate participants. Because some trial staff were students or enrolled may help to explain our relatively high retention in the recent graduates from urban Freetown who were assigned to deferred arm postenrollment. Clinical trials in the United States rural district locations, they were not always familiar with the have shown that participant retention tends to decline among people and location of the villages within the districts. Although those randomized to a placebo arm when the intervention is not we noted higher retention at the rural Tonkolili and semiurban oer ff ed to all at the end of the trial [ 8]. Bombali District sites, neither road infrastructure nor differ - Retention rates in the STRIVE trial are comparable to those ences in participant population explain why these sites were observed in other Ebola vaccine clinical trials conducted during more successful than others. the 2014–2016 Ebola outbreak in West Africa, despite differ - The ongoing Ebola outbreak also created unique barriers ent study designs, lengths of follow-up, and participant popu- to follow-up. Ebola response activities appropriately took lations. The World Health Organization “ Ebola Ca Suffit ” ring precedence over study activities, so there were times when vaccination trial [9] in Guinea, which enrolled village residents study procedures had to be modified or even suspended. For who were contacts or contacts of contacts of a case, reported example, there were several times during the trial when study 94% participant retention, although follow-up was <3 months. staff could not conduct home visits because the village was In Liberia, the PREVAIL trial [10] achieved 97.7% retention of under quarantine due to an Ebola case. In one district, sta ff 1500 participant volunteers residing in Monrovia at 6  months at a large hospital were quarantined for 21  days after being and 98.3% at 12  months. es Th e 3 West African Ebola vac - exposed to an Ebola case. Many of these staff members were cine trials all used similar approaches to follow-up: obtaining STRIVE trial participants who consequently could not be detailed contact information, providing a study cell phone or contacted for that month’s assessment if they were not reach- trial hotline number, conducting home visits to track default- able by phone. ers, and employing dedicated staff to conduct follow-up. These Perhaps the biggest challenge to follow-up was displace- results validate the effectiveness of these approaches. ment of participants. Participant characteristics we identified e S Th TRIVE trial invested substantial time and expertise in as associated with higher risk of becoming lost to follow-up— ongoing community engagement with the Ebola and health specifically, younger age and employment as a frontline Ebola facilities and with political and religious leaders in the districts worker—were likely strongly related to displacement. Beginning where the trial was implemented. In nonemergency settings, in November 2015, Ebola facilities began to close and response large pediatric trials in Gambia [11, 12] have similarly achieved activities started winding down. Healthcare and frontline high follow-up of >90% using such methods, including com- Ebola workers, many of whom had moved to the district cen- munity engagement and using dedicated follow-up staff to stay ters to work on the response, returned to their homes, oen in ft in contact with participants through telephone calls and home remote villages, making telephone follow-up more difficult and visits if they missed a follow-up appointment. Other charac- home visits virtually impossible. Many young nurse graduates teristics of the STRIVE trial, such as development of trusting enrolled in the STRIVE trial while awaiting their government relationships between study staff and participants, rigorous posting, a universal first step aer co ft mpleting training. During follow-up efforts, and provision of ongoing staff training, have follow-up, some received postings to health facilities in remote been shown to improve retention in various vaccine trials in the districts, far from where they had enrolled in the STRIVE trial, United States and Africa [13–15]. where cell phone service was unreliable and home visits unfea- er Th e were challenges to conducting follow-up in a sible, posing challenges to retention. Finally, some participants resource-limited setting with a large cohort of participants reported that there was less motivation to complete the trial as spread over an expansive geographical area. Rural districts Ebola receded from the public mind. Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S73 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Potential conifl cts of interests. P.  D.  and C.  R. P.  report grants from CONCLUSIONS the Biomedical Advanced Research and Development Authority during e S Th TRIVE trial maintained high retention of participants, the conduct of the study. All authors have submitted the ICJME Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors con - demonstrating the feasibility of a study design with both sider relevant to the content of the manuscript have been disclosed. a deferred vaccination arm and a long follow-up period. Furthermore, conducting safety assessment follow-up using References telephone interviews rather than in-person visits facilitated 1. Schulz KF, Grimes DA. Sample size slippages in randomised trials: exclusions and enrollment of a large number of participants across multi- the lost and wayward. Lancet 2002; 359:781–5. 2. Robinson KA, Dennison CR, Wayman DM, Pronovost PJ, Needham DM. ple Ebola-ae ff cted districts and resulted in high retention and Systematic review identifies number of strategies important for retaining study completion rates. participants. J Clin Epidemiol 2007; 60:757–65. 3. Samai M, Seward JF, Goldstein ST, et al. The Sierra Leone trail to introduce a vac - Key lessons learned include the value of the following: (1) cine against Ebola: an evaluation of rVSVΔG-ZEBOV-GP vaccine tolerability collecting extensive locating information, including telephone and safety during the West Africa Ebola outbreak. J Infect Dis 2018; 217:s6–15. 4. Widdowson MA, Schrag SJ, Carter RJ, et  al. Implementing an Ebola vaccine numbers of close relatives and friends who would know how study—Sierra Leone. MMWR Suppl 2016; 65:98–106. to reach participants; (2) flexibility to conduct home outreach 5. Carter RJ, Idriss A, Widdowson MA, et al. Implementing a multi-site clinical trial in the midst of an Ebola outbreak: lessons learned from the Sierra Leone trial to for defaulter tracking; (3) use of closed-user-group cell phone introduce a vaccine against Ebola. J Infect Dis 2018; 217:s16–23. technology and hotlines to facilitate communication to report 6. Callis A, Carter VM, Albert AP, et  al. Lessons learned in clinical trial commu- illness or a change in contact information; and (4) employing an nication during an Ebola outbreak: the implementation of STRIVE. J Infect Dis 2018; 217:s40–7. adequate number of staff who are familiar with the participant 7. Fewtrell MS, Kennedy K, Singhal A, et al. How much loss to follow-up is accepta- population and local areas. Future trials in outbreak settings can ble in long-term randomised trials and prospective studies? Arch Dis Child 2008; 93:458–61. consider using these strategies to oer in ff terventions to persons 8. McCarthy-Keith D, Nurudeen S, Armstrong A, Levens E, Nieman LK. residing in wide geographical areas not tethered to a single clin- Recruitment and retention of women for clinical leiomyoma trials. Contemp Clin Trials 2010; 31:44–8. ical site. 9. Henao-Restrepo AM, Camacho A, Longini IM, et al. Efficacy and effectiveness of an rVSV-vectored vaccine in preventing Ebola virus disease: final results from the Guinea ring vaccination, open-label, cluster-randomised trial (Ebola Ça Suffit!). Notes Lancet 2017; 389:505–18. Acknowledgments. We express our gratitude to the trial participants 10. Kennedy SB, Bolay F, Kieh M, et al. Phase 2 placebo-controlled trial of two vac- and thanks to the dedicated Sierra Leone Trial to Introduce a Vaccine cines to prevent Ebola in Liberia. N Engl J Med 2017; 377:1438–47. Against Ebola (STRIVE) staff from COMHAS, University of Sierra Leone, 11. Cutts FT, Zaman SM, Enwere G, et  al. Efficacy of nine-valent pneumococcal the Sierra Leone Ministry of Health and Sanitation, the Centers for Disease conjugate vaccine against pneumonia and invasive pneumococcal disease in Control and Prevention (CDC) Sierra Leone Country Office, and the CDC e Ga Th mbia: randomised, double-blind, placebo-controlled trial. Lancet 2005; 365:1139–46. STRIVE teams in Atlanta and Sierra Leone for their tireless work. 12. Idoko OT, Owolabi OA, Odutola AA, et al. Lessons in participant retention in the Disclaimer. The findings and conclusions in this report are those of course of a randomized controlled clinical trial. BMC Res Notes 2014; 7:706. the author(s) and do not necessarily represent the official position of the 13. Johnson RE, Williams RD, Nagy MC, Fouad MN. Retention of under-served Centers for Disease Control and Prevention. women in clinical trials: a focus group study. Ethn Dis 2003; 13:268–78. Financial suppoort. e t Th rial was funded by the U.S. Centers for 14. Kennedy SB, Neaton JD, Lane HC, et al. Implementation of an Ebola virus disease Disease Control and Prevention, the Biomedical Advanced Research and vaccine clinical trial during the Ebola epidemic in Liberia: design, procedures, Development Authority, and the National Institutes of Health, with addi- and challenges. Clin Trials 2016; 13:49–56. tional support from the CDC Foundation. 15. Larson GS, Baseler BR, Hoover ML, et  al. Conventional wisdom versus actual Supplement sponsorship. This work is part of a supplement sponsored outcomes: challenges in the conduct of an Ebola vaccine trial in Liberia during the international public health emergency. Am J Trop Med Hyg 2017; 97:10–5. by the Centers for Disease Control and Prevention. S74 • JID 2018:217 (Suppl 1) • Carter et al http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Infectious Diseases Oxford University Press

Participant Retention in a Randomized Clinical Trial in an Outbreak Setting: Lessons From the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE)

Loading next page...
