Molló et al. BMC Family Practice (2019) 20:25 https://doi.org/10.1186/s12875-019-0916-9 STUDY PROTOCOL Open Access INTEGRA study protocol: primary care intervention in type 2 diabetes patients with poor glycaemic control Àngels Molló1,2,3†, Anna Berenguera4,5†, Esther Rubinat6,7,8,9, Bogdan Vlacho10, Manel Mata11,12,13,14,15, Josep Franch16, Bonaventura Bolíbar17,18 and Dídac Mauricio19,20,21* Abstract Background: The management of hyperglycaemia and associated cardiovascular risk factors in patients with type 2 diabetes mellitus (T2DM) may reduce diabetes-related complications. The strategy to broaden the knowledge base of primary care professionals to improve health care has mainly been prompted by the current reality of limited resources and access to specialized care. The main objective of this study is to assess the effectiveness of comprehensive interventions focused on treatment intensification, decrease clinical inertia and reduce possible barriers to treatment adherence in patients with poorly controlled diabetes in a primary care setting. Methods: This is a two-phase mixed method study, whose aims are the development of complex interventions and the assessment of their effectiveness. The main study outcome is a change in glycated haemoglobin (HbA1c) levels. The INTEGRA study is divided into two phases. Phase 1: A qualitative study with a phenomenological approach using semi-structured interviews with the objective of determining the factors related to the participants and health care professionals that influence the development and implementation of a specific intervention strategy aimed at patients with poor glycaemic control of T2DM in primary care. Phase 2: Exploratory intervention study to be conducted in Primary Health Care Centres in Catalonia (Spain), including 3 specific health care areas. The intervention study has two arms: Intervention Group 1 and 2. Each intervention group will recruit 216 participants (the same as in the control group) between the ages of 30 and 80 years with deficient glycaemic control (HbA1c > 9%). The control group will be established based on a randomized selection from the large SIDIAP (Sistema d’Informació per al desenvolupament de la Investigació en Atenció Primària) database of patients with comparable socio-demographic and clinical characteristics from the three provinces. Discussion: This study is a comprehensive, pragmatic intervention based on glycaemic treatment intensification and the control of other cardiovascular risk factors. It is also aimed at improving treatment adherence and reducing clinical inertia, which could lead to improved glycaemic control and could likewise be feasible for implementation in the actual clinical practice of primary care. Trial registration: Clinicaltrials.gov. registration number. NCT02663245; January 25, 2016. Keywords: Glycated haemoglobin, Clinical inertia, Intervention, Primary care, Type 2 diabetes, Treatment intensification * Correspondence: didacmauricio@gmail.com † Àngels Molló and Anna Berenguera contributed equally to this work. 19 Department of Endocrinology & Nutrition, CIBER of Diabetes and Associated Metabolic Diseases (CIBERDEM), Health Sciences Research Institute and University Hospital de la Santa Creu i Sant Pau, Barcelona, Spain 20 Institut Universitari d’Investigació en Atenció Primària Jordi Gol, Gran Via de Les Corts Catalanes, 591 atico, 08007 Barcelona, Spain Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Molló et al. BMC Family Practice (2019) 20:25 Page 2 of 11 Background values (in the group receiving feedback, mean HbA1c was Type 2 diabetes mellitus (T2DM) is a chronic disease 7.5% vs. 8.2% in the control group) was also observed. with a significant socioeconomic impact due to its high However, in the Rochester study [13], although a reduc- prevalence, the impact of associated complications, and tion of more than 50% was observed in clinical inertia, this high rate of mortality [1]. The elevated costs associated did not translate into an improvement in the control pa- with the disease and its impact on patient quality of life rameters after one year. Therefore, although interventions are currently the target of many health plans and gov- can be effective in reducing clinical inertia, they are not al- ernment strategies. The benefits of proper control of ways effective in improving metabolic control. T2DM and the related cardiovascular risk factors, espe- Thus, there is a need to design strategies that help cially blood pressure and lipids, are widely accepted [2]. achieve the desired treatment goals that can be simul- However, studies in Spain show that although evidence taneously implemented in real-life clinical practice in the of the improved care of T2DM patients exists, the thera- primary care setting and the best way to achieve this peutic goals are often not achieved in real-life clinical goal is to use the mixed methodology by previously ex- practice [3, 4]. ploring the patients’ own perspective, in order to design A meta-analysis performed by Tricco et al. regarding a proper implementation strategy. Previous studies have the evaluation of strategies used in clinical trials for the shown that specialised Diabetes Unit improve glycaemic improvement of quality procedures used in the manage- control [11]. Our main intervention was designed to ment of diabetes shows evidence that the most notable evaluate if a local monographic consultation run by pri- improvements are observed when baseline levels of gly- mary healthcare professionals could be effective in the cated haemoglobin (HbA1c) are high, especially over 8%, context of real-world primary healthcare practice for the in diabetic patients [5]. This suggests that interventions management of very poor controlled diabetic patients. should focus on patients with