Original Article

Efficacy of communication skills training on colorectal cancer screening by GPs: a cluster randomised controlled trial I. AUBIN-AUGER, MD, PHD, PROFESSOR, Departement de M edecine G en erale, Universit e Paris Diderot, Sorbonne Paris , Paris, and EA Recherche Clinique Coordonn ^pital, M Cite ee Ville-Ho ethodologies et Soci et e (REMES), Paris,  C. LAOUENAN, MD, PHD, DOCTOR, IAME, UMR 1137, INSERM, Paris, Universit e Paris Diderot, Sorbonne Paris Cit e, ^pital Bichat, Service de Biostatistique, Paris, J. LE BEL, MD, DOCTOR, D epartement de M edecine Paris, and AP-HP, Ho G enerale, Universite Paris Diderot, Sorbonne Paris Cit e, Paris, IAME, UMR 1137, INSERM, Paris, and Universit e Paris Diderot, Sorbonne Paris Cite, Paris, A. MERCIER, MD, PHD, PROFESSOR, D epartement de M edecine G en erale, Rouen University, Rouen, D. BARUCH, MD, DOCTOR, D epartement de M edecine G en erale, Universit e Paris Diderot, ^pital, M Sorbonne Paris Cite, Paris, and EA Recherche Clinique Coordonn ee Ville-Ho ethodologies et Soci et e (REMES), Paris, J.P. LEBEAU, MD, PROFESSOR, D epartement de M edecine G en erale, Tours University, Tours, A. YOUSSEFIAN, MD, DOCTOR, Departement de Medecine G en erale, Universit e Paris Diderot, Sorbonne Paris Cit e, Paris, T. LE TRUNG, MD, DOCTOR, Prevention Sant e Val d’Oise, Cergy Saint Christophe, France, L. PEREMANS, MD, PHD, PROFESSOR, Department of Public Health, Vrije Universiteit Brussel, Brussels, and Department of Primary and Interdisciplinary Care, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, & P. VAN ROYEN, MD, PHD, PROFESSOR, Department of Primary and Interdisciplinary Care, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium  AUBIN-AUGER I., LAOUENAN C., LE BEL J., MERCIER A., BARUCH D., LEBEAU J.P., YOUSSEFIAN A., LE TRUNG T., PEREMANS L. & VAN ROYEN P. (2015) European Journal of Cancer Care Efficacy of communication skills training on colorectal cancer screening by GPs: a cluster randomised controlled trial Colorectal cancer (CRC) mass screening has been implemented in France since 2008. Participation rates remain too low. The objective of this study was to test if the implementation of a training course focused on communication skills among general practitioners (GP) would increase the delivery of gaiac faecal occult blood test and CRC screening participation among the target population of each participating GP. A cluster randomised controlled trial was conducted with GP’s practice as a cluster unit. GPs from practices in the control group were asked to continue their usual care. GPs of the intervention group received a 4-h educational training, built with previous qualitative data on CRC screening focusing on doctor–patient communication with a follow-up of 7 months for both groups. The primary outcome measure was the patients’ participation rate in the target population for each GP. Seventeen GPs (16 practices) in intervention group and 28 GPs (19 practices) in control group participated. The patients’ participation rate in the intervention group were 36.7% vs. 24.5% in the control group (P = 0.03). Doctor–patient communication should be developed and appear to be one of the possible targets of improvement patients adherence and participation rate in the target population for CRC mass screening.

Keywords: colorectal cancer screening, communication skills, general practice, educational intervention.

Correspondence address: Isabelle Aubin-Auger, 40 rue Carnot, 95230 Soisy sous Montmorency, France (e-mail: [email protected]).

