Research review 047
Randomised study
Short answer
Sending self-sampling kits directly resulted in more HPV samples being returned than letters requiring recipients to order a kit themselves. A personal offer at a general practice could also work well, but reached only part of the target group. The study shows why both how effective an offer is and how many people receive it need to be examined.
What did the researchers investigate?
The researchers tested different ways of offering self-sampling to people overdue for cervical screening. Some general practices were randomly selected to offer kits when an eligible person attended a consultation. In addition, a subset of people was individually randomised to direct kit mailing, a letter inviting them to order a kit, or no such systematic offer.
The trial therefore included both practice-level and individual-level allocation. The same people could be included in both parts. Figures from the two components must not be added together as if they represented different participants.
What did they find?
Direct mailing led to more returned samples
A total of 12.3% returned a self-sample when the kit was mailed directly, compared with 4.8% when the letter asked recipients to order it. The difference is 7.5 percentage points, or roughly eight additional returned samples per hundred recipients.
A personal offer at the practice had high uptake among those who actually received it. However, only some of all eligible people were reached this way. High participation among recipients can therefore translate into a more modest change across the entire target population.
Among 71 people with a positive self-sample, 51, approximately 72%, had undergone clinician-collected sampling or colposcopy within six months. This shows that initial testing and further follow-up are separate steps. The small subgroup numbers do not provide a reliable basis for ranking follow-up between the approaches.
How certain can we be?
Randomisation strengthens comparisons of offers, particularly direct mailing versus an order invitation. However, there were only 13 practices, whose patient populations differed. A small number of practices makes it harder to be confident that all differences were evenly distributed.
Offering a kit during a consultation depended on the person attending the practice and staff actually offering the kit. The finding therefore reflects both patient choices and service implementation. Limited follow-up and small numbers of positive tests make the long-term benefits more uncertain.
Who do the findings apply to?
The study took place at 13 general practices in Hounslow, London, England. Practice-level allocation included 12,657 women, while 6,400 of these also participated in the individual-level allocation of offers. They were aged 25–64 and at least six months overdue for screening.
The area had low screening coverage and considerable ethnic diversity. Among people whose ethnicity was known, approximately 47% were White, 37% Asian and 8% Black. About 42% had never been screened. The findings are especially relevant to this type of target population, not automatically to everyone in a national programme.
What might this mean in everyday life?
For programmes aiming to reach more people, removing the need to order a kit may help. Personal contact may also be valuable when available. The study supports considering combinations of offers while tracking who falls between services. An approach that works for frequent GP visitors will not necessarily reach those who have little contact with primary care.
What do we still need to know?
Personal offers and postal mailing should be studied together. This may show how to reach both people attending other medical appointments and those seldom in contact with services.
Everyone with a positive test should be followed to determine whether they receive the necessary examinations. Researchers should also compare costs and how many people each approach reaches. A high return rate among those receiving a personal offer means little to people who never receive one.
Sources
The publication covered by this review
Lim AWW, Landy R, Rigney J, North B, Sasieni PD. Impact of mode of offer of self-sampling to people overdue cervical screening on screening participation: a randomised controlled trial. EClinicalMedicine, 2025.
Other studies and professional sources mentioned in the text
Background source: Mailing Vaginal or Urine Self-Sampling Kits versus Conventional Invitation Letters: A Randomized Trial to Assess the Effect on Cervical Cancer Screening Attendance. 2025. Read the publication.
Background source: Opportunistic offering of self-sampling to non-attenders within the English cervical screening programme: a pragmatic, multicentre, implementation feasibility trial with randomly allocated cluster intervention start dates (YouScreen). 2024. Read the publication.
Last reviewed: 30 October 2026.
Source basis: The publication's full text.
Editorial update, 1 November 2026: Language, explanations and references were reviewed. The source-review date is stated above.