poorer control. With re- gard to strategies aimed at professionals and organiza- tions that demonstrated efficacy, the most efficient were Methods/design feedback from the information obtained in audits, train- Aims of the study ing of professionals, and organizational changes (such as Following this line of thought, the aim of this study is to electronic records, clinical reminders, and case manage- assess whether glycaemic control improves in very ment, in addition to financial incentives). poorly controlled type 2 diabetic patients as a result of Treatment adherence and clinical inertia play an im- interventions based on treatment intensification and on portant role in cases of insufficient glycaemic control. an increase in the patients’ adherence to treatment with Therapeutic or clinical inertia is defined as the failure to a diabetes-targeted clinic specifically held for those initiate or to intensify treatment when indicated [6], patients. thereby preventing or delaying the benefits of proper The hypothesis is that in patients with inadequately control. In the DIAMOND [7] study conducted in Span- controlled T2DM, a preliminary comprehensive detec- ish primary care centres, mean HbA1c was 8.1% over an tion intervention developed based on a monographic average of 2.9 years, with values of > 7% before change, consultation interview and a complex intervention to when switching from monotherapy to combination ther- improve the self-efficacy of patients could improve the apy was observed. Additionally, in an assessment of the effectiveness and cost-effectiveness in controlling gly- GEDAPS group carried out in Catalonia, clinical inertia caemia and other metabolic and risk factor parameters was detected in 33% of patients, and the mean HbA1c compared to the usual practice. level to perform the treatment change was 8.4%. [8]. The secondary aims are to determine: Sometimes, clinical inertia is related to the non-com- pliance of the patient [3, 9]. As seen in the Grant et al. – The percentage of patients who achieve HbA1c levels [10] study, interventions oriented to improve patient ad- of < 7% and < 8%. herence could consequently decrease clinical inertia. – The improvement of lipidaemic control measured by However, there are few studies on interventions aimed the mean concentration of low-density lipoprotein cho- specifically at reducing clinical inertia. lesterol(LDL-C), non-high density lipoprotein choles- Feedback to professionals is another possible key fac- terol (non-HDL-C), and triglycerides in the tor for improving clinical inertia. In an American study intervention groups relative to the control group. [11, 12], feedback was given to the professionals on the – Whether the percentage of patients who achieve objectives achieved via alerts in the electronic records target levels of LDL-C, non-HDL-C, and triglycer- and/or weekly short meetings between professionals. As ides recommended by the Catalan Health Institute a result, treatment intensification increased from 35 to guidelines (REGICOR tables) [14] is higher in the 52% in 3 years, and a significant improvement in HbA1c intervention groups than in the control group. Molló et al. BMC Family Practice (2019) 20:25 Page 3 of 11 – Whether the percentage of diabetic patients with records of patients attended at primary care centers of the mean systolic blood pressure < 140 mmHg and mean same health care districts not participating in the study. diastolic blood pressure < 90 mmHg is higher in the This study is non-randomized because it would be dif- intervention groups than in the control group. ficult give a different treatment to an individual in the – Whether the adherence to the screening for chronic same primary care center treated by the same profes- complications associated with T2DM, according to the sional without it affecting the outcome in the standard protocol of the Catalan Health Institute (fundoscopy, care arm of the study. microalbuminuria, examination of the foot: arterial index and peripheral sensitivity), is better in patients Study setting who have participated in the specialized consultation This mixed method study will be conducted in primary unit than in the control group. care settings in 3 different health care territories of the – Whether the patient’s self-efficacy to implement Catalan Health Institute (Lleida, Barcelona, and Girona). changes in risk factors is different in intervention Each basic health care area will be selected with similar group 1 (IG-1) than in intervention group 2 (IG-2). characteristics and will be organized into the interven- – Whether the direct health costs of T2DM patients tion group 1 (IG-1), the intervention group 2 (IG-2), and who participated in the monographic consultation a control group. are different from those in the control group. – The impact on satisfaction and quality of life of Characteristics of participants patients. The inclusion criteria include the following: T2DM of – Whether the improvement in the control of more than one year of disease duration, age between 30 behavioural cardiovascular risk factors (smoking and 80 years, HbA1c > 9% in the last test performed in and exercise) is better in the intervention than the 12 months prior to study inclusion, with no change in the control groups. in antidiabetic treatment within the previous 3 months. – The improvement in the adequacy and intensification The changes in the dose of a drug in the antidiabetic of treatment (clinical inertia). treatment will be not considered as exclusion criteria. The exclusion criteria include the following: T2DM Study design controlled by endocrinologist at the moment of inclu- A mixed method study will be carried out, which will sion, systemic glucocorticoid treatment (ATC code: compare poorly controlled T2DM patients (poor gly- H02AB) or