Accepted 9 February 2015 DOI: 10.1111/ecc.12310 European Journal of Cancer Care, 2015

© 2015 John Wiley & Sons Ltd

I N TR O DU C TI O N Colorectal cancer (CRC) is the third most common cancer in western countries, including the USA and Europe. CRC mortality remains high, though better survival rates are

AUBIN-AUGER ET AL.

reached if the disease is detected at early stages (Huang et al. 2005). Various screening strategies to detect asymptomatic pre-clinical CRC and adenomas known to precede CRC have been developed worldwide, all beginning at the age of 50 (Huang et al. 2005). Immunochemical faecal occult blood test (iFOBT) has been recommended for population screening by the latest European guidelines for quality assurance for CRC screening in 2011 (Von Karsa et al. 2012). They defined the acceptable level of uptake participation rate over 45% and the desirable level over 65%. In France, mass screening has been established since 2008, with a two-step strategy: a gaiac FOBT (gFOBT) delivered by general practitioners (GPs) every 2 years for the average risk population and a colonoscopy performed when the gFOBT test is positive or for high-risk patients (Goulard et al. 2008). Patients between 50 and 74 years of age receive an invitation from a local screening organisation every 2 years as part of the national health system. They are urged to ask their GP for the test during the next consultation. iFOBT will be soon implemented. Whatever the screening strategy, participation rates remain too low worldwide. Most of the European countries don not reach the acceptable participation rate defined by the guidelines. In France, 68% of the eligible population had not been screened in the past period (Goulard et al. 2008). To understand this problem, we conducted comprehensive studies to explore doctor and patient perspectives using GPs’ focus groups and patients’ interviews (Aubin-Auger et al. 2011). French GPs experienced few difficulties with early adopters coming and asking for the test. With these patients, the main obstacle for the GPs was managing the consultation time as these patients came with several medical issues to be dealt with (Aubin-Auger et al. 2011). Later adopters or non-compliant patients were much more of a challenge to convince. To have a better overview of the gFOBT delivery in real life, 35 consultations with 9 GPs in which patients came and asked for gFOBT were audio-recorded to analyse doctor–patient communication (Aubin-Auger et al. 2013b). Based on a Roter analysis (Roter 1977; Roter & Larson 2002), the data showed that the communication style during these consultations was not patient-centred. A content analysis was also performed, leading us to develop a checklist for GPs that includes all of the important clinical items to be discussed during the consultation. Educational material and a training programme were developed based on triangulation of the previous qualitative data focusing on the six components of the patient-centred clinical method (Stewart et al. 2003). Using a cluster randomised controlled trial (cRCT), we tested the hypothesis that implementation of a training 2

course among GPs focusing on communication skills and based on previous qualitative data would increase the delivery of gFOBT among the target population of each participating GP and the number of GPs improving patients’ participation rate in CRC screening and reaching the acceptable and desirable uptake rate defined by the European guidelines. METHODS Design and setting This trial aimed to assess the efficacy of communication skills training on CRC screening by GPs, as measured by the rate of gFOBTs delivered and those actually performed among the target population of each participating GP. This cluster randomised, controlled, primary care trial used a two-arm parallel design. The rationale for choosing a cluster design was to minimise ‘contamination’ between practitioners, i.e. the phenomenon of GPs who had taken the training course on new communication skills passing on the knowledge to physicians at the same practice who had not taken the training course. Thus, the cluster and unit of randomisation was the GP practice. All practices in the Val d’Oise d epartement (county) in France were eligible to participate. This specific location in the suburbs of Paris was chosen because the local screening organisation (PSVO, Pr evention Sant e Val d’Oise) had decided to be involved in a research project to improve the CRC screening participation rate. This agreement allowed us to share all of their data, including a list of all GPs and their addresses. The target population for the screening in this area included 273 882 adults from 50 to 74 years of age. We excluded 11 900 adults (4.3%) belonging to the high-risk population for CRC as defined by French national guidelines (HAS 2004). Thus, the eligible population for our study totalled 261 982 adults. The mean CRC screening participation rate in this d epartement in 2011 was 27.6% (with a French national average of 32%) (InVS 2013). This trial was approved as a non-interventional study by the French Ethics Committee of Ile-de-France IV (no. IRB 00003835) and registered in ClincalTrials.gov (#NCT02012829).

Randomisation All GPs (n = 878) at practices (n = 598) in the Val d’Oise partement, with the exception of GPs at investigators’ de practices and doctors exclusively using complementary therapies (acupuncture therapy, homoeopathy, etc.) (n = 32 GPs), received a letter explaining the study in June © 2015 John Wiley & Sons Ltd

Improve patients’ participation rate in CRC mass screening

2011. Among 585 practices, an independent biostatistician randomised 50 practices per arm in October 2011 based on a computer-generated randomisation list. This resulted in 69 GPs in the intervention group and 73 GPs in the control group. Two team members (IA and AY) telephoned all of the GPs at the randomised practices at the end of October 2011. The involvement of practices and GPs in the trial is summarised in Figure 1.