 
/lp/ou_press/participant-retention-in-a-randomized-clinical-trial-in-an-outbreak-RkI9RH6eTs

References (15)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Infectious Diseases Society of America
ISSN
0022-1899
eISSN
1537-6613
DOI
10.1093/infdis/jiy094
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Participant Retention in a Randomized Clinical Trial in an Outbreak Setting: Lessons From the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) Rosalind J Carter, Reynold G B Senesi, Peter Dawson, Ibrahim Gassama, S A S Kargbo, Carey R Petrie, Mohamed Hashim Rogers, Mohamed Samai, Elizabeth T Luman Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 The Journal of Infectious Diseases SUPPLEMENT ARTICLE Participant Retention in a Randomized Clinical Trial in an Outbreak Setting: Lessons From the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) 1 2 3 2 4 3 Rosalind J. Carter, Reynold G. B. Senesi, Peter Dawson, Ibrahim Gassama, S. A. S. Kargbo, Carey R. Petrie, 2 2 1 Mohamed Hashim Rogers, Mohamed Samai, and Elizabeth T. Luman 1 2 3 Centers for Disease Control and Prevention, Atlanta, Georgia; College of Medicine and Allied Health Sciences, University of Sierra Leone, Freetown, Sierra Leone; The Emmes Corporation, Washington, DC; Ministry of Health and Sanitation, Freetown, Sierra Leone XX e S Th ierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE), a phase 2/3 trial of an investigational Ebola vaccine, was implemented in April 2015 in Sierra Leone. Healthcare and Ebola frontline workers were randomized to immediate (within 7 days) XXXX or deferred (18–24 weeks) vaccination and observed from enrollment to 6 months postvaccination for safety and Ebola virus disease. er Th e were concerns that high retention and compliance would be difficult to achieve, particularly for the deferred group. Trial sta ff conducted monthly calls to participants and home visits if needed. Retention was defined as completion of the final assessment at 6 months postenrollment and postvaccination. Full compliance was defined as completion of all monthly assessments before and 6 months aer vaccin ft ation and vaccination per protocol. Logistic regression was used to identify demographic characteristics asso - ciated with these outcomes. Of 8626 participants enrolled, 7979 (92.5%) were retained postenrollment (95.2% in immediate vaccina- tion arm, 89.8% in deferred arm). Overall, 68.8% were fully compliant postenrollment (73.4% in immediate arm, 64.2% in deferred arm). Among 7979 vaccinated participants, 95.9% were retained 6 months postvaccination, with no significant difference between study arms. Frontline workers and younger participants were least likely to be retained and had a lower likelihood of full compliance. High retention of participants in a vaccine clinical trial in a low-resource setting, even among those assigned to deferred vaccina- tion, was achievable. Younger participants and frontline workers may require additional follow-up strategies and resources. Clinical Trials Registration. ClinicalTrials.gov [NCT02378753] and Pan African Clinical Trials Registry [PACTR201502001037220]. Keywords. Ebola; Ebola vaccine; retention; losses to follow-up; randomized clinical trial. In randomized clinical trials, poor participant retention can participant follow-up. First, there was a lack of clinical trial introduce bias and reduce study power, ae ff cting the general - experience and research infrastructure in Sierra Leone. Second, izability, validity, and reliability of results [1]. Factors associ- available resources focused on immediate outbreak response, ated with retention can include the following: study complexity; and it was critical that those resources not be diverted for the GOVERNMENT stigma associated with the disease being studied; not wanting sake of the trial. Third, STRIVE trial participants were health - to disclose trial participation; cost to participate, such as trans- care and frontline Ebola response workers who worked long portation costs and time off from work; health of participants; hours and provided essential clinical or response-related ser- educational level; and occupation [2]. The Sierra Leone Trial vices at the height of the outbreak, and many were displaced to Introduce a Vaccine Against Ebola (STRIVE) is a phase 2/3 due to job loss when the outbreak was waning. In this paper, unblinded, randomized trial of an investigational Ebola vac- we describe the methods used to reduce losses to follow-up and cine, rVSVΔG-ZEBOV-GP (Merck), to evaluate the safety and examine factors associated with retention and compliance with efficacy of the vaccine and was conducted in Sierra Leone. The the study protocol. This trial was registered under ClinicalTrials. study enrolled more than 8000 healthcare and frontline Ebola gov (NCT02378753) and Pan African Clinical Trials Registry workers as study participants [3, 4]. Conducting a clinical trial (PACTR201502001037220). in Sierra Leone during an Ebola outbreak posed challenges to METHODS Description of Sierra Leone Trial to Introduce a Vaccine Against Ebola Participants were enrolled at 7 STRIVE trial sites in 5 districts Received 14 December 2017; editorial decision 14 February 2018; accepted 23 February 2018. highly aeff cted by Ebola virus disease (Western Area Urban Presented in part: IDWeek, October 26–30, 2016; Abstract 749; New Orleans, LA. [Freetown], Western Area Rural, Bombali, Tonkolili, and 3 sites Correspondence: R.  J. Carter, PhD, CDC 1600 Clifton Road, MS A04, Atlanta, GA 30329 (RDC6@cdc.gov). in Port Loko) from April 9 through August 15, 2015 [5]. To be The Journal of Infectious Diseases 2018;217(S1):S65–74 eligible for the trial, participants had to be at least 18 years of age Published by Oxford University Press for the Infectious Diseases Society of America 2018. and be actively working as a healthcare provider (in a clinical or This work is written by (a) US Government employee(s) and is in the public domain in the US. DOI: 10.1093/infdis/jiy094 nonclinical role) or frontline Ebola worker (eg, contact tracer, Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S65 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 ambulance crew, burial team). Participants had to assert that Control and Prevention Institutional Review Board, the they were planning to reside in Sierra Leone for 6 months aer ft Pharmacy Board of Sierra Leone, and the US Food and Drug enrollment and agree to be contacted monthly by telephone. Administration. All participants gave written informed consent. Detailed contact information was collected during enrollment, Participant Follow-Up including the following: participant address with important Participants were followed for 6  months postvaccination. nearby landmarks; participant personal cell phone number; and Several strategies were used to maximize participant retention name and cell phone number of an alternative contact, typically and to ensure that participants would be reachable (Table  1). a close