orlistat treatment (ATC code: A08AB01) caemic control was defined as an HbA1c value > 9% in (chronic or during the two months prior to inclusion), the last test performed in the 12 months prior to study estimated life expectancy < 2 years, active treatment for inclusion), treated with two different comprehensive ap- malignancy (other than basal cell cancer or squamous proaches with a control group receiving only usual clin- cell skin cancer), serious mental disorders (psychosis, bi- ical care. The mixed methodology for the development polar affective disorder, severe depressive episode with of complex interventions composed of two phases: a first psychotic symptoms), class III or IV (New York Heart phase consisting of a qualitative study carried out Association functional classification) heart failure through personal interviews with patients and profes- (ICD-10: I50), dementia, renal transplantation or current sionals, and a second quantitative phase that will corres- dialysis treatment, history of drug and/or alcohol abuse pond to the execution of the clinical trial and in which (ICD-10: F10-F14), pregnancy or lactation, treatment the findings of the first phase will be applied. with immunosuppressive drugs, haemoglobinopathies We designed this study considering that the interven- (ICD-10: D58.2), or chronic anaemia, body mass index tion group 2 will be the control group for Intervention (BMI) > 45 kg/m2, or any conditions that the investigator group 1, where the specific monographic consultation is considered could prevent the patient from completing implemented. Group Intervention 1 includes the main the study. intervention for which we will test the effectiveness to improve glycaemic control. In parallel to Intervention Study phases group 2, we decided to include an additional comparison The INTEGRA study is divided into two distinct phases. group consisting of type 2 diabetic subjects attending Phase 1 is a qualitative research study, which is aimed at primary care centres managed by our institution in our identifying viable strategies to overcome barriers to region, with subjects selected according to the same treatment; these strategies have been included in the study criteria. To select this latter group, we describe the intervention study (phase 2). Phase 2 is a controlled, use of the Sistema d´Informació per al Desenvolupament non-randomized, interventional, pragmatic study that in- de la investigació en Atenció Primària (SIDIAP) data- volves primary care centres in 3 regions in Catalonia base, which contains anonymised electronic health (Lleida, Girona and Barcelona). Molló et al. BMC Family Practice (2019) 20:25 Page 4 of 11 Phase 1: Qualitative study Analysis A thematic content analysis was performed Design A qualitative study with a phenomenological ap- based on the information obtained from the interviews proach was conducted to identify the barriers and facili- [17]. The data was analysed in the following manner: tators for the management of poorly controlled T2DM after successive readings of the transcribed interviews patients. This approach was taken in order to identify [18], the researchers attained some pre-analytical insight psychosocial factors influencing glycaemic control. into the data. Next, four researchers conducted the follow- Briefly, we collected information on the patient’s percep- ing analytical steps: (a) identification of the relevant sub- tion on the communication of the diagnosis, cognitive jects and texts; (b) fragmentation of the text into units of representation of the disease (knowledge, cause, symp- meaning; (c) text codification with a mixed strategy: toms, duration, consequences and control of the dis- Leventhal model and emerging codes from the data; (d) ease), emotions associated with the disease (e.g., fears or creation of categories by grouping the codes based on the worries about the future), and their cognitive and emo- criterion of similarity; (e) analysis of each category; and (f) tional attitudes regarding strategies to control diabetes elaboration of a new text with the results. These results (diet, physical activity, and pharmacological activity). Fi- were subsequently discussed by the research team mem- nally, patients described their perceived relationship with bers until a consensus was reached (triangulation). the health-care professionals and gave their input re- The results of the qualitative study have been pub- garding the design of the INTEGRA study (e.g., visit lished recently [19]. schedule or use of information and communication technologies). The results of this phase of the study were Phase 2: Interventional study used to draw recommendations and to design strategies Design A controlled, non-randomized interventional in order to optimize the patient’s adherence and disease study that involves primary care centres of 3 health care control during the intervention phase of the INTEGRA districts in Catalonia (Lleida, Girona, and Barcelona). study. Thus, other components that may be needed for the intervention, such as the use of Information and Participants The participants are T2DM patients with Communication Technologies like internet platforms or poor glycaemic control that meet all the inclusion cri- mobile applications, was also determined. teria and sign the informed consent. Inclusion criteria for primary health care Centres Recruitment strategies and participants At each pri- The participating Primary Care centres must meet the mary care centre, a liaison was selected to facilitate the following criteria to be eligible: I) experience of more communication between the centre and the research than 5 years using the eCAP™ programme (electronic team. One member of the research team contacted the clinical records of Primary Care Centres in Catalonia) to liaison at