Introduction

Diabetes is a chronic disease that affects more than 500 million people in the world and more than 4 million Canadians.1 While type 1, type 2 and gestational diabetes are the most common, rare monogenic forms of diabetes are well-recognised,2 3 including the maturity-onset diabetes of the young (MODY) affecting 1–5 cases per 10 000.4 Among the 14 types of MODY, the most frequent representing 50% of the MODY is the MODY2 (MIM125851), a rare hereditary disease with an autosomal dominant transmission initially identified in France.5 The prevalence of MODY2 in the European population has been estimated at around 1%6 but can vary from one region to another due to regional variations. In addition, the MODY2 prevalence is highly dependent on both access to genetic testing and the training of healthcare professionals (HCPs) for its diagnosis.7 MODY2 is due to the frequent variant c.676G>A (p.Val226Met, rs148311934) located in the glucokinase (GCK) gene and is characterised by slightly elevated fasting hyperglycaemia but without any long-term vascular complications.8 9 MODY2, therefore, does not require any pharmacological treatment or specific dietary restrictions except during pregnancy in the case of the mother having MODY2, and the fetus does not have any pathological variant of the GCK gene.10 11 However, this monogenic diabetes is often confused with type 1 or type 2 diabetes and would represent 2 –5% of diabetes.10 12 Misdiagnosis exposes pregnant women and fetus to pregnancy complications but also patients with MODY2 to antidiabetic medications.10 This inadequate management results in constraints imposed on patients13 and additional costs for the healthcare system. Currently, in the Saguenay-Lac-Saint-Jean population, a region of the Quebec Province in Canada, a family physician, or a nurse practitioner suspecting MODY2 refers patients to different specialists to finally undergo genetic testing to confirm the diagnosis, and between suspicion and result, an average of 8–14 months elapsed. A genetic test has been developed and clinically validated by our team to detect the frequent GCK c.676G>A variant carriers among patients with suspected MODY2. This PCR test is performed on DNA isolated from buccal mucosa cells collected by patients themselves using a swab. In Canada, genetic testing for the diagnosis of MODY2 is currently reserved for genetic specialists and some physicians. However, access to this genetic test by primary care HCPs, coupled with appropriate training, would improve the diagnosis and management of patients with MODY2 in addition to provide important data on misdiagnosed cases.14 Thus, we hypothesise that implementing a MODY2 screening test in primary care would improve the detection, diagnosis and management of this pathology. Implementation science promotes the adoption and integration of evidence-based practices, interventions and policies into routine healthcare and public health settings to improve the impact on population health.15 Thus, this study aims to evaluate the implementation in primary care of genetic testing for the screening of MODY2 (iMOgene study).

Methods and analysisOverarching study approach and conceptual frameworks

This study includes implementation sciences, pharmacogenetics and diabetes management, three scientific research fields represented by the research team. Indeed, we teamed up with endocrinologists, biologists, genetic counsellors and primary care physicians to conceive the study, define the aims, discuss the study design and secure funding. These coresearchers meaningfully participated in the first steps of the study (phase 1), consisting of a consensual design and validation of the protocol. Thus, during this first phase, the research team validated:

The use of the RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) model16 17 to evaluate the implementation in primary care of the MODY2 screening test.

The training content through video capsules and its evaluation with a pre–post questionnaire according to the Kirkpatrick model.18

The inclusion criteria and the care pathway of patients following the result of the MODY2 screening test.

The clinical settings from primary care for implementing the MODY2 screening test.

According to phase 1 (figure 1), we draw on the RE-AIM model,16 17 a conceptual framework to guide our research into the understanding of patients’ and HCPs’ experiences of using the MODY2 screening test. As previously described,19 20 Reach is considered the representativeness of individuals who are willing to participate in the study and can be appreciated with the absolute number or the percentage of individuals who participated in the study. Effectiveness is considered as the impact of the intervention on outcomes. Adoption refers to the representativeness of settings and intervention agents willing to initiate the proposed research programme. The intervention agents’ fidelity can appreciate the Implementation of the various elements of an intervention’s protocols. Finally, Maintenance is the 6-month impact of a programme on outcomes and the extent to which the programme becomes institutionalised. For each item of the RE-AIM framework, qualitative methods can be employed to complete or to precise the evaluation.

Figure 1

Protocol of the iMOgene study.