relative or friend who would know how to reach the par- Each participant who enrolled was given a new cell phone and ticipant if investigators had difficulty reaching them on their a SIM card that provided free access to a “closed-user-group.” study or personal telephone. This allowed free calls for the study staff to contact partici - At enrollment, participants were individually randomized to pants each month and for participants to call the trial hotline. immediate (<7  days) or deferred (18–24 weeks) vaccination. This hotline was available for 24 hours a day and 7 days a week Starting at approximately 17 weeks aer enr ft ollment, partici - throughout the trial for procedural questions and referral for pants in the deferred group were contacted to return to the trial medical issues. site where they were rescreened for eligibility, and those still eligible were vaccinated. Once deferred participants were vac- Monthly Calls cinated, they were referred to as “crossover vaccinated.” Each month trained study staff conducted a structured tele - Community sensitization to build awareness and support for phone interview with participants that included questions the trial was initiated 4 months before the trial’s launch, followed to identify new or ongoing health events, including Ebola by targeted information sessions conducted at local healthcare virus disease and newly recognized pregnancy for women. facilities, specialized Ebola facilities, and District Ebola Response Participants who reported a new or ongoing health issue were Centers (Ebola coordination centers) to recruit participants. referred to the study nurse for medical care as needed. Two call Senior staff from the Ministry of Health and Sanitation and attempts were made each day (morning and evening) for 3 days the College of Medicine and Allied Health Sciences conducted until the participant was contacted, with staff having discretion approximately 50 stakeholder sensitization meetings with com- of continuing attempts for up to 7 days to maximize the chance munity and government leaders at the national, district, and local of reaching participants. Home visits were conducted if the par- levels [6]. es Th e activities forged important relationships that ticipant was not reachable by telephone aer a ft t least 6 attempts. supported community outreach throughout the trial. Details of At every call or visit, locator and contact information was veri- STRIVE trial procedures are described elsewhere [3, 5]. fied and updated if necessary. e p Th rotocol was approved by the Sierra Leone Ethics and If the participant could not be contacted within 14 days aer ft Scientific Review Committee, the US Centers for Disease the targeted monthly assessment date, that month’s assessment Table 1. STRIVE Trial Strategies to Promote Participant Retention Category Description of Strategy Considerations and Notes Equipment Cell phone/SIM card provided at enrollment • At time of st udy planning, unclear whether most adults would own a cell phone. Communications Closed-user-group • Participant could make free calls to trial hotline and study staff to report changes in health and request medical triage. • Free calls facilitated contact between participants and staff be- tween scheduled monthly calls. • Study staff could make free calls to participants’ study cell phone in addition to personal telephone (multiple contacts). Community engagement Discussions with community leadership • Ongoing community engagement to identify and address rumors. Leadership at local healthcare and Ebola facilities kept in- • Engagement of hospital leadership important for keeping partic- formed throughout the trial ipants engaged in study over time including returning for cross- over vaccination. Flexible appointments Study staff worked in 2 shifts, which allowed calls and home • Because participants were working at time of enrollment, they had visits to participants outside of working hours as needed to schedule appointments taking into account their work hours. Multiple locator If participant could not be reached by telephone after 6 • Multiple methods to track participants including study, personal information attempts, study staff attempted to reach them through rel- and a relative's cell phone number, home address. ative telephone number or a home visit Staff training and Ongoing staff training to improve communication skills, con- • Staff assigned to specific areas for tracking participants at home. feedback fidentiality during home visits. Progress on percentage of Teams provided feedback on success of making all telephone persons due for follow-up reached discussed at weekly calls; troubleshoot problems together. staff meetings Abbreviations: STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola. S66 • JID 2018:217 (Suppl 1) • Carter et al Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 was recorded as not completed. Study staff attempted to con - including those who were never vaccinated. For postvaccination tact the participant for the subsequent monthly calls using the retention, we compared vaccinated participants from the imme- same procedures described above. For the final assessment diate group to vaccinated participants in the deferred group (ie, call, 6 months postvaccination (or postenrollment if not vacci- crossover participants). For the immediate vaccination group, the nated), contact was attempted for up to 28 days aer t ft he target postenrollment and postvaccination time periods are the same, date before terminating the participant from the trial. and the final assessment occurred 6  months aer enr ft ollment/ Telephone calls and home visits were conducted by approxi- vaccination. For the deferred vaccination group, the final posten - mately 100 recent nursing and pharmacy school graduates who rollment assessment occurred 18–24 weeks aer ft enrollment (the were based at the 3 follow-up coordination and data management assessment immediately before vaccination or at 6 months if not centers (follow-up centers): Western Area Urban (for participants vaccinated), and the final postvaccination assessment occurred from Western Area Urban and Western Area Rural districts), 6 months aer cr ft ossover vaccination. Bombali (for participants from Bombali and Tonkolili districts), Outcome: Full Compliance and Port Loko (for participants from the 3 sites in Port Loko We also evaluated full compliance with the study protocol for District). Aer t ft he main protocol training with all study sta,ff postenrollment and postvaccination activities. Participants in refresher training sessions were conducted at individual follow-up the immediate vaccination group who were vaccinated within sites to cover protocol updates, interpersonal communication 7  days of enrollment and completed all 6  monthly follow-up skills, and maintain confidentiality while tracking participants interviews were defined as being in full compliance postenroll - in their communities. Retention rates were regularly reviewed