each primary care centre. These liaisons, to- optimize the use of the software as a source of informa- gether with the primary care professionals, identified the tion and intervention (warnings, alerts, etc.); II) centres patients who meet the criteria of poor glycaemic control with more than 10 general practitioners, to obtain a and contacted the potential participants. large number of study subjects; III) centres belonging to different health care areas, to increase external validity Sampling The sampling for the study was opportunistic of the study; and IV) acceptance by the majority of pro- [15]. Although not a theoretical sample, it was taken into fessionals in the centre. account the following variables: gender, age, years of progression of T2DM, and type of treatment (oral anti- Intervention design In each basic health area, two diabetics, insulin, or a combination). intervention groups and one control group were included. Two different interventions focused on treatment intensi- fication, namely reducing inertia by professionals and re- Techniques for generating information A psychologist ducing the possible barriers to treatment adherence have carried out 8 to 10 semi-structured individual interviews been implemented, one in each intervention group. In in each Primary Health Care Centre. This interview order to reduce inertia by primary care professionals, to technique is especially useful when it is important to get closer to the objectives of health control, and to allow collect the subjective opinions of societal actors, and be- reproducibility in the current context of primary care cen- comes even more valuable when wanting to explore di- ters, the intervention was carried out through multiple in- verse points of view that represent the different attitudes tegrated strategies. To improve the intervention that might exist regarding the subject of investigation replicability, the Template for Intervention Description [16]. The interviews were audio-recorded and subse- and Replication guide was followed [20]. Each health care quently transcribed systematically and verbatim. district has a contact in a hospital with a Department of Molló et al. BMC Family Practice (2019) 20:25 Page 5 of 11 Endocrinology who provides specialized support for pro- Control group fessionals if needed. The control group will consist of a randomized selection of patients who meet the inclusion criteria and have the Intervention group 1 (IG-1) same sociodemographic and clinical characteristics as Intervention 1 consists of a comprehensive strategy with the patients from the intervention groups, including the the following components: three participating health care districts. Patients allo- cated in this group will receive routine clinical care with – A specialized consultation carried out by a general the usual control by the family doctor and nurse accord- practitioner and a nurse that will provide the ing to the current Catalan Health Institute protocol. The participant with a complete assessment of their case, data of patients in the control group will be obtained and a series of recommendations will be set for case from the electronic medical record (SIDIAP) database management, together (only if required) with the [23]. The inclusion period for the control group will be specialized team from Endocrinology and Nutrition from 9 December 2015 (the date when the first patient using a telematics communication system. The entered the study) until 31 October 2017 (end of the strategy recommendation includes management recruiting period). Data on diabetic status, glycaemia measures aimed at optimizing blood glucose, blood and all other study variables that were recorded in the pressure, lipid profile, and other cardiovascular risk electronic health records for patients in the intervention factors, as well as the detection of diabetes- groups will be extracted from the SIDIAP database for associated complications following the guidelines of 12 months for each patient. Participants will be followed the Catalan Health Institute [21] and the guidance up until they experience the outcomes of interest, die, document of the Department of Health of the local leave the SIDIAP database (e.g., change of address) or government (Generalitat de Catalunya) [22]. complete the follow-up (31 October 2018). If the inter- – Basic training in clinical practice guidelines: consulting vention groups reach the study sample earlier than 31 professionals will participate in workshops related to July 2017, the data for the control group will be followed clinical practice guidelines, thus broadening the until 12 months from the date when the last patient professionals’ knowledge base with regard to would be enrolled in the intervention group. diabetes by providing them with an increased degree of autonomy in diabetes case management. Study variables The variables are collected at different – Training of professionals in coaching: all time-points during the study (Table 1). These include: professionals in the primary care centres will sociodemographic variables (age, sex), anthropometric participate in a 7-h training programme for coaching variables (height, weight, BMI, blood pressure), clinical to be able to impart practical theoretical content on variables (drugs and dosages), smoking habits (smoking, the following subjects: strategies for active listening; non-smoker, ex-smoker), presence of multiple disorders strategies for communication without value (Clinical Risk Group and identified in the computerized judgement; support strategies to develop self- patient clinical history and the number of health prob- management skills for diabetes, hypertension, and lems as a proxy asset of multi-morbidity), presence of hyperlipidaemia; strategies to provide social and microvascular