Intervention General practitioners from practices in the control group were asked to continue their usual care, as if they were

not participating in this trial. The whole procedure is the same all over the country, i.e. all adults aged 50–74 of each GP’s target population received a letter every 2 years from the local cancer prevention association encouraging them to visit their GP and ask for the test. Patients must then send the test and a completed identity form, contained in an envelope to a central laboratory. If they do not, they received a reminder 3 months later. Finally, another 3 months later, those who did not go to see their GP received a test directly at home. General practitioners from both groups had to send their patient list to the research team to calculate their eligible population. They were also asked to list

All practices and GPs in Val d’Oise département, France (n = 598 practices; 878 GPs)

Excluded (n = 13 practices; 32 GPs) ♦ practices of GP investigators

Enrollment

♦ GPs exclusively

using complementary therapies

Assessed for eligibility (n = 585 practices; 846 GPs)

Randomized (n = 100 practices; 142 GPs)

Allocation Allocated to intervention (n = 50 practices; 69 GPs) ♦ Accepted to participate (n = 23 practices; 24 GPs) • Received allocated intervention (n = 17 practices; 18 GPs) ♦

Allocated to control group (n = 50 practices; 73 GPs) ♦ Accepted to participated (n = 35 practices; 50 GPs) ♦

Did not participate (n = 15 practices; 23 GPs)

Did not participate/receive intervention (n = 27 practices; 45 GPs)

Follow-Up Lost to follow-up (n = 1 practices; 1 GPs)

Lost to follow-up (n = 16 practices; 22 GPs)

Did not collect data

Did not collect data

Analysis Analysed (n = 16 practices; 17 GPs)

Analysed (n = 19 practices; 28 GPs)

GP, general practitioner Figure 1. Study flow chart.

© 2015 John Wiley & Sons Ltd

3

AUBIN-AUGER ET AL.

all gFOBT screening delivered during the 7-month study period. The list of gFOBTs performed per GP was supplemented with the data sent by the local screening organisation (PSVO). In the intervention group, participating GPs received a 4-h training course focusing on patient-centred care based on the six components of the patient-centred clinical method. Two different scenarios for a video aiming to improve communication with patients were developed: one for a compliant patient, another for a non-compliant patient. Two videos were made featuring a doctor and a simulated patient (more details is available in Aubin-Auger et al. 2013a). Interactive methods were used to deliver educational training, including role playing and presentation of the video followed by an interactive discussion (Berkhof et al. 2011). Three identical sessions, involving five to seven participants, were organised during the same week in December 2011. To limit the scope of investigation solely to the exploration of doctor–patient communication, the GPs from both the intervention and control groups received a written memo summarising the main clinical items that needed to be checked before delivering the gFOBT (eligibility, technical explanations, etc.). A memo summarising the main communication skills was delivered only to the GPs of the intervention group. Although physicians allocated to the intervention arm were aware of the group they were part of, patients and outcome assessors remained blind to the allocation.

Outcome measures The primary outcome measure was each GP’s patient CRC screening participation rate in the target population during the 7-month study period. For the intervention group, this period started immediately after the training course. The participation rate was the number of patients who performed a gFOBT during the study period divided by each GP’s eligible population (adults from 50 to 74 years of age minus adults excluded from the screening) during that same period. Since it is recommended that CRC be performed at least every 2 years, the total eligible population for 2 years was extrapolated from the 7-month study period. The secondary outcomes were as follows: the number of GPs reaching a patient participation rate of 65% during the study period, consistent with the desirable uptake rate defined by European guidelines for quality assurance in CRC screening (Von Karsa et al. 2012); the number of GPs reaching a participation rate of 45% consistent with the 4

acceptable uptake rate specified by the same guidelines; and the number of gFOBTs delivered by GPs during the study period, the number of gFOBTs performed per GP in eligible patients during the study period and the difference between the two because we posited that the training course could lower this difference. All of the outcome data were measured at the individual GP level.

