| dc.description.abstract | Abstract Background: Participants from vulnerable populations in most modern clinical trials are underrepresented thereby impacting the generalizability of clinical trial results. This is particularly worrisome – where the intervention is behavioral and the outcome is cancer screening– as both are not well understood but are essential to addressing cancer health disparities and closing gaps in care. We used the World Health Organization’s (WHO) social determinants of health to frame results of clinical trial participation, long-term follow-up, and cervical cancer prevention outcomes among vulnerable participants- women leaving incarceration. Methods: Three years of follow-up data from a cervical cancer behavioral intervention clinical trial that took place in three county jails between 2014 and 2016 was used to report three outcomes: a) clinical trial completion for women in jails; b) factors associated with best practice community follow-up of women three years after the jail-based intervention; c) cervical cancer health literacy and up-to-date Pap testing at three-year follow-up. Results: Of the 261 women who consented to be in the original jail-based clinical trial (about half of the women incarcerated at local jails were recruited on any given day), 114/142 (80.3%) of those assigned to the intervention group completed the intervention. 70/119 (58.8%) of those in the waitlist control group completed the intervention. The primary factor associated with completion of the jail-based intervention was being assigned to the intervention group in Week 1 (by Week 2 when the control group received the intervention, 32 had been discharged from jails or transferred, 5 were cancelled by the intervention study staff, and 12 for other reasons). One hundred eleven of the 182 (61%) of intervention completers were retained at three years post-intervention and completed all pre, post, and year 3 assessments. Predictors of long-term follow-up included food security and lower utilization of public benefits. Among these participants, with regards to the cervical health literacy, there were statistically significant gains in three of the eight cervical health literacy domains when compared to post-intervention scores and seven of the eight domains when compared to pre-intervention scores (p < 0.05). Statistically significant predictors of cervical health literacy scores at year three included age within the confidence domain (b = 0.03, p = 0.030), education level within the susceptibility domain (b = – 0.82, p = 0.006), food insecurity within the barrier domain (b = 0.70, p = 0.006), having a past cervical cancer diagnosis within the severity domain (b = –1.15, p = 0.012), having received public benefits within the motivation domain (b = 0.62, p = 0.038) and having experienced racism within the self-efficacy domain ( b = – 0.90, p = 0.033). Despite these improvements to cervical health literacy, up-to-date Pap testing remained steady at 74% compared to 75% at baseline (p = 0.679). Conclusion: These findings illuminate how WHO social determinants of health predict clinical trial participation, long-term follow-up, and outcomes among vulnerable women. The study offers insights into how clinical trials for behavioral approaches to cancer prevention can be designed, implemented, and evaluated when targeting cancer prevention among vulnerable populations. | |