by ment and postvaccination. Deferred participants were consid- the STRIVE Data Safety and Monitoring Board and presented at ered in full compliance postenrollment if they were vaccinated weekly staff meetings at the 3 participant follow-up centers, where per protocol (approximately  18–24 weeks aer enr ft ollment) staff discussed strategies for contacting hard-to-reach participants. and completed every monthly assessment from enrollment to Duration of Follow-Up vaccination. Deferred participants who were never vaccinated Participants in the immediate vaccination group were vaccinated but completed every monthly interview through month 6 aer ft within 7  days of enrollment and called monthly for 6  months enrollment were also considered fully compliant postenroll- postvaccination (Figure 1). Deferred vaccination group partici- ment. Full compliance postvaccination in crossover participants pants were observed from enrollment until they were vaccinated were those who were vaccinated per protocol and completed all and then for 6 months postvaccination (Figure  1). Participants 6 monthly interviews aer vaccin ft ation (regardless of interview in both groups who were never vaccinated (eg, refused vaccina- completion between enrollment and vaccination). tion, or for the deferred group, ineligible at the time of vaccina- Statistical Methods tion) were observed for 6 months postenrollment. Participants were analyzed as randomized to either immediate or deferred groups. Participants who died during follow-up Data Analysis Outcome: Retention (n = 25) are excluded from these analyses of retention (although We defined retention as completion of the final monthly assess - they are included in the reports of the trial results [3]). ment, and we measured it at 2 time points: postenrollment We calculated the proportion of participants retained and and postvaccination (Figure  1). For postenrollment retention, in full compliance, and we identified factors associated with we compared results for the immediate and deferred groups, retention and full compliance using univariate and multivariate Vaccinated Immediate Group Post-vaccination follow-up N = 4311 Unvaccinated Enrollment and Immediate vaccination Deferred Group a Post-enrollment follow-up Post-vaccination follow-up N=4315 Enrollment Cross-over vaccination  0123456789 10 11 12 Months Includes 12 participants randomized to the deferred group who were vaccinated in error with the immediate group Figure 1. Postenrollment and postvaccination follow-up of the Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) participants. Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S67 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 stepwise logistic regression models adjusted for randomization participants were vaccinated during the crossover period, and assignment (immediate vs deferred), age, sex, education, occu- 12 (<1%) of the deferred participants were vaccinated immedi- pation, facility type, and enrollment site. Retention models also ately in error (Figure 2). These last 12 participants were included included vaccination status. The Western Area Rural site was in all retention and full compliance analyses as vaccinated and selected as the referent group for site comparisons because it deferred randomization arm participants. For analyses of full had the highest enrollment of the nonurban sites. A  stepwise compliance, they were considered noncompliant because they model selection approach was used to determine the best subset were not vaccinated per protocol. of predictors. e Th criterion used to enter the model was P = .10, Most participants were male (60.6%) and the median age was with P = .01 as the criterion to stay in the model once additional 31 years (range, 18–79 years) (Table 2). The majority of partici - predictors were added. Odds ratios (ORs) and 95% confidence pants were nurses (33.3%) or frontline workers (48.2%); 44.2% intervals (CIs) are presented for each level of predictor com- completed secondary school, and 41.6% had tertiary education. pared with its reference groups. Education was related to occupation, with all (100%) doctors and a very high proportion of pharmacists (87%) and nurses RESULTS (76%) having tertiary education. Frontline workers came from diverse educational backgrounds: the majority (59%) completed Of 8626 participants included in the analysis, 4311 were secondary school, whereas 25% had a primary school education randomized to immediate vaccination and 4315 to deferred or less. This diversity reflects the variety of technical roles front - vaccination. Overall, 7979 (92.5%) of 8626 randomized line workers performed during the Ebola response, including participants were vaccinated; 4157 (96.4%) of the immedi- contact tracing, ambulance driving, cleaning Ebola facilities, ate participants were vaccinated, 3810 (88.9%) of the deferred Ineligible: 127 Assessed for Eligibility: 8815 Declined Participation: 15 Excluded from Retention Study: 47 Randomized: 8673 Invalid Consent: 22 Died: 25 Retained Post-Enrollment: 7979 (93%) Retention Study Participants: 8626 Fully Compliant Post-Enrollment: 5934 (69%) Retained Post-Enrollment: 3875 (90%) Retained Post-Enrollment: 4104 (95%) Randomized Immediate: 4311 Randomized Deferred: 4315 Fully Compliant Post-Enrollment: 2769 (64%) Fully Compliant Post-Enrollment: 3165 (73%) Crossover Vaccinated: 3810 (89%) Vaccinated: 4157 (96%) Retained Post-Vaccination: 3628 (95%) Retained Post-Vaccination: 4012 (97%) Fully Compliant Post-Vaccination: 2455 (64%) Fully Compliant Post-Vaccination: 3165 (73%) Immediately Vaccinated (in Error) : 12 (<1%) Not Vaccinated (Declined): 154 (4%) Not vaccinated: 493 (11%) Declined: 352 No Longer Eligible: 100 Lost to Follow-up: 32 Other: 9 All Vaccinated: 7979 Retained Post-Vaccination: 7649 (96%) Fully Compliant Post-Vaccination: 5620 (70%) Figure 2. Sierra Leone Trial to Introduce a Vaccine Against Ebola (STRIVE) participant retention and compliance CONSORT diagram. Included in the vaccinated group; not eligible for crossover vaccination. S68 • JID 2018:217 (Suppl 1) • Carter et al Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Table 2. Retention, Compliance, and Participant Characteristics by Randomization Arm and Vaccination Status: STRIVE (n = 8626) Immediate Vaccination Deferred Vaccination Crossover All Vaccinated All Randomized (N = 4311) (N = 4315) (N = 3810) (N = 7979) (N = 8626) Study Variable n (%) n (%) n (%) n (%) n (%) c d d Retained 4104 (95.2) 3875 (89.8) 3628 (95.2) 7649 (95.9) 7979 (92.5) Fully Compliant 3165 (73.4) 2769 (64.2) 2455 (64.4) 5620 (70.4) 5934 (68.8) Male Gender 2611 (60.6) 2617 (60.6) 2454 (64.4) 5032 (63.1) 5228 (60.6) Age (Tertiled) (years) 18–27 1423 (33.0) 1452 (33.7) 1283 (33.7) 2658 (33.3) 2875 (33.3) 28–35 1470 (34.1) 1406 (32.6) 1213 (31.8) 2619 (32.8) 2876 (33.3) >35 1418 (32.9) 1457 (33.8) 1314 (34.5) 2702 (33.9) 2875 (33.3) Primary Occupation Nurse 1443 (33.5) 1428 (33.1) 988 (25.9) 2336 (29.3) 2871 (33.3) Doctor 13 (0.3) 