Sample size To detect an increase in the patient participation rate from 30% in the control group to 40% in the intervention group 7 months after the intervention, with 90% power at a 5% significance level, 477 patients were required for each of the study groups. Sample size was determined using a method that takes the cluster effect into account. An intra-cluster correlation coefficient (CIC) of 0.05, commonly used in primary care studies, was assumed, with a process variable as primary criterion (Campbell et al. 2005; Cals et al. 2009). Mean cluster size was calculated based on a pilot study with half of the GPs from the same d epartement (i.e. 1.5 GPs per practice). Based on this, a design effect was calculated as 1 + (mo 1) 9 CIC with mo being the number of eligible patients per practice. With the number of eligible patients per GP being 300, mo was 450. In addition, the design effect was 23.5. The sample sizes in the intervention and control groups were then multiplied by this design effect to give an effective sample size (Eldridge et al. 2006). The number of patients per arm was 477 9 23.5 = 11 210 and the number of practices per arm was 11 210/mo = 25. Finally, the number of GPs required per arm was 25 9 1.5 = 38. Thus, 50 practices were randomised per arm to ultimately enrol 25 practices per arm (with 50% refusal expected).

Statistical analysis The effect of the intervention on the primary outcomes (patient participation rate) was analysed taking into account of the design effect due to cluster sampling (i.e. within practice correlation) using generalised linearmixed effects model with group (intervention vs. control) as fixed effect and practice as random effect (using SAS procedure GLIMMIX). The same analysis was used to measure the effect of the intervention on the secondary outcomes. All tests were two-sided and the alpharisk was set at 0.05. SAS 9.3 software (SAS Institute Inc, Cary, NC) was used for statistical computations. Data are shown as mean  standard deviation (SD) for continuous variables, and number and % for categorical variables. © 2015 John Wiley & Sons Ltd

Improve patients’ participation rate in CRC mass screening

RESULTS GPs and patients characteristics Among the 50 practices randomised per study group, 23 practices (24 GPs) accepted to take part in the intervention group and 35 practices (50 GPs) in the control group. There were multiple reasons for declining the invitation: no interest in the topic, the impression of already doing their best, too much work and no time for continuous medical education (CME), and previous training (Fig. 1). A total of 18 GPs were trained during three identical courses. Six GPs who had initially accepted did not come in the end because they were too busy or ill. At the end of the study period (7 months from December 10, 2011 to July 9, 2012), 17 GPs (16 practices) in the intervention group and 28 GPs (19 practices) in the control group were still active and sent in their data (Fig. 1). Finally, 45 GPs (35 practices) ultimately participated in the trial. Among these 45 GPs, the mean cluster size was 1.3. The mean cluster size in the entire departement was 1.5 (878 GPs and 598 practices). The baseline characteristics of participating GPs in both groups were comparable in terms of sex, age, year of practice setup, certification and location (Table 1). They were also comparable to French national GP characteristics in terms of sex and age. The GPs were predominantly male (62%) and worked at group practices (69%) in urban areas (93%), with an overall mean age of 51 (9.8) years.

Table 1. Baseline characteristics of all 45 participating general practices in Val d’Oise, France Intervention group (n = 17 GPs) GP characteristics at baseline Female GPs, n (%) 5 Age of GPs in 51.0 years (SD) Years since GP 19.6 certification (SD) Years since setup of GP 18.4 practice (SD) Number of GPs per 2.2 practice, (SD) Number of GPs at practices, n (%) One 7 Two 4 More than two 6 Practice with 14 secretary, n (%) Location, n (%) Suburban 1 Urban 16

Control group (n = 28 GPs)

(29.4) (12.5)

12 (42.9) 51.3 (8.1)

(14.2)

21.4 (8.6)

(14.4)

18.4 (9.2)

(1.3)

2.4 (1.0)

(41.2) (23.5) (35.3) (82.4)

7 8 13 23

(5.9) (94.1)

2 (7.1) 26 (92.9)

(25.0) (28.6) (46.4) (82.1)

In the intervention group, the GPs delivered 415 gFOBTs among their eligible population, compared to 549 in the control group, during the study period. In the intervention group, the number of gFOBTs performed in the GPs’ eligible population was 383, compared to 567 in the control group, during the study period. During the same period, 20 652 tests were performed in the entire department, within a population of 261 982 eligible patients (273 882 adults from 50 to 74 years of age minus 4.3% non-eligible adults). This meant a participation rate of 7.9% for the population to be screened over 7 months, or an extrapolated participation rate of 27.1% over 2 years. The number of eligible patients per GP during the study period was 64 (46) in the intervention group and 71 (28) in the control group (P = 0.6). Mean age of eligible patients was 60.9 years (6.8) and 48.9% were male in the intervention group, and 59.1 years (6.8) and 54.9% in the control group.

