In a recent study published in the Journal of the American Medical Association, researchers examined the effectiveness of opt-in and direct-mail strategies for offering self-sampling kits of human papillomavirus (HPV) to populations based on cervical cancer screening history.
Background
In the United States (US), over 50% of cervical cancers are diagnosed among people overdue for screening. Adherence to cervical cancer screening declined from 86% in 2005 to 77% in 2019, which was further reduced by coronavirus disease 2019 (COVID-19)-related disruptions. US guidelines recommend primary screening of individuals aged 30–65 by HPV testing of clinician-collected samples.
This is a preferred approach or recommended alternative to Papanicolaou/HPV co-testing or Papanicolaou testing. Papanicolaou testing is less sensitive than HPV testing for the diagnosis of cervical pre-cancer. Further, self-collected samples can be used in HPV tests with comparable sensitivity and specificity to clinician-collected samples.
Notably, self-screening eliminates the requirement for clinic appointments and negativity regarding pelvic examinations. Although several countries have introduced HPV self-sampling, the Food and Drug Administration has yet to approve it. Trials conducted on HPV self-sampling did not specifically evaluate kit uptake rates among people adherent to screening or with an unknown screening history.
About the study
In the present study, researchers evaluated the efficacy of opt-in and direct-mail strategies for offering HPV self-sampling kits. The team used electronic health records from the Kaiser-Permanente Washington (KPWA) care system to identify individuals who were due (for screening in ≤ three months) or overdue (Papanicolaou testing > 3.25 years or co-testing > 5.25 years) or had an unknown screening history (no recorded history).
Inclusion criteria were age 30–64, female sex, intact cervix, KPWA insurance, having a primary care clinician, and being due or overdue for screening. The team stratified eligible subjects by screening history and randomized them into education, usual care, opt-in, or direct-mail groups. The usual care group did not receive study-related interventions or contact.
The education group received the usual care and educational materials to motivate them for screening. The direct-mail group received usual care and HPV self-sampling kits. The opt-in group received the same intervention as the direct-mail group, except that participants had to request a self-sampling kit. The kits contained dry testing swabs and patient specimen labels.
Samples were tested, and results were documented and reported per usual care. Test results that required follow-up were routed to the primary care clinician and a centralized licensed nurse. The nurse monitored adherence through follow-up visit completion (in-clinic colposcopy for individuals positive for HPV-16 or HPV-18 or reflex Papanicolaou testing for those positive for other high-risk HPVs).
The primary outcome of the study was screening completion within six months after randomization. Secondary outcomes included incremental cost-effectiveness ratios, screening initiation, time from randomization to screening completion, and completion of recommended follow-up after positive test results.
Findings
The team randomized 31,355 individuals with an average age of 45.9 years. Most participants who were due (73.4%) or overdue (73.6%) for screening or had an unknown screening history (64.2%) were White. Baseline characteristics were similar across study groups. Most individuals completed the six-month follow-up.
Among participants due for screening, 61.7% in the direct-mail group and 51.1% in the opt-in group completed screening, compared to 47.6% in the education group. Further, among overdue participants, 35.7% in the direct-mail group completed screening compared to 18.8% in the education group. Among participants with an unknown screening history, 18% in the opt-in group completed screening compared to 15.9% in the education group.
The time to screening completion was shorter for opt-in and direct-mail groups than for the education group. It was also shorter for participants who returned HPV kits. About 9% of those who returned kits required follow-up (in-clinic reflex Papanicolaou testing or colposcopy). There were no unexpected adverse events. However, expected adverse events (vaginal pain and dysuria) were reported as mild discomfort.
Conclusions
In sum, directly mailing HPV self-sampling kits resulted in more than 14% higher screening among participants who were due or overdue for cervical cancer screening than education alone. By contrast, the opt-in strategy minimally increased screening. The direct-mail group also had a shorter time to screening completion.
Notably, the absolute difference in screening completion between direct-mail and education groups was nearly 17% among overdue individuals. Overall, the findings suggest prioritizing the direct-mail approach for those who are due or overdue for cervical cancer screening. For people with an unknown screening history, exploring alternative outreach strategies is required.
Journal reference:
- Strategies to Increase Cervical Cancer Screening with Mailed Human Papillomavirus Self-Sampling Kits. Rachel L. Winer, John Lin, Melissa L. Anderson, Jasmin A. Tiro, Beverly B. Green, Hongyuan Gao, Richard T. Meenan, Kristina Hansen, Angela Sparks, Diana S. M. Buist. JAMA, 2023.
doi: 10.1001/jama.2023.21471 https://jamanetwork.com/journals/jama/fullarticle/2812323#:~:text=Our%202014%2D2017%20HOME%20trial,%25)%2C%20similar%20to%20international%20trials.