11 (0.3) 7 (0.2) 20 (0.3) 24 (0.3) Pharmacist 20 (0.5) 19 (0.4) 19 (0.5) 39 (0.5) 39 (0.5) Allied health professional 77 (1.8) 72 (1.7) 58 (1.5) 130 (1.6) 149 (1.7) Community health worker 85 (2.0) 93 (2.2) 68 (1.8) 153 (1.9) 178 (2.1) Laboratory worker 133 (3.1) 139 (3.2) 139 (3.6) 269 (3.4) 272 (3.2) Frontline worker 2065 (47.9) 2089 (48.4) 1340 (35.2) 3377 (42.3) 4154 (48.2) Surveillance worker 246 (5.7) 250 (5.8) 170 (4.5) 410 (5.1) 496 (5.8) Other/Not reported 229 (5.3) 214 (5.0) 128 (3.4) 352 (4.4) 443 (5.1) Facility Type Clinic setting 796 (18.5) 794 (18.4) 837 (22.0) 1626 (20.4) 1590 (18.4) Ebola Facility 1491 (34.6) 1493 (34.6) 686 (18.0) 2119 (26.6) 2984 (34.6) Hospital 1722 (39.9) 1711 (39.7) 1365 (35.8) 3020 (37.8) 3433 (39.8) Other/Not reported 302 (7.0) 317 (7.3) 922 (24.2) 1214 (15.2) 619 (7.2) Education Tertiary 1810 (42.0) 1778 (41.2) 1500 (39.4) 3209 (40.2) 3588 (41.6) Secondary 1874 (43.5) 1937 (44.9) 1773 (46.5) 3614 (45.3) 3811 (44.2) Primary 233 (5.4) 216 (5.0) 191 (5.0) 420 (5.3) 449 (5.2) None 374 (8.7) 363 (8.4) 338 (8.9) 709 (8.9) 737 (8.5) Other/Not reported 20 (0.5) 21 (0.5) 8 (0.2) 27 (0.3) 41 (0.5) Enrollment Site District (Town) Western Area Rural 941 (21.8) 949 (22.0) 849 (22.3) 1746 (21.9) 1890 (21.9) Western Area Urban 1624 (37.7) 1620 (37.5) 1370 (36.0) 2915 (36.5) 3244 (37.6) Bombali 599 (13.9) 603 (14.0) 544 (14.3) 1137 (14.2) 1202 (13.9) Tonkolili 366 (8.5) 365 (8.5) 338 (8.9) 703 (8.8) 731 (8.5) Port Loko (Port Loko Town) 433 (10.0) 429 (9.9) 398 (10.4) 823 (10.3) 862 (10.0) Port Loko (Lunsar) 197 (4.6) 199 (4.6) 172 (4.5) 365 (4.6) 396 (4.6) Port Loko (Kaffu Bullom) 151 (3.5) 150 (3.5) 139 (3.6) 290 (3.6) 301 (3.5) Abbreviation: STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola.  There were no significant differences in demographic characteristics between the participants randomized to the immediate and deferred vaccination arms. Participants randomized to deferred vaccination arm who were vaccinated 18–24 weeks after enrollment. Participants randomized to immediate vaccination who were retained 6 months after enrollment. Participants who were retained 6 months after vaccination. Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student, and vaccinator. Includes dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabber (persons taking postmortem skin/mucosal swabs for Ebola testing). burial duties, and social mobilizing. There were no significant participants were retained 6 months postvaccination: 96.5% of differences in demographic characteristics between immediate immediate vaccinated participants and 95.2% of crossover vac- and deferred randomized groups. cinated participants. In the analysis of full protocol compliance, 5934 (68.8%) of the 8626 participants were in full compliance Retention and Full Compliance With Protocol postenrollment: 73.4% of participants in the immediate vaccina- Overall, 7979 (92.5%) of the 8626 participants randomized tion group and 64.2% of those assigned to deferred vaccination. were retained 6  months postenrollment: 95.2% of participants Of the vaccinated participants, 70.4% were in full compliance assigned to immediate vaccination and 89.8% of those assigned postvaccination: 73.4% of the immediate vaccinated and 64.4% to deferred vaccination (Table 2). Almost all (95.9%) vaccinated of the crossover vaccinated participants. Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S69 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Factors Associated With Retention nurses the second highest (99%), and doctors the lowest (90%). In multivariate analysis, 6-month postenrollment retention was Compared with nurses, doctors (aOR = 0.1; 95% CI, 0.02–0.5), higher among participants >27 years (adjusted OR [aOR] = 1.3, frontline workers (aOR  =  0.3; 95% CI, 0.2–0.4), community 95% CI, 1.1–1.6 for those 28–35; and aOR = 1.9, 95% CI, 1.6–2.4 health workers (aOR = 0.3; 95% CI, 0.1–0.8), and those no lon- for those >35) and those who were vaccinated (aOR: 3.5; 95% ger employed at the time of crossover vaccination (aOR = 0.3; CI: 2.9–4.3) (Table 3). Significantly lower retention was associ - 95% CI, 0.2–0.4) had significantly lower retention. ated with (1) frontline workers (aOR = 0.6; 95% CI, 0.5–0.7) and Factors Associated With Full Protocol Compliance surveillance workers (aOR  =  0.6; 95% CI, 0.4–0.9) compared Factors associated with full compliance postenrollment and with nurses and (2) enrollment at the Port Loko Town site com- postvaccination were similar. Randomization to immediate pared with the Western Area Rural site (aOR  =  0.5; 95% CI, vaccination arm, age >27  years, sites in Bombali and Tonkoli 0.4–0.7). Districts, and occupation as a laboratory worker were all associ- Retention 6  months postvaccination was associated with ated with a higher likelihood of full protocol compliance during older age (aOR  =  1.5, 95% CI, 1.2–2.0 for those 28–35; and both postenrollment and postvaccination periods in multivar- aOR  =  2.2, 95% CI, 1.7–3.0 for those >35) and enrollment at iate analyses (Table  5). Bombali and Port Loko (Lunsar, Kaffu sites in Bombali (aOR  =  2.2; 95% CI, 1.3–3.6) and Tonkoli Bullom) District sites generally had higher rates of full compli- (aOR = 3.0; 95% CI, 1.5–6.3) Districts compared with Western ance compared with Western Area Rural. Frontline and com- Area Rural in adjusted analyses (Table  4). Among occupa- munity health workers and those working at Ebola facilities tional groups, pharmacists had the highest retention (100%), Table 3. Factors Associated With Retention 6 Months Postenrollment Among Randomized Participants in the STRIVE Trial (n = 8626) Retained Univariable Multivariable Factor-Level Factor-Level N (%) Odds Ratio (95% CI) P Value Odds Ratio (95% CI) P Value Randomization Arm Deferred Vaccination 3875 (89.8) ref <.0001 Immediate Vaccination 4104 (95.2) 2.3 (1.9–2.7) Vaccination Status Unvaccinated 3958 (88.8) ref <.0001 ref <.0001 Vaccinated 4021 (96.4) 3.4 (2.8–4.1) 3.5 (2.9–4.3) Gender Male 4792 (91.7) ref .0003 Female 3187 (93.8) 1.4 (1.2–1.6) Age (tertiled) (years) 18–27 2587 (90.0) ref <.0001 ref <.0001 28–35 2668 (92.8) 1.4 (1.2–1.7) 1.3 (1.1–1.6) >35 2724 (94.7) 2.0 (1.6–2.5) 1.9 (1.6–2.4) Occupation Nurse 2723 (94.8) ref <.0001 ref .0002 Doctor 22 (91.7) 0.6 (0.1–2.6) 0.5 (0.1–2.3) Pharmacist 37 (94.9) 1.0 (0.2–4.2) 0.9 (0.2–3.9) Allied health professional 143 (96.0) 1.3 (0.6–3.0) 1.2 (0.5–2.7) Community health worker 163 (91.6) 0.6 (0.3–1.0) 0.7 (0.4–1.2) Laboratory worker 260 (95.6) 1.2 (0.7–2.2) 1.1 (0.6–2.1) Frontline worker 3777 (90.9) 0.5 (0.5–0.7) 0.6 (0.5–0.7) Surveillance worker 446 (89.9) 0.5 (0.4–0.7) 0.6 (0.4–0.9) Other/Not reported 408 (92.1) 0.6 (0.4–0.9) 0.6 (0.4–0.9) Facility Type Clinic setting 1468 (92.3) ref .0276 Ebola Facility 2744 (92.0) 1.0 (0.8–1.2) Hospital 3207 (93.4) 1.2 (0.9–1.5) Other/Not reported 560 (90.5) 0.8 (0.6–1.1) Enrollment Site Western Area Rural 1741 (92.1) ref <.0001 ref <.0001 Western Area Urban 3061 (94.4) 1.4 (1.1–1.8) 1.2 (0.9–1.5) Bombali 1117 (92.9) 1.1 (0.9–1.5) 0.9 (0.7–1.3) Tonkolili 670 (91.7) 0.9 (0.7–1.3) 0.7 (0.5–1.0) Port Loko (Port Loko Town) 747 (86.7) 0.6 (0.4–0.7) 0.5 (0.4–0.7) Port Loko (Lunsar) 360 (90.9) 0.9 (0.6–1.3) 0.7 (0.5–1.0) Port Loko (Kaffu Bullom) 283 (94.0) 1.4 (0.8–2.2) 1.1 (0.7–1.8) Abbreviations: CI, confidence interval; ref, reference category; STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola.  Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student and vaccinator. Includes dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabbers (persons taking post mortem skin/mucosal swabs for Ebola testing). S70 • JID 2018:217 (Suppl 1) • Carter et al Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Table 4. Factors Associated With Postvaccination Retention Among Vaccinated Participants, STRIVE (n = 7979) Retained Univariable Multivariable Factor-Level Factor-Level N (%) Odds Ratio (95% CI) P Value Odds Ratio (95% CI) P Value Randomization Arm Crossover Vaccination 3628 (95.2) ref .0039 Immediate Vaccination 4012 (96.5) 1.4 (1.1–1.7) Gender Male 4748 (94.6) ref <.0001 ref .0056 Female 2892 (98.1) 3.0 (2.3–4.0) 1.6 (1.2–2.3) Age (tertiled) (years) 18–27 2491 (93.8) ref <.0001 ref <.0001 >27–35 2539 (96.3) 1.7 (1.3–2.2) 1.5 (1.2–1.9) >35 2610 (97.6) 2.6 (2.0–3.5) 2.2 (1.7–3.0) Occupation Nurse 2312 (99.0) ref <.0001 ref <.0001 Doctor 18 (90.0) 0.1 (0.02–0.4) 0.1 (0.02–0.5) Pharmacist 39 (100) >99.9 (<0.01 to >99.9) >99.9 (<0.01 to >99.9) Allied health professional 124 (95.4) 0.2 (0.1–0.5) 0.2 (0.1–0.5) Community health worker 145 (96.0) 0.2 (0.1–0.6) 0.3 (0.1–0.7) Laboratory worker 261 (97.0) 0.3 (0.1–0.7) 0.5 (0.2–1.1) Frontline worker 3176 (94.3) 0.2 (0.1–0.3) 0.2 (0.2–0.4) Surveillance worker 393 (95.9) 0.2 (0.1–0.4) 0.4 (0.2–0.8) Not currently working 836 (93.6) 0.2 (0.1–0.2) 0.2 (0.1–0.4) Other/Not reported 336 (95.7) 0.2 (0.1–0.4) 0.3 (0.2–0.6) Facility Type Clinic setting 1536 (94.6) ref .0002 Ebola Facility 2037 (96.2) 1.4 (1.1–2.0) Hospital 2921 (96.9) 1.8 (1.3–2.4) Other/Not reported 1146 (94.6) 1.0 (0.7–1.4) Enrollment Site Western Rural 1655 (95.0) ref <.0001 ref <.0001 Western Urban 2783 (95.7) 1.2 (0.9–1.6) 0.8 (0.6–1.1) Bombali 1117 (98.2) 3.0 (1.8–4.9) 2.2 (1.3–3.6) Tonkolili 695 (98.9) 4.6 (2.2–9.6) 3.0 (1.5–6.3) Port Loko (Port Loko Town) 768 (93.4) 0.8 (0.5–1.1) 0.7 (0.5–1.0) Port Loko (Lunsar) 345 (94.4) 1.0 (0.6–1.6) 0.6 (0.3–1.0) Port Loko (Kaffu Bullom) 277 (95.5) 1.1 (0.6–2.1) 0.9 (0.5–1.6) Abbreviations: CI, confidence interval; ref, reference category; STRIVE, Sierra Leone Trial to Introduce a Vaccine Against Ebola.  Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student, and vaccinator. Includes pharmacist, dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabbers (persons taking post mortem skin/mucosal swabs for Ebola testing). Current occupation information was updated for deferred crossover participants during the crossover period. Many were no longer working as the Ebola outbreak subsided, and Ebola response jobs ended. Crossover participants were offered vaccination regardless of work status. had lower likelihood of completing all 6 assessments compared closed and response activities slowed towards the end of the with nurses during both postenrollment and postvaccination outbreak was a sizeable challenge to follow-up. Nonetheless, the follow-up. strategies used in the STRIVE trial, including collecting exten- sive locating information, using closed-user-group cell phone DISCUSSION technology and hotlines to facilitate communication, and hir- e S Th TRIVE trial achieved high participant retention through ing adequate staff who were familiar with the participant pop - 6  months of follow-up despite the challenges of implement- ulation and local areas resulted in overall high retention and ing a vaccine trial during an ongoing Ebola outbreak in a should be considered for future trials in low-resource and other resource-limited setting. The trial, which enrolled and vac - emergency settings. cinated almost 8000 participants, retained more than 92.5% Evaluating and reporting retention in randomized clinical of participants 6  months postenrollment and 97.4% postvac- trials is important because high rates of loss to follow-up (in cination. Participation in every monthly interview was less ranges >20%) can compromise the validity of the trial results complete, with 68.8% and 70.4% achieving full compliance in [7], particularly if the retention rate is different between the postenrollment and postvaccination interviews, respectively. study arms. We observed somewhat lower retention among par- We observed lower retention among Ebola frontline workers. ticipants randomized to the deferred vaccination group com- es Th e workers generally held temporary employment posi - pared with the immediate arm (89.8% vs 95.2%, respectively) in tions during the outbreak, and displacement as Ebola facilities the first 6 months postenrollment. The unblinded design of the Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S71 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 S72 • JID 2018:217 (Suppl 1) • Carter et al Table 5. Predictors of Full Compliance Among Randomized Participants Postenrollment and Postvaccination 6 Months Postenrollment (All Participants, n = 8626) 6 Months Postvaccination (All Vaccinated, n = 7979) Compliant Univariable Multivariable Compliant Univariable Multivariable Odds Ratio Factor- Level Factor- Level Odds Ratio Factor Level Odds Ratio Factor- Level N (%) (95% CI) P Value N (%) P Value N (%) (95% CI) P Value (95% CI) P Value Randomization Arm Deferred Vaccination 2769 (64.2) ref <.0001 ref <.0001 2455 (64.4) ref <.0001 ref <.0001 Immediate Vaccination 3165 (73.4) 1.5 (1.4–1.7) 1.6 (1.4–1.7) 3165 (76.1) 1.8 (1.6–1.9) 1.9 (1.7–2.1) Enrollment Site Western Rural 1189 (62.9) ref <.0001 ref <.0001 1118 (64.1) ref <.0001 ref <.0001 Western Urban 2176 (67.1) 1.2 (1.1–1.4) 0.9 (0.8–1.1) 2021 (69.5) 1.3 (1.1–1.4) 1.1 (0.9–1.2) Bombali 989 (82.3) 2.7 (2.3–3.3) 2.1 (1.8–2.6) 861 (75.7) 1.7 (1.5–2.1) 1.5 (1.2–1.8) Tonkolili 523 (71.5) 1.5 (1.2–1.8) 1.2 (1.0–1.4) 595 (84.6) 3.1 (2.5–3.9) 2.6 (2.1–3.3) Port Loko (Port Loko) 523 (60.7) 0.9 (0.8–1.1) 0.8 (0.6–0.9) 508 (61.8) 0.9 (0.8–1.1) 0.8 (0.7–1.0) Port Loko (Lunsar) 296 (74.7) 1.8 (1.4–2.2) 1.4 (1.1–1.9) 284 (78.0) 2.0 (1.5–2.6) 1.7 (1.3–2.3) Port Loko (Kaffu Bullom) 238 (79.1) 2.2 (1.7–3.0) 1.7 (1.3–2.3) 233 (80.3) 2.3 (1.7–3.1) 2.0 (1.4–2.7) Age (tertiled) years 18.0–27.5 1820 (63.3) ref <.0001 ref <.0001 1747 (65.8) ref <.0001 ref <.0001 27.5–35.4 2017 (70.1) 1.4 (1.2–1.5) 1.3 (1.2–1.5) 1896 (71.9) 1.3 (1.2–1.5) 1.3 (1.1–1.4) 35.4–79.5 2097 (72.9) 1.6 (1.4–1.8) 1.4 (1.3–1.6) 1977 (73.9) 1.5 (1.3–1.7) 1.3 (1.2–1.5) Gender Male 3568 (68.2) ref .1761 3466 (69.0) ref .0001 Female 2366 (69.6) 1.1 (1.0–1.2) 2154 (73.1) 1.2 (1.1–1.4) Occupation Nurse 2071 (72.1) ref <.0001 ref <.0001 1769 (75.8) ref <.0001 ref <.0001 Doctor 19 (79.2) 1.5 (0.6–3.9) 1.3 (0.5–3.4) 16 (80) 1.3 (0.4–3.8) 1.0 (0.3–3.0) Pharmacist 25 (64.1) 0.7 (0.4–1.3) 0.6 (0.3–1.2) 25 (64.1) 0.6 (0.3–1.1) 0.6 (0.3–1.1) Allied Health Professional 101 (67.8) 0.8 (0.6–1.2) 0.8 (0.5–1.1) 93 (71.5) 0.8 (0.5–1.2) 0.8 (0.5–1.2) Community health worker 110 (61.8) 0.6 (0.5–0.9) 0.7 (0.5–0.9) 97 (64.2) 0.6 (0.4–0.8) 0.6 (0.4–0.8) Laboratory worker 218 (80.1) 1.6 (1.2–2.1) 1.6 (1.2–2.2) 216 (80.3) 1.3 (0.9–1.8) 1.5 (1.1–2.0) Frontline worker 2716 (65.4) 0.7 (0.7–0.8) 0.8 (0.8–0.9) 2281 (67.7) 0.7 (0.6–0.8) 0.8 (0.7–0.9) Surveillance worker 361 (72.8) 1.0 (0.8–1.3) 1.2 (1.0–1.5) 312 (76.1) 1.0 (0.8–1.3) 1.2 (0.9–1.6) Other/Not reported 313 (70.7) 0.9 (0.8–1.2) 1.0 (0.8–1.2) 256 (72.9) 0.9 (0.7–1.1) 0.9 (0.7–1.1) Not currently working NA NA NA 555 (62.2) 0.5 (0.5–0.6) 0.9 (0.6–1.2) Facility Type Clinic setting 1140 (71.7) ref <.0001 ref <.0001 1154 (71.1) ref <.0001 ref <.0001 Ebola Facility 1855 (62.2) 0.7 (0.6–0.7) 0.8 (0.7–0.9) 1393 (65.8) 0.8 (0.7–0.9) 0.8 (0.7–0.9) Hospital 2524 (73.5) 1.1 (1.0–1.3) 1.2 (1.0–1.3) 2279 (75.6) 1.3 (1.1–1.4) 1.1 (0.9–1.3) Other/Not reported 415 (67) 0.8 (0.7–1.0) 1.0 (0.8–1.2) 794 (65.5) 0.8 (0.7–0.9) 1.0 (0.7–1.3) Vaccination Status Unvaccinated 2769 (62.1) ref <.0001 Vaccinated 3165 (75.9) 1.9 (1.8–2.1) Abbreviations: CI, confidence interval; ref, reference category; NA, not applicable. Includes nurse, midwife, community health nurse, nursing aide, maternal-child health aide, nursing student, and vaccinator. Includes pharmacist, dentist, medical counselor, nutritionist, and physiotherapist. Includes contact tracers, ambulance crew, burial workers, cleaners, and swabbers (persons taking post mortem skin/mucosal swabs for Ebola testing). Current occupation information was updated for deferred crossover participants during the crossover period. Many were no longer working as the Ebola outbreak subsided, and Ebola response jobs ended. Crossover participants were offered vaccination regardless of work status. Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 STRIVE trial may have contributed to higher postenrollment outside of Freetown oen h ft ad poor cell phone coverage, and retention in the immediate vaccination arm because trial sta ff there were periods when cellular service was down nationwide. may have prioritized contacting vaccinated participants—espe- Furthermore, because not all parts of the country are on the cially early in the study—because of concerns about possible national electric grid, participants frequently lacked access to vaccine-associated adverse events. However, we believe the electricity to charge their study cell phone. Home visits were modest difference in retention across trial arms is unlikely to oen t ft ime-, labor-, and resource-intensive due to poor quality ae ff ct the validity of our trial results [ 3]. of roads and limited transportation options, especially in the Participants vaccinated immediately aer ft enrollment and districts. Approximately 1  day per week, trial staff had use of those vaccinated aer a de ft lay of 18–24 weeks had similar high a study car to make field visits, but even then they oen h ft ad rates of retention, 96.5% and 95.2%, respectively, 6 months post- to use motorbikes from the main road to the villages where vaccination, suggesting that use of a delayed intervention arm participants lived due to lack of paved roads. Street names and is a feasible option for randomized clinical trials in emergency house numbers were frequently missing in both rural areas and settings where a placebo control arm would not be feasible. The densely populated urban areas, making it difficult to physically assurance that the intervention would be oer ff ed to everyone locate participants. Because some trial staff were students or enrolled may help to explain our relatively high retention in the recent graduates from urban Freetown who were assigned to deferred arm postenrollment. Clinical trials in the United States rural district locations, they were not always familiar with the have shown that participant retention tends to decline among people and location of the villages within the districts. Although those randomized to a placebo arm when the intervention is not we noted higher retention at the rural Tonkolili and semiurban oer ff ed to all at the end of the trial [ 8]. Bombali District sites, neither road infrastructure nor differ - Retention rates in the STRIVE trial are comparable to those ences in participant population explain why these sites were observed in other Ebola vaccine clinical trials conducted during more successful than others. the 2014–2016 Ebola outbreak in West Africa, despite differ - The ongoing Ebola outbreak also created unique barriers ent study designs, lengths of follow-up, and participant popu- to follow-up. Ebola response activities appropriately took lations. The World Health Organization “ Ebola Ca Suffit ” ring precedence over study activities, so there were times when vaccination trial [9] in Guinea, which enrolled village residents study procedures had to be modified or even suspended. For who were contacts or contacts of contacts of a case, reported example, there were several times during the trial when study 94% participant retention, although follow-up was <3 months. staff could not conduct home visits because the village was In Liberia, the PREVAIL trial [10] achieved 97.7% retention of under quarantine due to an Ebola case. In one district, sta ff 1500 participant volunteers residing in Monrovia at 6  months at a large hospital were quarantined for 21  days after being and 98.3% at 12  months. es Th e 3 West African Ebola vac - exposed to an Ebola case. Many of these staff members were cine trials all used similar approaches to follow-up: obtaining STRIVE trial participants who consequently could not be detailed contact information, providing a study cell phone or contacted for that month’s assessment if they were not reach- trial hotline number, conducting home visits to track default- able by phone. ers, and employing dedicated staff to conduct follow-up. These Perhaps the biggest challenge to follow-up was displace- results validate the effectiveness of these approaches. ment of participants. Participant characteristics we identified e S Th TRIVE trial invested substantial time and expertise in as associated with higher risk of becoming lost to follow-up— ongoing community engagement with the Ebola and health specifically, younger age and employment as a frontline Ebola facilities and with political and religious leaders in the districts worker—were likely strongly related to displacement. Beginning where the trial was implemented. In nonemergency settings, in November 2015, Ebola facilities began to close and response large pediatric trials in Gambia [11, 12] have similarly achieved activities started winding down. Healthcare and frontline high follow-up of >90% using such methods, including com- Ebola workers, many of whom had moved to the district cen- munity engagement and using dedicated follow-up staff to stay ters to work on the response, returned to