Primary outcome The patient participation rate per GP among eligible patients during the study period was 36.7 (20.3) in the intervention group and 24.5 (10.1) in the control group (P = 0.03) (Table 2). Of note, the patient participation rate per GP over the 2 years before the study period (from December 10, 2009 to December 9, 2011) was 24.3 (9.1) in the intervention group and 24.7 (9.5) in the control group (P = 0.8). Secondary outcomes One GP in the intervention group reached a patient participation rate of 65% during the study period, whereas none did so in the control group (P = 0.2). Four GPs in the intervention group reached the acceptable participation rate of 45%, as did two GPs in the control group (P = 0.3). The difference between the number of gFOBTs delivered and those actually performed during the study period was 1.9 (18.9) in the intervention group and 0.6 (4.3) in the control group (P = 0.5) because some gFOBTs performed were probably delivered before the study period. Details on the numbers of gFOBTs delivered and those actually performed are provided in Table 2. D I S C U S S I ON Main findings

GP, general practitioner; SD, standard deviation.

© 2015 John Wiley & Sons Ltd

An intervention focusing on doctor–patient communication showed efficacy in improving patient participation. 5

AUBIN-AUGER ET AL.

Table 2. Primary and secondary outcomes after 6 months in each group

Primary outcome Patient participation rate per GP among eligible patients during study period, mean (SD) Secondary outcomes Number of GPs with patient participation rate of 65% during the study period, n (%) Number of GPs with patient participation rate of 45% during the study period, n (%) Number of gFOBTs delivered during study period, mean (SD) Number of gFOBTs actually performed per GP among eligible patients during study period, mean (SD) Difference between gFOBTs delivered and those actually performed per GP during the study period, mean (SD)

Intervention group (n = 17 GPs)

Control group (n = 28 GPs)

P

36.7 (20.3)

24.5 (10.1)

0.03

1 4 24.4 22.5

0 2 19.6 20.3

0.2 0.3 0.3 0.6

(5.9) (23.5) (24.1) (15.9)

1.9 (18.9)

(0) (7.1) (9.0) (10.3)

0.6 (4.3)

0.5

GP, general practitioners; gFOBT, gaiac faecal occult blood test; SD, standard deviation.

Using patient-centred care and giving GPs skills to motivate non-compliant patients appeared to be good ways of increasing their CRC screening participation rate.

Study limitations According to a systematic review undertaken by Berkhof et al. (2011), training programmes are effective if they last for at least 1 day. There were many difficulties in getting GPs involved in the intervention group, so a half-day course was organised. The choice of scheduling the intervention during the month of December was probably a mistake as practices have a heavy workload at that time of year due to winter epidemics. GPs were already too busy and did not take time for CME. Given the high dropout rates of practices after randomisation, we could not exclude the possibility of a selection bias. As always in cRCT, participants were recruited into the trial after clusters were randomised into intervention or control groups, we could not exclude the possibility of a recruitment bias. An intention-to-treat analysis was not done because the data on GPs who were randomised but did not participate in the study have not been collected. Indeed in cRCT, the situation is more complex compared to individually randomised trials (Giraudeau & Ravaud 2009). No statistical significance was found for the secondary outcomes, probably due to a lack of statistical power. The hypothesis that improving communication skills should reduce the difference between the numbers of tests delivered and those actually performed could not be proved. Though our calculations called for 25 practices per arm, 16 practices actually took part in the intervention group and 19 actually took part in the control group. 6

Strengths This cRCT is particularly original because the intervention content focused on communication skills between GPs and their patients and was based on qualitative data previously collected from patients and doctors in primary care settings (Aubin-Auger et al. 2011, 2013b). Patient-centred care, though recognised as a component of high-quality health care, is not yet taught at every French medical university. Participants in the intervention group showed interest in this original training course, exploring patient perspectives, in particular with non-compliant patients. It was more of a challenge to convince patients who came to consultations without any request for the screening, and the second video gave them the skills and specific arguments to make progress in this area. Improving communication skills could also prove useful for GPs in screening or preventive care to reach shared decisions (Austoker et al. 2012). Another intervention towards GPs has been performed in France and showed the same trends, but all of the participating GPs were volunteers (Thu-Thon et al. 2013).

