Increasing participation in cervical cancer screening: Telephone contact with long‐term non‐attendees in Sweden. Results from RACOMIP,a randomized controlled trial

Non‐participation is the foremost screening‐related risk factor for cervical cancer. We studied the effectiveness and cost‐effectiveness of an intervention to increase participation in the context of a well‐run screening program. Telephone contact with non‐attendees, offering an appointment to take a smear, was compared with a control group in a population‐based randomized trial in western Sweden. Of 8,800 randomly selected women aged 30–62, without a registered Pap smear in the two latest screening rounds, 4,000 were randomized to a telephone arm, another 800 were offered a high‐risk human papillomavirus (HPV) self‐test by mail (not reported in this article) and 4,000 constituted a control group. Endpoints were frequency of testing, frequency of abnormal smears and further assessment of abnormal tests. Participation during the following 12 months was significantly higher in the telephone arm than in the control group, 718 (18.0%) versus 422 (10.6%) RR: 1.70, 95% confidence interval (CI): 1.52–1.90]. The number of detected abnormal smears was 39 and 19, respectively (RR: 2.05, 95% CI: 1.19–3.55). The respective numbers of further assessed abnormalities were 34 and 18 (RR: 1.89, 95% CI: 1.07–3.34). Twice as many high‐grade intraepithelial neoplasia (CIN2+) were detected and treated in the telephone arm: 14 and 7, respectively. Telephone contact with women who have abstained from cervical cancer screening for long time increases participation and leads to a significant increase in detection of atypical smears. Cost calculations indicate that this intervention is unlikely to be cost‐generating and this strategy is feasible in the context of a screening program.