their homes, oen in ft in contact with participants through telephone calls and home remote villages, making telephone follow-up more difficult and visits if they missed a follow-up appointment. Other charac- home visits virtually impossible. Many young nurse graduates teristics of the STRIVE trial, such as development of trusting enrolled in the STRIVE trial while awaiting their government relationships between study staff and participants, rigorous posting, a universal first step aer co ft mpleting training. During follow-up efforts, and provision of ongoing staff training, have follow-up, some received postings to health facilities in remote been shown to improve retention in various vaccine trials in the districts, far from where they had enrolled in the STRIVE trial, United States and Africa [13–15]. where cell phone service was unreliable and home visits unfea- er Th e were challenges to conducting follow-up in a sible, posing challenges to retention. Finally, some participants resource-limited setting with a large cohort of participants reported that there was less motivation to complete the trial as spread over an expansive geographical area. Rural districts Ebola receded from the public mind. Retaining Participants in a Vaccine Trial • JID 2018:217 (Suppl 1) • S73 Downloaded from https://academic.oup.com/jid/article/217/suppl_1/S65/4999148 by DeepDyve user on 14 July 2022 Potential conifl cts of interests. P.  D.  and C.  R. P.  report grants from CONCLUSIONS the Biomedical Advanced Research and Development Authority during e S Th TRIVE trial maintained high retention of participants, the conduct of the study. All authors have submitted the ICJME Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors con - demonstrating the feasibility of a study design with both sider relevant to the content of the manuscript have been disclosed. a deferred vaccination arm and a long follow-up period. Furthermore, conducting safety assessment follow-up using References telephone interviews rather than in-person visits facilitated 1. Schulz KF, Grimes DA. Sample size slippages in randomised trials: exclusions and enrollment of a large number of participants across multi- the lost and wayward. Lancet 2002; 359:781–5. 2. Robinson KA, Dennison CR, Wayman DM, Pronovost PJ, Needham DM. ple Ebola-ae ff cted districts and resulted in high retention and Systematic review identifies number of strategies important for retaining study completion rates. participants. J Clin Epidemiol 2007; 60:757–65. 3. Samai M, Seward JF, Goldstein ST, et al. The Sierra Leone trail to introduce a vac - Key lessons learned include the value of the following: (1) cine against Ebola: an evaluation of rVSVΔG-ZEBOV-GP vaccine tolerability collecting extensive locating information, including telephone and safety during the West Africa Ebola outbreak. J Infect Dis 2018; 217:s6–15. 4. Widdowson MA, Schrag SJ, Carter RJ, et  al. Implementing an Ebola vaccine numbers of close relatives and friends who would know how study—Sierra Leone. MMWR Suppl 2016; 65:98–106. to reach participants; (2) flexibility to conduct home outreach 5. Carter RJ, Idriss A, Widdowson MA, et al. Implementing a multi-site clinical trial in the midst of an Ebola outbreak: lessons learned from the Sierra Leone trial to for defaulter tracking; (3) use of closed-user-group cell phone introduce a vaccine against Ebola. J Infect Dis 2018; 217:s16–23. technology and hotlines to facilitate communication to report 6. Callis A, Carter VM, Albert AP, et  al. Lessons learned in clinical trial commu- illness or a change in contact information; and (4) employing an nication during an Ebola outbreak: the implementation of STRIVE. J Infect Dis 2018; 217:s40–7. adequate number of staff who are familiar with the participant 7. Fewtrell MS, Kennedy K, Singhal A, et al. How much loss to follow-up is accepta- population and local areas. Future trials in outbreak settings can ble in long-term randomised trials and prospective studies? Arch Dis Child 2008; 93:458–61. consider using these strategies to oer in ff terventions to persons 8. McCarthy-Keith D, Nurudeen S, Armstrong A, Levens E, Nieman LK. residing in wide geographical areas not tethered to a single clin- Recruitment and retention of women for clinical leiomyoma trials. Contemp Clin Trials 2010; 31:44–8. ical site. 9. Henao-Restrepo AM, Camacho A, Longini IM, et al. Efficacy and effectiveness of an rVSV-vectored vaccine in preventing Ebola virus disease: final results from the Guinea ring vaccination, open-label, cluster-randomised trial (Ebola Ça Suffit!). Notes Lancet 2017; 389:505–18. Acknowledgments. We express our gratitude to the trial participants 10. Kennedy SB, Bolay F, Kieh M, et al. Phase 2 placebo-controlled trial of two vac- and thanks to the dedicated Sierra Leone Trial to Introduce a Vaccine cines to prevent Ebola in Liberia. N Engl J Med 2017; 377:1438–47. Against Ebola (STRIVE) staff from COMHAS, University of Sierra Leone, 11. Cutts FT, Zaman SM, Enwere G, et  al. Efficacy of nine-valent pneumococcal the Sierra Leone Ministry of Health and Sanitation, the Centers for Disease conjugate vaccine against pneumonia and invasive pneumococcal disease in Control and Prevention (CDC) Sierra Leone Country Office, and the CDC e Ga Th mbia: randomised, double-blind, placebo-controlled trial. Lancet 2005; 365:1139–46. STRIVE teams in Atlanta and Sierra Leone for their tireless work. 12. Idoko OT, Owolabi OA, Odutola AA, et al. Lessons in participant retention in the Disclaimer. The findings and conclusions in this report are those of course of a randomized controlled clinical trial. BMC Res Notes 2014; 7:706. the author(s) and do not necessarily represent the official position of the 13. Johnson RE, Williams RD, Nagy MC, Fouad MN. Retention of under-served Centers for Disease Control and Prevention. women in clinical trials: a focus group study. Ethn Dis 2003; 13:268–78. Financial suppoort. e t Th rial was funded by the U.S. Centers for 14. Kennedy SB, Neaton JD, Lane HC, et al. Implementation of an Ebola virus disease Disease Control and Prevention, the Biomedical Advanced Research and vaccine clinical trial during the Ebola epidemic in Liberia: design, procedures, Development Authority, and the National Institutes of Health, with addi- and challenges. Clin Trials 2016; 13:49–56. tional support from the CDC Foundation. 15. Larson GS, Baseler BR, Hoover ML, et  al. Conventional wisdom versus actual Supplement sponsorship. This work is part of a supplement sponsored outcomes: challenges in the conduct of an Ebola vaccine trial in Liberia during the international public health emergency. Am J Trop Med Hyg 2017; 97:10–5. by the Centers for Disease Control and Prevention. S74 • JID 2018:217 (Suppl 1) • Carter et al

Journal

The Journal of Infectious DiseasesOxford University Press

Published: May 18, 2018

Keywords: follow-up; ebola virus disease; vaccines; ebola vaccines; disease outbreaks; home visits

There are no references for this article.