Implementation Even though such an intervention showed efficacy, the desirable uptake rate of over 65% specified by European guidelines on quality assurance in CRC screening and diagnosis has not yet been reached (Von Karsa et al. 2012). Only one doctor in the intervention group reached this rate. The way such an intervention is implemented needs to be discussed. In a French study on GPs’ desire to participate in primary care research, approximately 30% of the participants were willing to be investigators. Preventive care was one of their fields of interest, albeit more for cardiovascular diseases than for cancer screening © 2015 John Wiley & Sons Ltd

Improve patients’ participation rate in CRC mass screening

(Supper et al. 2011). Other educational materials, such as online videos, already tested for other topics such as antibiotic prescriptions, should be promoted to achieve better implementation.

Interventions to increase the CRC screening participation rate Main interventions focus on patients and not on doctors such as patient reminders or phone calls. Interventions can be tailored to the unique characteristics of a single person or untailored for a target population. Printed untailored interventions delivered by clinical practices, organisations or screening programmes were effective for gFOBT CRC screening (Sabatino et al. 2012). Interventions to reduce structural barriers to gFOBT screening, such as time, distance or administrative procedures, were shown to be sufficiently effective and were thus recommended (Sabatino et al. 2012). Providers’ attitudes are modifiable factors for improving CRC screening and are an interesting target (Gimeno Garcia 2012). Le Breton et al. (2012) showed that targeted actions to improve CRC screening should be mainly directed to GPs. Some GPs did not participate in the RCT because they did not feel concerned. Incentives or other motivations should be developed to improve GP participation. In France, a new payment for performance (P4P) scheme for primary care physicians was gradually introduced in 2009 (Saint-Lary et al. 2012). In 2013, this P4P scheme was extended to various indicators (percentage of hypertensive patients and diabetics who reached the targets, etc.) but none involve CRC screening. Provider assessment and feedback is recommended with sufficient evidence to increase delivery and promotion of CRC screening (Sabatino et al. 2012). Reminders directed towards health care providers are also efficient (Weller & Campbell 2009; Siddiqui et al. 2011). Multi-level interventions focusing on the various determinants involved in the screening process, i.e. patients, health care providers, health care organisation, information sources (the media, family, doctors, etc.), should prove useful in achieving this rate (Weller & Campbell 2009; Aubin-Auger et al. 2011). Continuous medical education is not yet mandatory in France. Support from national academies and health care organisations will probably prove useful in developing CME and research in primary care on this topic. The support from the local PSVO organisation has been perfect and could be expanded.

© 2015 John Wiley & Sons Ltd

Invitation strategies Different strategies are used to involve the target population in mass screening programmes: through GPs or through an invitation by direct mail including a sampling kit. Organised screening programmes, based on invitation letters or on GP involvement, were consistently effective in increasing participation compared to spontaneous screening. A Flemish study showed that inviting people by a direct mail invitation, and including a faecal sampling set (iFOBT), resulted in much higher participation rates than inviting people through the GP (van Roosbroeck et al. 2012). Other data were contradictory, in an Australian study, associating a GP to an invitation to screen achieved better participation and re-participation than an invitation from a centralised screening unit (Zajac et al. 2010). A previous French study explored different recruitment methods and concluded that such a programme required GPs’ involvement and close coordination between practitioners and health care insurance agencies (Launoy et al. 1996). Our data showed that doctors’ communication skills are a good target to increase participation rates. Direct invitations do not allow this opportunity. In conclusion, in this cluster randomised intervention study, it was shown that an educational programme focusing on doctor–patient communication was effective at increasing patient participation. Using patient-centred care and giving GPs the skills to motivate non-compliant patients appeared to be good ways to increase their CRC screening participation rate. Nonetheless, participation rates have remained low. Feasible multi-level interventions should be developed to improve screening programme results.