Our study also includes training for healthcare providers to anchor, at least, the study design, inclusion criteria, diagnosis, management and care journeys of patients with MODY2. This training will be evaluated with a pre–post questionnaire matching the four Kirkpatrick model levels: Reaction, Learning, Behaviour and Results.18 21 The Kirkpatrick model is considered the best-known worldwide tool for analysing and evaluating the effectiveness and results of training and educational programmes.18 22 It is also used to establish and evaluate short-term and long-term outcomes for implementing an intervention in primary care.23

Finally, our study includes focus groups that will be held at the end of the patients’ inclusion period. We used the RE-AIM model to design the interview guide. Thus, we conduct an implementation pilot study supported by the RE-AIM model.17 19 20

Intervention

The second phase of the study (phase 2) includes training HCPs by online asynchronous and presential trainings. This training includes three video capsules supplying information on (capsule #1) the screening, diagnostic and management of MODY2, (capsule #2) how to realise the sample and (capsule #3) the inclusion criteria and how to propose the study to patients. A pre–post questionnaire to evaluate this training according to the Kirkpatrick model has been designed. This questionnaire includes questions on the training content, the confidence level of HCPs on the diagnosis and the management of MODY2, and the HCPs’ commitment to apply the knowledge they acquired during the training. By measuring changes in knowledge, the pre–post questionnaire will allow the quantification of the learning outcomes (level 2) achieved through the training. In addition, the pre–post questionnaire will also allow the evaluation of the intention of HCPs to apply their new knowledge, which is an indicator of potential behavioural change (level 3). This questionnaire has been reviewed and validated by two HCPs from the research team. All the video capsules and the pre–post questionnaires will be available online with password-protected access. In addition, an accredited 1-hour conference given by an endocrinologist and a genetic counsellor will be offered to all participant HCPs. Once trained, HCPs can include patients during the third phase (phase 3). According to inclusion criteria (see below) and after explaining the study and filling out the identification questionnaire, HCPs propose the study to patients, obtain oral consent from patients and provide them with an oral mucosa cell sample kit. All explanations on how to realise the sample are included in the kit. In addition, patients read and sign the information and consent letter and fill out a sociodemographic questionnaire. This questionnaire, the information and consent form and the samples are then sent to the pharmacogenetics research and transfer laboratory in which the genotyping of the frequent variant c.676G>A of the GCK gene is realised through a TaqMan PCR.24 The genotyping result is sent to the HCP who proposed the study to the patient. Three months after sample receipt, a patient experience survey is sent to the patients. After a 12-month inclusion time, HCPs fill out a survey and participate in semistructured focus groups to evaluate their experience of participating in this study. The fourth phase (phase 4) then concerns the design and validation of guidelines for a scale-up of the study in Quebec.

All positive patients will be referred to the genetics service of the CIUSSS-SLSJ (Centre intégré universitaire de santé et de services sociaux of Saguenay-Lac-Saint-Jean) to confirm the MODY2 in a reference laboratory. Once the MODY2 is confirmed, these patients will be managed by the HCPs following the standard of care recommendations.25

All negative patients could be considered without a MODY2. However, if the MODY suspicion persists, the patient could be referred to the endocrinology service of the CIUSSS-SLSJ to pursue the diagnostic investigation with other rare GCK pathogenic mutations or other genes responsible for other MODY forms.25

Sample kit

The sample kit delivered to patients includes an information and consent form in two copies, a sociodemographic questionnaire, a one-page explanation on how to perform the oral mucosa cells sample, two sterile standard cotton swabs with a protective cap and a prestamped envelope. All the relevant information is available in the sample kit, but if needed, the patient can access additional information on the study through a QR code linked to a website and to a video capsule explaining how to perform the sample. The kit has been tested and validated by three diabetic and volunteer patients before starting the study. Once the buccal mucosa cells sample is performed, the patient should put one exemplary of the signed information and consent form, the filled socio-demographic questionnaire and the two sterile standard cotton swabs in the prestamped envelope sent by regular post. Once they arrive at the laboratory, the swabs are denominated and made available for genotyping. The consent forms and other information provided by the patients are stored in a locked cabinet to protect confidentiality.

Inclusion criteria

This study will be proposed to all physicians and nurse practitioners from two primary care clinics in Quebec, Canada (one public-funded and one private). These clinics have been selected through the professional network of the research team.

Non-pregnant patients included in this study must have fasting blood glucose over or equal to 6.1 mmol/L or a glycated haemoglobin (haemoglobin A1c) level over or equal to 6% and at least one of the following criteria: having a family member already diagnosed with MODY2, or being diagnosed diabetic before age 35, or having a persistent postpartum gestational diabetes without Body Mass Index (BMI) >35 kg/m², age over 40, ethnic origin as First Nation people, African, Asian, Hispanic and Arab, or having a BMI<25 kg/m².25–27

Pregnant patients can be included in the present study if they have a blood glucose over or equal to 5.1 mmol/L or a blood glucose over or equal to 10.0 mmol/L 1 hour after ingestion or over or equal to 8.5 mmol/L 2 hours after ingestion during an oral glucose tolerance test, or a glycated haemoglobin (haemoglobin A1c) level over or equal to 6.1% and at least one of the following criteria: having a family member already diagnosed with MODY2, or being diagnosed diabetic before age 35, or having a BMI<25 kg/m² before pregnancy.2

Non-inclusion criteria

Patients with current oral corticosteroid therapy, antipsychotic medication in progress, polycystic ovary syndrome, cystic fibrosis, anti-GAD (glutamic acid decarboxylase) or anti-islet autoantibodies, history of diabetic ketoacidosis, history of pancreatitis or pancreatic surgery will not be included to avoid confusing conditions.