complications (diabetic retinopathy, dia- emotional support; strategies to motivate lifestyle betic nephropathy, and diabetic neuropathy), presence of changes; strategies for medication adherence; and macrovascular complications (ischaemic heart disease, strategies to access community resources. stroke, peripheral vascular disease, and heart failure), la- – The professionals will attend a 2-h training boratory variables according to the standard procedures programme to update their training for reviewing of the laboratory of reference (HbA1c, blood count, the practical cases discussed in consultation using a renal function- creatinine, glomerular filtration rate, coaching strategy point of view. urinary excretion of albumin), liver function (Alanine – Interventions based on patients’ SMS phone Aminotrasferase (ALT), Aspartate Aminotransferase messages (sending a message 1 time / week during (AST), alkaline phosphatase, Gammaglutamil transferase the first three months) to promote the change in (GGT)), lipid profile (total cholesterol, HDL-C, triglycer- behaviour and to maintain healthy lifestyle changes ides, LDL-C, and non-HDL-C), and episodes of severe to improve diabetes control. hypoglycaemia (episodes of hypoglycaemia that require the intervention of a third party for recovery will be Intervention group 2 (IG-2) recorded). Patients allocated to the IG-2 will receive the same per- Variables related to healthcare costs are also collected sonalized intervention described for the IG-1, without and include the costs derived from the medication used the monographic consultation (Fig. 1). (cost per patient per month of medication prescribed Molló et al. BMC Family Practice (2019) 20:25 Page 6 of 11 Control group n= 216 Centres enrolment Intervention group 1+ n=72 (Follow up: 0, 3, 6, 12 m) Lleida Intervention group 2++ n=72 (Follow up: 0, 3, 12 m) Phase 1- Qualitative study 8-10 patients/professionals Intervention group 1+ n=72 Semi-structured interviews Centres of three (Follow up: 0, 3, 6, 12 m) geographical areas of Barcelona of Catalunya Intervention group 2++ n=72 (Follow up: 0, 3, 12 m) Intervention group 1+ n=72 Phase 2- Clinical trial (Follow up: 0, 3, 6, 12 m) Girona Intervention group 2++ n=72 (Follow up: 0, 3, 12 m) Fig. 1 Study flow chart. *Control group: usual clinical care with the usual control by the family doctor and nurse according to the current CPG protocol. +Intervention 1: Usual clinical care with the usual control + Monographic consultation + Basic training in clinical practice guidelines + Training in coaching + 2-h training program to update the coaching strategy + Intervention based on patients SMS phone messages. ++Intervention 2: Usual clinical care with the usual control + Basic training in clinical practice guidelines + Training in coaching + 2-h training program to update the coaching strategy + Intervention based on patients SMS phone messages. m: months according to prices set by the Health Ministry), labora- and hypoglycaemia from 0 (never perceived) to 6 tory tests used (based on prices listed in the Catalan (perceived most of the time), and they are scored Health Institute service portfolio), supplementary tests, separately. professional costs (the cost of patient visits), cost derived – Quality of life: this is measured using the validated from the referrals to specialists, use of the emergency Spanish version of the Diabetes Quality of Life department, outpatient and hospital admissions, tempor- Questionnaire (ESDQoL) [26], which consists of 21 ary disability, and consumption of testing strips for items, of which 19 relate to specific life domains. measuring blood glucose. The items are scored on a 5-point scale of impact. The INTEGRA study includes variables evaluated The effects of diabetes on each of the domains are using specific questionnaires: weighted according to their importance to the quality of life of patients, and the weighted average – Physical activity: this is assessed using the Spanish impact is obtained. version of the Brief Physical Activity Assessment – Satisfaction of the health professionals before and Tool questionnaire [24]. It contains two questions after the interventions proposed in the study is that are scored on a scale from 0 to 4. The sum of assessed by a questionnaire divided into 4 sections; the scores of both questions are classified the first three assess aspects of organization, content, as follows: ≥ 4 total score = “sufficiently” active and structure, and the fourth refer to the overall (encourage the patient to continue their activity) or assessment. 0 to 3 total score = “insufficiently” active (encourage – Clinical inertia is evaluated through the questionnaire the patient to increase their activity). created by the RedGEDAPS [27]. This questionnaire – Patient treatment satisfaction: this is assessed using consists of 5 questions with two possible answers (yes the Diabetes Treatment Satisfaction Questionnaire or no) that professionals should answer in each visit (DTSQ) [25] in both the “status” and “change” when HbA1c > 9%. versions validated for Spanish patients. It comprises – Self-management of the disease is evaluated by the 8 sections, each of which is rated on a scale from 0 Spanish validated version of the Patient Activation to 6 points. The score of satisfaction with treatment Measure questionnaire [28]. This questionnaire is the sum of the 6 sections of the questionnaire, evaluates patients’ ability to play an active role in their which can range from very satisfied (36 points) to health care by assessing knowledge, skills, confidence, very dissatisfied (0 points). The remaining 2 sections and behaviours. It consists of 13 questions in which measure the perceived frequency of hyperglycaemia there are 5 possible response options (strongly disagree, Molló