A UT H OR CO N TR I B U T IO N S IAA contributed to the original idea and conception of the study. IAA, AM, JL, AY, LP and PVR conceived the study protocol and participated in its design and coordination and helped to draft the manuscript. CL participated in the design of the study and performed the statistical analysis and helped to draft the manuscript. TLT, DB, JPL participated in participants recruitment and coordination. All authors read the draft, made suggestions and approved the final manuscript.

C ON F L I C T O F I N TE R E S T The authors declare that they have no competing interests.

7

AUBIN-AUGER ET AL.

F U N DI N G

ACKNOWLEDGEMENTS

The study was realised with the support of the French local mass screening organisation Prevention Sant e Val d’Oise (PSVO) for the education courses and data collection.

Dani ele Delobel for secretarial work, French local mass screening organisation Pr evention Sant e Val d’Oise (PSVO) for the education courses and data collection, and Stephen Martin (Tradensys) for English language.

R E F E R E N CE S Aubin-Auger I., Mercier A., Lebeau J.P., Baumann L., Peremans L. & van Royen P. (2011) Obstacles to colorectal screening in general practice: a qualitative study of GPs and patients. Family Practice 28, 670–676. Aubin-Auger I., Mercier A., Le Bel J., Bombeke K., Baruch D., Youssefian A., Lebeau J.P., Le Trung T., Peremans L. & van Royen P. (2013a) From qualitative data to GP training on CRC screening. Journal of Cancer Education 28, 439–443. Aubin-Auger I., Mercier A., Mignotte K., Lebeau J.-P., Bismuth M., Peremans L. & van Royen P. (2013b) Delivering the faecal occult blood test: more instructions than shared decisions. A qualitative study among French GPs. The European Journal of General Practice 19, 150–157. Austoker J., Giordano L., Hewitson P. & Villain P. (2012) European guidelines for quality assurance in colorectal cancer screening and diagnosis. First Edition– Communication. Endoscopy 44(Suppl. 3), SE164–SE185. Berkhof M., van Rijssen H.J., Schellart A.J., Anema J.R. & van der Beek A.J. (2011) Effective training strategies for teaching communication skills to physicians: an overview of systematic reviews. Patient Education and Counseling 84, 152–162. Le Breton J., Journy N., Attali C., Le Corvoisier P., Brixi Z., Bastuji-Garin S. & Chevreul K. (2012) Improving participation in colorectal cancer screening: targets for action. Preventive Medicine 55, 488–492. Cals J.W., Butler C.C., Hopstaken R.M., Hood K. & Dinant G.J. (2009) Effect of point of care testing for C reactive protein and training in communication skills on antibiotic use in lower respiratory tract infections: cluster randomised trial. British Medical Journal 338, b1374. Campbell M.K., Fayers P.M. & Grimshaw J.M. (2005) Determinants of the intracluster correlation coefficient in cluster randomized trials: the case of implementation research. Clinical Trials 2, 99–107.

8

Eldridge S.M., Ashby D. & Kerry S. (2006) Sample size for cluster randomized trials: effect of coefficient of variation of cluster size and analysis method. International Journal of Epidemiology 35, 1292–1300. Gimeno Garcia A.Z. (2012) Factors influencing colorectal cancer screening participation. Gastroenterology Research and Practice 2012, 483417 doi: 10.1155/ 2012/483417 Giraudeau B. & Ravaud P. (2009) Preventing bias in cluster randomised trials. PLoS Medicine 6, e1000065. Goulard H., Boussac-Zarebska M., AncellePark R. & Bloch J. (2008) French colorectal cancer screening pilot programme: results of the first round. Journal of Medical Screening 15, 143–148. HAS (2004) Indications for lower gastrointestinal endoscopy (excluding population screening). Available at: http://www.has-sante.fr/portail/upload/ docs/application/pdf/Endoscopy_guideline s.pdf (accessed 5 January 2013). Huang C.S., Lal S.K. & Farraye F.A. (2005) Colorectal cancer screening in average risk individuals. Cancer Causes and Control 16, 171–188. InVS (2013) Taux de participation au programme de depistage organise du cancer colorectal 2011-2012. Available at: http://www.invs.sante.fr/Dossiers-th ematiques/Maladies-chroniques-et-trau matismes/Cancers/Evaluation-des-progr ammes-de-depistage-des-cancers/Evaluat ion-du-programme-de-depistage-du-can cer-colorectal/Indicateurs-d-evaluation/Ta ux-de-participation-au-programme-de-dep istage-organise-du-cancer-colorectal-20112012 (accessed 5 January 2013). Launoy G., Herbert C., Vallee J.P., Desoubeaux N., Reaud J.M., Ollivier V., Bouvier V., Flachs A., Arsene D., Valla A. & Gignoux M. (1996) Mass screening for colorectal cancer in France. Experience in 165,000 people in the department of Calvados. Gastroenterologie Clinique Et Biologique 20, 228–236. van Roosbroeck S., Hoeck S. & van Hal G. (2012) Population-based screening for colorectal cancer using an immunochemical faecal occult blood test: a comparison of two invitation

