Increasing uptake of colorectal cancer screening

BMJ 2009; 338 doi: http://dx.doi.org/10.1136/bmj.a2658 (Published 08 January 2009) Cite this as: BMJ 2009;338:a2658
  1. Anthony Jerant, associate professor
  1. 1Department of Family and Community Medicine, University of California Davis School of Medicine, Sacramento, CA 95817, USA
  1. afjerant{at}ucdavis.edu

    New approaches hold promise in motivating the 50% of people who do not attend

    Colorectal cancer is a leading cause of mortality in developed countries,1 2 yet uptake of screening for this cancer is lower than for other mass cancer screening interventions.3 Globally only about half of the eligible population undergoes colorectal cancer screening after prompts such as mail or telephone invitations, doctors’ recommendations during visits, and mass media campaigns.4 5 Thus, finding effective, feasible ways to motivate the other half of the population to undergo screening is an important priority for public health worldwide. The study by Hoff and colleagues (doi:10.1136/bmj.a2794) looked at one promising approach.6

    The authors used data from 12 960 people in the Norwegian Colorectal Cancer Prevention Trial 1 to investigate associations between the uptake of screening (flexible sigmoidoscopy alone or combined with faecal occult blood testing) and the timing of mailed invitations to screening appointments scheduled for six to seven weeks after the date of mailing.6 They found that uptake of screening was higher when people were invited for a screening appointment within one to two weeks of their birthday and when they received the screening invitation during their birthday week or during December, regardless of the date of their birthday.

    These findings indicate that mailed invitations to prescheduled screening appointments, a simple approach used in many offices, might be more effective if mailings and appointments are linked to birthdays or the winter holidays, or both. Hoff and colleagues’ analyses were post hoc and observational and therefore cannot prove that the intervention caused the rise in the uptake of screening. Because the opportunity costs of new interventions may offset potential gains, randomised controlled trials are needed to compare the effect of screening invitations and appointments timed to birthdays and holidays with non-timed mailings and appointments.

    The most useful interventions that change health behaviour are typically those developed in accordance with theoretical models of health behaviour that in research studies have been shown to predict health behaviour.7 Among other advantages, this allows the variables or constructs within the model that drive or mediate the effects of interventions to be identified.8 Unfortunately, Hoff and colleagues were unable to look at why screening invitations linked to birthdays and winter holidays might be effective. They suggest the reason may be increased perception of ageing; if this is true, mailed invitations timed to birthdays or winter holidays might be more effective if their content plays on that perception. Other explanations are equally plausible. For example, birthdays and holidays may simply serve as reminders to get things done as the personal or calendar year winds down and a new one begins.

    Further research is needed to clarify the reasons for the effect, but regardless of the explanation, the absolute increase in uptake of colorectal cancer screening that might be expected from timing invitations and appointments around birthdays and holidays is relatively small. Hoff and colleagues found about 3% greater uptake associated with timing around birthdays, and about 8% greater uptake with timing around the winter holidays.

    So how can we motivate people to take up colorectal cancer screening? A variety of innovative methods will be needed, because even a single powerful approach might improve uptake by only 10-15%. Several new approaches are emerging, such as the use of patient navigators (people trained to help patients negotiate the healthcare system) and the redesign of office systems and processes (for example, using patient medical record prompts to identify those who are overdue for screening, or training ancillary office staff to participate in preventive counselling).9 Interventions that are personally tailored to psychological mediators of health behaviour may be especially promising. These interventions capitalise on the growing body of research indicating that a small number of psychological constructs mediates a wide array of health behaviours.7

    One well studied psychological construct is self efficacy, or people’s confidence in their ability to complete the tasks required to attain a goal.10 Personally tailored interventions can be delivered in several ways. With an interactive computer approach, participants who have low self efficacy for colorectal cancer screening on a pre-intervention questionnaire could be provided with detailed information that increases their confidence for undergoing screening. In contrast, people with high self efficacy might receive brief affirmation and then be directed to a module aimed at other psychological mediators of screening behaviour for which their personal standing is less favourable, such as perceived barriers to screening (for example, fear of pain during sigmoidoscopy).

    Meta-analyses suggest that interventions tailored to psychological mediators of health behaviour may improve care more than non-tailored interventions.11 12 The time is right for their wider application and evaluation, including determination of their cost effectiveness.

    In addition, even highly effective new interventions could increase colorectal cancer screening overall, yet increase sociodemographic inequalities in screening.4 5 Interventions must therefore be designed and deployed so they are salient and accessible to a wide range of people, regardless of sociodemographic characteristics. Fortunately, personal tailoring, patient navigators, and thoughtful redesign of office systems all seem to avoid worsening disparities in care and may even mitigate disparities.9


    Cite this as: BMJ 2009;338:a2658


    • Competing interests: None declared.

    • Provenance and peer review: Commissioned; not externally peer reviewed.