et al. BMC Family Practice (2019) 20:25 Page 7 of 11 Table 1 Study timeline, variables and procedures IG-1 and IG-2 Control group Data collected Assessment (months) Source Assessment Source 0 3 6 12 Month 0 & Month 12 Inclusion/Exclusion criteria 1+2 1+2 1 1+2 e-CRF § SIDIAP Sociodemographic variables and comorbidities: 1+2 1 1 1+2 eCAP § SIDIAP 1. Age, height, gender, weight, BMI, duration of T2DM, medication (active component and dose), etc.... 2. Pluripathology, clinical risk group Main variable: HbA1c 1+2 1 1 1+2 e-CRF § SIDIAP eCAP Lipid profile and other laboratory variables: 1+2 1 1 1+2 eCAP § SIDIAP Total cholesterol, HDL, LDL, triglycerides, non-HDL, urea, creatinine, e-CRF glomerular filtration rate, urine albumin, AST, ALT, FA, GGT. Systolic and diastolic blood pressure Smoking habit 1+2 1 1 1+2 eCAP § SIDIAP e-CRF Sedentary lifestyle 1+2 1+2 1 1+2 e-CRF Detection of macrovascular complications 1+2 1 1 1+2 eCAP § SIDIAP Ischaemic heart disease, stroke, peripheral artery disease, heart failure e-CRF Chronic microvascular complications detection 1+2 1 1 1+2 eCAP § SIDIAP Diabetic retinopathy, diabetic nephropathy (albuminuria, renal failure), e-CRF diabetic neuropathy. Costs: 1+2 1+2 eCAP § SIDIAP Cost of medication, laboratory tests, additional tests, cost of staff e-CRF (visits), referrals, IT Costs: 1+2 1+2 e-CRF Urgent care, outpatient and hospital admissions Satisfaction and quality of life of patients DTSQ / ESDQoL 1+2 1 1+2 e-CRF Professional Satisfaction 1+2 e-CRF Specific Questionnaire intensification/treatment adherence 1+2 1 1+2 e-CRF Occurrence of severe hypoglycaemia episodes 1+2 1 1 1+2 e-CRF SIDIAP “Patient Activation Measure” questionnaire (instrument for 1+2 1+2 1+2 e-CRF disease self-evaluation) For the IG-1 (monographic + training of professionals + additional measures originated in phase 1), clinic visits will be carried out at baseline, 3, 6, and 12 months. For the IG-2 (training of professionals + additional measures originated in phase 1), clinic visits will be carried out at the time of inclusion (month 0), 3, and 12 months disagree, agree, strongly agree, N/P). The scores range Form (e-CRF), specifically designed for the purpose of in value from 38.6 to 53.0 (on a 0–100 theoretical point the study. scale). The consultation with the specialists is carried out in addition to any others that are deemed necessary, and the number and frequency of visits is derived from the Data collection interventions proposed to the patients (pharmacological In the intervention groups (groups 1 and 2), sociodemo- and non-pharmacological). graphic, anthropometric, clinical, and laboratory vari- ables are obtained from the clinical history and Evaluation of the effectiveness between the two anamnesis of the subjects. The data obtained through intervention groups the questionnaires are collected during the clinic visit The e-CRF has been used to perform an evaluation at according to the subjects’ responses. The variables re- baseline, 3, 6, and 12 months. Secondary aims, such as lated to health care costs are obtained from the SIDIAP patient’s self-efficacy to implement changes in risk fac- database. In the control groups, all data will be obtained tors, the direct health costs of T2DM patients, and the from the SIDIAP database. impact on satisfaction and quality of life of patients, is For both groups, variables related to data are collected assessed using the e-CRF data, and the other informa- as described in Table 1 in the Electronic Case Report tion will be obtained from SIDIAP database. Molló et al. BMC Family Practice (2019) 20:25 Page 8 of 11 Evaluation of the effectiveness in the three study groups trends. Initially, basal characteristics of all groups will be Data at baseline and at 12 months of the population of evaluated to establish homogeneity in age, socio-demo- all three types of learning centres that meet the defined graphic, comorbidities, laboratory parameters, concomi- criteria for inclusion and exclusion will be compared. tant medication and diabetes complications. An initial, Criteria will be adequately defined using the ICD-10 descriptive comparison between groups of all variables codes, thoroughly specifying the temporary windows will be performed to evaluate their baseline balance. of the accepted values. Only the outcomes that in- Statistical significance will be assessed by the chi-square clude SIDIAP variables (as indicated in the Table 1) or t-test between the groups. Exploratory analyses will will be able to be evaluated; even those that are in be applied when necessary. the e-CRD will not be able to be assessed. This evalu- The main variable is set as the difference in mean ation of effectiveness will be performed in two pos- HbA1c values between the final visit and baseline. The sible ways: i) following the patients who meet the difference will be evaluated in terms of the evolution of inclusion criteria in 2015 until the end of the study HbA1c as contrasted between groups by paired in 2018 (cohort study), or ii) a cross-sectional study t-Student, or if the assumption of normality is not met in 2015 and another in 2018. for the right non-parametric. If differences in the characteristics of the patients in Sample size calculation each group are observed in the descriptive section of the The sample size was calculated using the GRANMO study, a linear regression of the main variable (parameter v.7.12 (IMIM, Barcelona) programme. Considering an evolution during the study) will be adjusted. This linear alpha risk of 0.05 and a beta risk of less than 0.2 in a bi- regression adjusted for the variables identified in the de- lateral contrast, it was necessary to include 72 patients scriptive section, or the ability to adjust the model by in each group (intervention and control) to detect a dif- setting a