strategies. Cancer Epidemiology 36, e317–e324. Roter D.L. (1977) Patient participation in the patient-provider interaction: the effects of patient question asking on the quality of interaction, satisfaction and compliance. Health Education Monographs 5, 281–315. Roter D. & Larson S. (2002) The Roter interaction analysis system (RIAS): utility and flexibility for analysis of medical interactions. Patient Education and Counseling 46, 243–251. Sabatino S.A., Lawrence B., Elder R., Mercer S.L., Wilson K.M., DeVinney B., Melillo S., Carvalho M., Taplin S., Bastani R., Rimer B.K., Vernon S.W., Melvin C.L., Taylor V., Fernandez M. & Glanz K. (2012) Effectiveness of interventions to increase screening for breast, cervical, and colorectal cancers: nine updated systematic reviews for the guide to community preventive services. The American College of Preventive Medicine 43, 97–118. Saint-Lary O., Boisnault P., Naiditch M., Szidon P., Duhot D., Bourgueil Y. & Pelletier-Fleury N. (2012) Performance scores in general practice: a comparison between the clinical versus medicationbased approach to identify target populations. PLoS ONE 7, e35721. Siddiqui M.R., Sajid M.S., Khatri K., Kanri B., Cheek E. & Baig M.K. (2011) The role of physician reminders in faecal occult blood testing for colorectal cancer screening. European Journal of General Practice 17, 221–228. Stewart M., Brown J., Weston W., McWhinney I., McWilliam C. & Freeman T. (2003) Patient-Centered Medicine: Transforming the Clinical Method. Radcliffe Medical Press, Abingdon, UK. Supper I., Ecochard R., Bois C., Paumier F., Bez N. & Letrilliart L. (2011) How do French GPs consider participating in primary care research: the DRIM study. Family Practice 28, 226–232.  Thu-Thon E., Charles R., Froger P.,  & Trombert B. Meyrand B., Pare E. (2013) Impact d’une formation pour les medecins generalistes sur la participation au depistage du cancer colorectal. Sante Publique 25, 775–783.

© 2015 John Wiley & Sons Ltd

Improve patients’ participation rate in CRC mass screening

Von Karsa L., Patnick J. & Segnan N. (2012) European guidelines for quality assurance in colorectal cancer screening and diagnosis. First Edition-Executive summary. Endoscopy 44(Suppl. 3), SE1– SE8.

© 2015 John Wiley & Sons Ltd

Weller D.P. & Campbell C. (2009) Uptake in cancer screening programmes: a priority in cancer control. British Journal of Cancer 101(Suppl. 2), S55–S59. Zajac I.T., Whibley A.H., Cole S.R., Byrne D., Guy J., Morcom J. & Young G.P.

(2010) Endorsement by the primary care practitioner consistently improves participation in screening for colorectal cancer: a longitudinal analysis. Journal of Medical Screening 17, 19–24.

9

Efficacy of communication skills training on colorectal cancer screening by GPs: a cluster randomised controlled trial.

Colorectal cancer (CRC) mass screening has been implemented in France since 2008. Participation rates remain too low. The objective of this study was ...
157KB Sizes 0 Downloads 6 Views