Numerous organisational and cognitive factors influencing inadequate coverage in the actual practice of mass screening for CRC. These factors are derived from both patients and healthcare and administration professionals. The low participation is partly due to a lack of awareness of both the illness itself and the early detection programmes, but also to the existence of barriers for the conduct and results of the tests. It is essential to inform the population in question about the magnitude of CRC, the importance of early detection, the benefits and risks of participating in this type of programme and the need to coordinate and involve the different health professionals and institutions that participate directly or indirectly in a screening program.
Direct recommendation by the family physician has been described as one of the strongest predictors for the performance of CRC screening, while the non-involvement of this level of care in the recommendations is one of the main reasons for it not being carried out [36, 37]. Data published in our sector show that 89% of subjects would accept CRC screening if their primary care physician or nurse suggested it; a percentage that is very different from the data presented in the current population programmes . On the other hand, most of the eligible population in countries with a long history of CRC screening have shown that they have never received such a recommendation . The reality of the primary care professional’s offices with an overload of care, preventive and bureaucratic tasks influence the poor level of recommendation for CRC screening in the target population attending the clinic for other reasons.
There are clinical studies in place that support the effectiveness of electronic reminders in clinical practice. Nease et al. found a significant increase of 9% in terms of the performance of FOBT, despite a low rate of revision for electronic alerts (30%) . Sequist et al. found an increase in screening rates in those patients who attended the surgery on more than two occasions during the study period, although the difference was not significant, in part due to very high baseline screening rates already in existence and also owing to the fact that the colonoscopy was the test of choice of physicians when recommending screening, with an uptake rate of only 50% of patients . Nease and Sequist evaluate the acceptance and integration of reminders into medical practice with a good general level of acceptance. However, there are certain limitations, such as the moderate suitability of alerts activated in patients considered candidates for screening, possibly generating a tendency to wilfully overlook the reminders, or see them as an interference in the course of medical visits owing to care overload.
The following are worth mentioning as possible limitations of this study:
The selection of the CRCSP target population is based on data from patients included in Catalonia’s Registro Central de Asegurados (Registry of Users of the Catalan Health-care System). The percentage of patients on this register that are assigned to a PCC, and would therefore be invited to participate in the programme when it starts screening, but in actual practice reside at another address or attend another centre, accounts for 19% of the study population, much higher than the average of Catalonia, which was 8.1% according to 2012 figures. Alerts cannot be activated in the medical records of these subjects, as they do not have a physician assigned to the centre that will be participating in the screening program. This may involve a certain level of selection bias, but there is no reason to believe that this population attending a different centre to the one they are assigned may have some distinguishing feature in relation to the study groups and in any case, the control and intervention groups are distributed on a random basis.
While the intervention is directed at the population receiving care, this represents the majority of the assigned population as the duration of the intervention is one year. In 2011, 69% of patients aged between 50 and 69 made at least one visit to their centre involved in the study, where the overall average in Catalonia stands at 71%.
Losses during follow-up: changes of address, institutionalisation or death may occur during the course of the study. Any of these scenarios will be considered as the screening having not taken place.
External validity: This involves a study of urban population, but since the use of EMR is used across the board in primary care in Catalonia, no differences in the effectiveness of electronic reminders are forecast according to the scope of work.
Contamination between professionals: Since the unit of randomisation is the physician, certain contamination could occur between centre professionals. In order to minimise this, a training session on the computerised tool exclusively for professionals in the intervention group is provided. The decision to randomise by medical professional was made by significant socio-demographic differences existing in the reference population of the study centres and by the differences in basal participation found in other centres already screened in the same field, exceeding 10% on occasion.
The CRC screening programmes in Spain are population-based, providing access to the target population, and biennial iFOBT is the test that has been selected, which has shown better levels of acceptance and participation among the population. On the other hand, health professionals from the PCC have a long history in the use of EMR, with universal coverage of the population. In light of this, we are considering the introduction of a specific reminder in the primary care EMR of the target population for an early detection programme for CRC. The healthcare professional will provide the identification and recommendation directly to the patient when he/she attends his/her health professional for any other reason, resulting in increased participation, and thus improving its cost-effectiveness and quality indicators.