propensity score, will be evaluated. For all ference equal to or greater than 1% of HbA1c [5] be- models, the distribution of waste will be studied, as well tween the two groups. A common standard deviation of as the coefficient of determination (R2), to check for the 1.46 [6] and a loss rate to follow up of 20% were correct setting. assumed. As was done in Vinagre et al. [4], a subgroup analysis Considering that each physician/nurse that makes up will be performed by a) gender (male / female), b) age: < the Basic Care Unit will coordinate their diabetic pa- age 65 / ≥ 65, and c) according to HbA1c control levels tients, a correction may be made due to a possible con- (< 10% / ≥ 10%). tamination in the analysis between individuals by We will determine the prevalence of patients who assigning each Primary Care Team to intervention or achieved a concentration of HbA1c of less than 7% at control groups and taking into account the design effect. the end of the study. A categorical variable (No, Yes) will According to a study performed by our research group be calculated. The analysis of a significant decrease in [16], we estimated that each GP would have 15 potential HbA1c will be carried out using a logistic regression candidate patients with HbA1c > 9%. Assuming that the analysis adjusting by the effect of the study groups (as intra-class correlation coefficient for Primary Care is dummy variable) and the main factors previously deter- 0.05 [29], resulting in a 1.7 design effect, the required mined to be associated with the target of 7% (in a previ- sample size will be 42 * 1.7 = 72 individuals in each ous descriptive analysis or adjustment by the propensity group. score). The power of discrimination (C statistic and ROC Each control or intervention group consists of 3 basic curve) and calibration (Hosmer & Lemeshow test) for health areas, one in each geographical area, each with a the final models will be studied. The software we will mean of 10 BCUs and 15 individuals per basic care unit. use for analysis of the data will be R Core Team (2014). Thus, the sample finally calculated to obtain statistical R: A language and environment for statistical comput- significance was 72 subjects per centre, leading to a total ing. R Foundation for Statistical Computing, Vienna, of 648 study subjects. Austria (http://www.R-project.org/). First patient enrolled in the study INTEGRA on 09/ 12/2015. Data Management The researchers who collaborate in each of the primary Statistical analysis plan care centres are responsible for the accurate, complete, For data analysis, the IG-1, the IG-2, and the control and reliable collection of all data. An e-CRF has been cre- group will be compared. All analyses will be carried out ated for the study, and a data extraction from patient’s using bilateral tests with a nominal significance level of medical history will also be performed. An expert, inde- 0.05. However, if the comparisons obtain a p-value of pendent professional from the Primary Care Centres is re- less than 0.10, they will also be discussed as indicative of sponsible for periodically monitoring to corroborate the Molló et al. BMC Family Practice (2019) 20:25 Page 9 of 11 reliability of the data entered in the e-CRF with respect to HbA1c and major cardiovascular risk factors, well beyond electronic health records. The detailed description of the that achieved with the usual practices [35]. Some studies study-monitoring plan is shown in the Additional file 1. describe that patient-centred care is more valuable when targeted to patients with HbA1c > 8.5%. Patient-centred Ethics and dissemination care was most effective and cost-effective in those pa- The study is conducted according to the Helsinki Dec- tients, resulting in more improved glycaemic and meta- laration and Good Clinical Practice guidelines. The study bolic control but also improving quality-adjusted life years protocol was approved by the Ethics Committee of and costs of associated complications [36]. IDIAP Jordi Gol, Institute of Research in Primary Health In addition to the implementation of methodological Care. Trial Registration Number: P14/129. tools proposed by the INTEGRA study, identifying the Reporting of this trial will adhere to the most relevant factors that hinder the metabolic control of these pa- and up-to-date CONSORT statement [30] and its rele- tients will allow us to define new strategies for improv- vant extensions [31]. The results from this study will be ing patient care, especially for poorly controlled patients. published regardless of the outcomes. These strategies include improving adherence to treat- Confidentiality and anonymity of the data is en- ment proposed by primary care professionals, as well as sured according to Law 15/1999 of data confidential- eliminating the clinical inertia of these professionals ity, both in the implementation phase of the project treating patients who have difficulty achieving good gly- and in the resulting presentations and/or publications. caemic control targets. Individual data will be codified to ensure anonymity. There are some limitations in this study. It is well Only researchers and monitors will have access to the known that patients who voluntarily participate in re- data. search projects tend to be more motivated, which can be a common bias in all clinical trials where patients sign Discussion an informed consent. In the present study, we aim to re- According to the SIDIAP database, which contains all cruit a large group of patients in each centre. the data entered into the eCAP, the database used in The external validity of the study will be determined Catalan Health Institute, 79.6% of the patients have by the ability to recruit as many candidates as possible HbA1c values ≤8%, and the proportion of patients with because