Data collection and evaluation of the implantation

The patients’ experience survey allows to collect information divided into four items: (1) the influence of their participation, (2) the perception of their care management, (3) the perception of their trust in the research team and (4) the perception of their experience with participating in this study, in agreement with the RE-AIM model. After a 12-month inclusion period, HCPs will fill out an HCP experience survey, which is divided into four items in agreement with the RE-AIM model: (1) the contribution of genetic tests for the diagnosis of MODY2, (2) a long-term evaluation of the training, (3) their trust and relationship with the research team and (4) their opinion on the maintenance in primary care of the access of the genetic test for the screening of MODY2. HCPs will also participate in semistructured focus groups based on both the COREQ (consolidated criteria for reporting qualitative research) and the RE-AIM models to evaluate the facilitating and limiting factors for the implementation in primary care of new genetic testing for the diagnosis of MODY2. During these semistructured interviews, the access of genetic tests for the diagnosis of MODY2, the interactions with patients they have had during the study, the interactions they have had with the research team and how they feel and cope with the end of this study. The interview guidelines will also be flexible and will allow dynamic interactions to favour participant engagement and the richness of data. Each semistructured interview will take approximately 60–90 min and will be audio-recorded. Finally, all additional data or comments from the research group are included in a research team logbook for the entire study duration. This logbook will complete the documentation of the facilitators and barriers of the implementation.

Thus, in agreement with the RE-AIM model, the following data will be collected. Reach: the number of sample kits delivered, the number of received samples for genotyping, the number of pregnant women included compared with the number of total pregnant women with gestational diabetes, the percentage of HCP who considered that during the study a larger number of patients could have been included, the percentage of surveys perfectly completed by patients and by HCP, and qualitative data from the final semistructured interviews with HCP; Effectiveness: the percentage of patients with the c.676G>A variant, the incidence of MODY2 in the studied region during the study compared with the incidence during the 2 years before the study in the same region, the time between proposing the study to the patient and sending the results to HCP, the number of medication changes for positive patients at the end of the study, the mean score of HCPs at the post-training test compared with the mean score before the test, the patients’ opinion concerning their participation to this study, qualitative data from the final semistructured interviews with HCP; Adoption: the percentage of HCPs from the two-family primary care practices who participated in the study, characteristics of HCPs who participated compared with those who did not participate, qualitative data from the final semistructured interviews with HCPs; Implementation: the percentage of sample kits received accurately filled, the adaptations made during the study, the final cost of this study and qualitative data from the final semistructured interviews with HCP; Maintenance: the percentage of HCPs who would like to pursue the protocol, the patients’ opinion concerning the maintenance of the access of the genetic test in primary cares, if discussions have been started with decision-makers, qualitative data from the final semi-structured interviews with HCP.

Patient and public involvement statement

Three patients with Diabetes were involved in December 2023 to test and validate the oral mucosa cell sample kit.

Analysis

For the quantitative data, the continuous variables will be analysed by Student t-tests or an analysis of variance in case of normal distribution checked by Shapiro-Wilk test, or by Mann-Whitney tests or Kruskal-Wallis test in case of non-parametric distribution. The dichotomous variables will be analysed by Pearson χ2 tests when sample size is sufficient, or Fisher exact tests for small sample size. For all tests, the degree of statistical significance will be estimated for α=0.05. A Bonferroni correction will be applied to avoid type I error due to multiple testing.

For the qualitative data, the analysis will be facilitated by MAXQDA software and will be both deductive (based on the RE-AIM categories) and inductive, in order to accommodate emerging terms.28 The analysis will be performed through the steps of decontextualisation, recontextualisation, categorisation and compilation.28 Two team members will perform an initial analysis for the development of the coding template. The interviewing, transcription and coding protocols will be documented to improve the relevance of the interviews and analyses. In addition, trained qualitative analysts and interviewers will participate in this analysis, and detailed records of field notes, group meetings, coding disagreements and resolutions will be set up.

Timeline

The training of the HCPs started on 20 February 2024, and the first sample for genotyping was received on 1 March 2024. The end of the inclusion phase is scheduled for 19 February 2025.

Ethics and dissemination

The present protocol has been approved by the research ethic committee of the ‘Centre intégré universitaire de santé et de services sociaux of Saguenay-Lac-Saint-Jean’ (CIUSSS-SLSJ) on 9 January 2024 and by the ‘Comité central d’éthique de la recherche’ (CCER) of the ‘Ministère de la Santé et des Services Sociaux’ of Quebec (Canada) on 30 January 2024 (Nagano #2024-566, project #2023-027). The informed consent of participants will be obtained orally.

A knowledge translation (KT) plan will be developed to share and communicate the results of the present study to both the scientific community and relevant decision-makers and stakeholders in the Region of Saguenay-Lac-Saint-Jean (Quebec, Canada). This KT plan will include articles for publication in peer-reviewed open-access journals, conducting media interviews and using social media and a website to disseminate findings. We will also share any findings through regional, national and international conferences.