br Finally regarding potential confounders
Finally, regarding potential confounders, several studies have shown that young women are more likely than older women to have never screened (Catarino et al., 2016; Were et al., 2011); with the lowest screening rates reported by women aged 20–29 years (42–51%), re-lative to those aged 45–54 years, who have the highest screening rate (63%; AIHW, 2017). Thus, age group will be controlled in the planned study analyses in this study. Consistent with the prior relevant litera-ture, the number of psychological and practical screening barriers and facilitators, and individual screening barriers and facilitators were ex-amined in regards to women's screening status and prior screening. Specifically, it 146724-94-9 was hypothesized that:
1. The numbers of practical barriers and facilitators listed by women will more strongly relate to screening status than the number of psychological barriers.
2. Specifically, a lack of time (practical barrier) and GP status (prac-tical facilitator) will be most strongly related to women's screening status.
Finally, as a research question, the relationship between prior screening and numbers of barriers and facilitators and individual bar-riers and facilitators will be explored making no a priori assertions.
.005, and using up to five predictors in the logistic regression analyses. Conservative power and p-values were applied in the analyses to ac-commodate the larger number of young vs. older women in the sample.
2.2. Statistical analysis
Routine statistical analyses were conducted using the Statistical Package for the Social Sciences (Version 24) (IBM, 2015). Logistic re-gression analyses explored the relationships between screening status (or prior screening), age group, and the number of practical and psy-chological barriers (or facilitators) nominated by the women; χ2 ana-lyses then examined whether screening status (or prior screening) varied according to the individual barriers and facilitators they had nominated. Bonferroni-type adjustments were applied to the study analyses, with significance set at p < .005.
Ethics approval for the study was granted by the Australian National University Human Research Ethics Committee (Approval # 2015/614). The study was conducted in Canberra, Australia, and women were
recruited to the study from November 2015 to January 2016. They were recruited to the study by posters placed at the university campus and public areas in Canberra, Australia (e.g., shopping centers), and online advertisements placed on discussion boards (e.g., Whirlpool and Reddit). They were eligible to participate if they were: Australian re-sidents, fluent in English, sexually active, had not had a hysterectomy, and were aged 25–35-years (young women) or 45–55-years (older women). They were entered into a “lottery” to win one of 12 $20 gift vouchers.
Participants were asked to indicate their age, relationship status, and employment status. In the study analyses, age was binary coded as follows: young women (25–35-years) vs. older women (45–55-years). The study outcome variables were: screening status (i.e., up-to-date vs. overdue for screening by at least three months, including women who had never screened); and prior screening (i.e., prior screen vs. never screened). If a woman had previously screened, then they were asked to indicate the year and month of their most recent Pap test.
Cervical cancer screening barriers and facilitators were elicited by asking the women to list up to 10 factors that made pollen tube difficult to screen, and 10 factors that made it easier for them to screen, using free text. Their responses were coded as practical or psychological barriers or facilitators. Psychological factors included the emotional response to the test (e.g., anxiety about results, reassured by friendly doctor), knowl-edge gaps (e.g., correct age to start testing), and perceptions/attitudes (e.g., perceived cervical cancer risk). Discomfort caused by the proce-dure was coded as a psychological barrier as some of the women in-dicated that it had caused them distress. Practical factors were external factors that included the logistics of attending the screening (e.g., lack of time, nearby clinic). Using the definitions, one researcher (TM) coded each barrier and facilitator nominated by the women, and the codes were then verified by another researcher (RB). There were no cases of disagreement between the coding applied by the two re-searchers. Four scores were calculated for each woman: the number of practical barriers, psychological barriers, practical facilitators, and psychological facilitators (score range: 0–10).