Posted: 08/05/2011; Nat Rev Gastroenterol Hepatol © 2011 Nature Publishing Group
Novel interventions aimed at increasing colorectal cancer screening are needed to reduce mortality from this preventable disease. Two randomized controlled trials have found that a multicomponent outreach program increased screening rates by ~6% among patients with an expired colonoscopy order, while personalized electronic messages had no sustainable effect on screening rates.
Colorectal cancer (CRC) is the third most common cancer, and the second leading cause of cancer-related death, in the USA. Routine screening of individuals ≥50 years of age is a cost-effective yet underutilized strategy for reducing both the incidence and mortality of this largely preventable disease. While virtually all authoritative groups in the USA endorse various screening modalities, only 62% of the at-risk US population have had either fecal occult blood testing within the past year or lower endoscopy (colonoscopy or sigmoidoscopy) within the past 10 years.[1] The vast majority of individuals eligible for screening have never been screened and, therefore, these individuals stand to gain the greatest benefit from interventions aimed at increasing participation. Effective interventions at the level of the patient, the provider and the system have previously been evaluated, but the overall effect of individual strategies on screening rates has been relatively modest.[2] Moreover, the actual implementation and integration of these interventions into routine clinical practice has been hindered by the wide diversity in practice models, and constraints on physician time, cost and resources. Multilevel interventions that can be easily integrated into a diverse array of practice settings at low cost are needed. Two randomized controlled trials attempted to address this need by assessing the effect of a multicomponent outreach program[3]and personalized electronic messages[4] on CRC screening rates.
Cameron and colleagues evaluated the effect of a combined reminder-outreach intervention among patients with no history of CRC who had an expired order for a screening colonoscopy in an urban, primary care, internal medicine practice.[3] Eligible patients were identified using data contained in electronic health records. Individuals randomly assigned to the intervention group (n = 314) received a personalized letter from the primary care physician reminding them about their screening colonoscopy and reiterating the importance of screening, an educational brochure outlining common myths and questions regarding CRC screening, and a DVD about CRC screening. Patients in the intervention group also received a telephone call to confirm receipt of the mailing 2 weeks later. The control group (n = 314) received a personalized letter from the primary care physician reminding them about their colonoscopy and reiterating the importance of screening. At 3 months of follow-up, a higher number of patients in the intervention group had completed CRC screening compared with the control group (9.9% versus 3.2%, respectively; P = 0.001); screening rates increased in both groups at 6 months of follow-up (18.2% versus 12.1%; P = 0.03), but the absolute difference remained the same at ~6%. Interestingly, only 30% of patients in the intervention group who confirmed receipt of the mailing actually went on to watch the enclosed DVD. The investigators concluded that this multidimensional intervention had a "modest positive effect"; however, the relative improvement and absolute magnitude of the effect was small. Limitations of the study design included an inability to identify which components of the intervention were most effective, an inability to control for the quality of counseling by the ordering physician or reasons for the expired colonoscopy orders and, most importantly, the possibility that the follow-up phone call, which was conducted to evaluate the intervention, might have actually become part of the intervention itself.
In a separate study, Sequist et al. evaluated the effect of personalized electronic messages and risk assessment on CRC screening rates among 1,103 patients who were overdue for screening in 14 ambulatory health centers.[4]An online integrated personal-health-record messaging system was used to deliver patients in the intervention group (n = 552) both personalized electronic messages from their primary care physician and a web link to the 'Your Disease Risk' personalized risk assessment tool.[5] The control group (n = 551) did not receive electronic messages. At 1 month of follow-up, CRC screening rates were higher in the intervention group compared with the control group (8.3% versus 0.2%, respectively; P <0.001); however, this difference was not sustained at the 4-month primary end point (15.8% versus 13.1%; P = 0.18). Despite targeting a highly-motivated and Internet-competent population who had previously chosen to create an online 'MyHealth' portal,[6] only 54% of the intervention group actually viewed the electronic message within 4 months and only 16% accessed and completed the online risk assessment tool. In comparison to individuals who only viewed the electronic message, patients in the subgroup that completed the online risk assessment tool were more likely to complete screening (15% versus 30%, respectively; P = 0.06). The extent to which screening behavior varied among individuals identified to be at increased or decreased risk of CRC remains unknown. Apart from the limited effectiveness of this intervention, even among a highly-motivated patient population, the requirement that the patient have Internet access and be Internet competent means that it is unlikely to be of benefit among patients with the greatest need, namely those with limited finances and/or with low literacy.
Despite the modest results of the two studies, their approaches highlight the potential of information technology (IT) for addressing various barriers to CRC screening. For example, IT can be used to identify patients in need, generate reminders and track outcomes through the use of an integrated electronic health record, and deliver personalized messages via the Internet. Unfortunately, the two studies also demonstrate that the use of reminders has limited effectiveness in promoting CRC screening. Together, these studies suggest that interventions that replace the use of face-to-face interactions with a health-care provider to exchange information, explore patient preferences, clarify values and share in the decision-making process are unlikely to entice patients to participate in screening.[7] The extent to which system-level barriers, such as transportation or concern over bowel preparation, might have contributed to the relative lack of effectiveness is also unclear; a growing body of literature suggests that successful patient navigation programs can increase the completion of screening colonoscopy, with rates as high as 66–91% among diverse patient populations.[8,9] Despite their limited impact as stand-alone strategies, both of the highlighted interventions have tremendous dissemination potential in heath-care settings with pre-existing electronic patient health records or integrated personal-health-record messaging systems. By contrast, the relative cost-effectiveness of either intervention in settings lacking the necessary IT resources is debatable. Furthermore, how these findings will translate to other patient populations (such as those who are uninsured or underinsured) and to health-care settings with suboptimal CRC screening programs remains unknown.
In summary, the findings of Cameron et al.[3] and Sequist et al.[4] provide new perspectives on the potential utility of electronic patient health records and the Internet to facilitate the delivery of low-cost targeted interventions to promote CRC screening. The limited effectiveness of the interventions, however, highlights the need for a multidimensional approach that addresses barriers at the level of the patient, the provider and the system. Further work is still required to increase CRC screening rates in the USA and, as the poet Robert Frost wrote, we have "miles to go before [we] sleep".[10]