the profiles of the patients included in the study fair control (HbA1c ≤7%) is 56.1% [4]. show characteristics of low participation in clinical trials The INTEGRA study defines potential pragmatic new and less adherence to treatment recommendations. An strategies to improve metabolic control in patients with important strength of the study is that it will follow the very poorly controlled T2DM in the primary care setting CONSORT Statement for the reporting and publishing that, if demonstrated to be effective, could be imple- of clinical trials. Another strength of the study is that mented by professionals in their real-life practices at the Medical Research Council’s methodology allows for ICS. The research team agreed to define poorly con- the adaptation of the complex intervention’s compo- trolled T2DM patients as those who had HbA1c levels nents to the needs of participants based on the results of ≥9%. This is because at the time of the protocol writing, the qualitative study, and therefore an easier implemen- local and international guidelines were recommending tation in Primary care centres. the initiation of insulin therapy in type 2 diabetes when Targeting interventions at the high-risk population HbA1c levels were between 8.5 and 10% [32, 33]. Due to may allow for better use of resources, reduction of cost, this the lack of agreement, the ≥9% cut-off was chosen and reduction of side effects by avoiding unnecessary in accordance with the AACE Comprehensive Diabetes use of medications. Management Algorithm 2013 [34], which is actually a value in between all recommendations. Moreover, our Additional file population characteristics and primary health care pat- terns of treatment were taken in account. Indeed, those Additional file 1: Monitoring plan. The monitoring plan include process subjects with HbA1c levels > 9% are usually the ones for monitoring activities of the study, definition of the key information concerning the realization of the study, the verification of data sources, that most often need to be treated with insulin, but this essential documents of the study. (PDF 294 kb) therapy is frequently delayed or not even started by some general practitioners in our primary health-care Abbreviations centres [4]. Some studies report that a multifactorial tar- AQuAS: Agency Qualitat i Avaluació Sanitaries de Catalunya; SIDIAP: Sistema get intervention used for the management of T2DM is d´Informació per al desenvolupament de la Investigació en Atenció Primària; effective and feasible in clinical practice. An intensive BMI: Body mass index; DTSQ: Diabetes Treatment Satisfaction Questionnaire; e-CRF: Electronic Case Report Form; ESDQoL: Spanish version of the Diabetes intervention strategy delivered at the clinical level is as- Quality Of Life Questionnaire (ESDQoL); HbA1c: Glycated haemoglobin; sociated with a significant and durable improvement in ICH: International Council for Harmonisation of Technical Requirements for Molló et al. BMC Family Practice (2019) 20:25 Page 10 of 11 Pharmaceuticals for Human Use; IG-1: Intervention Group 1; IG-2: Intervention Publisher’s Note Group 2; LDL-C: low-density lipoprotein cholesterol; Non-HDL-C: non-high Springer Nature remains neutral with regard to jurisdictional claims in published density lipoprotein cholesterol; PIS: Patient Information Sheet; ICF: Informed maps and institutional affiliations. Consent Form; T2DM: Type 2 Diabetes Mellitus Author details 1 Centre d’Atenció Primària de Cervera, Lleida, Spain. 2Insitut Català de la Acknowledgements Salut. Institut Universitari d’Investigació en Atenció Primària Jordi Gol, Grup We would like to thank Sanofi for contributing to the translation of the DAP_CAT.), Barcelona, Spain. 3Membre del Grup d’Estudi de la Diabetis en manuscript. Additionally, we would like to thank all of the general practitioners Atenció Primària (GEDAPS) de la Societat Catalana de Medicina Familiar i and nurses. Comunitària (CAMFIC) i de la RedGDPS, Barcelona, Spain. 4Institut Universitari d’Investigació en Atenció Primària Jordi Gol, Barcelona, Spain. 5Universitat Funding Autònoma de Barcelona, Bellaterra, Cerdanyola del Vallès, Spain. 6Institut The study is partially supported by unrestricted grant from Sanofi. The Universitari d’Investigació en Atenció Primària Jordi Gol, Barcelona, Spain. 7 study has also intramural support from Institut Universitari d’Investigació Serra Húnter fellow - Facultad de Enfermería y Fisioterapia de la Universidad en Atenció Primària Jordi Gol. The funder does not have any role in de Lleida, Lleida, Spain. 8Grup de Recerca en Cures de la Salut (GRECS) – writing the study protocol. This funding source will not have any role IRBLleida, Lleida, Spain. 9Centro de Investigacion Biomedica en Red de during its execution, analyses, interpretation of the data, or decision to submit Diabetes y Enfermedades Metabolicas Asociadas (CIBERDEM), Instituto de results. Salud Carlos III (ISCIII), Gran Via Corts Catalanes, 587, àtic, 08007 Barcelona, Spain. 10Institut Universitari d’Investigació en Atenció Primària Jordi Gol, Gran Via Corts Catalanes, 587, àtic, 08007 Barcelona, Spain. 11Centre d’Atenció Availability of data and materials Primària La Mina, Barcelona, Spain. 12CIBER of Diabetes and Associated The datasets generated and/or analysed during the current study are not Metabolic Diseases (CIBERDEM), Madrid, Spain. 13Institut Català de la Salut, publicly available but are available from the corresponding author on Cerdanyola del Vallès, Spain. 14Institut Universitari d’Investigació en Atenció reasonable request. Primària Jordi Gol, Barcelona, Spain. 15Membre del Grup d’Estudi de la diabetis en Atenció Primària (GEDAPS) de la Societat Catalana de Medicina Authors’ contributions Familiar i Comunitària (CAMFIC) i de la RedGDPS, C. Mar, s/n 08930 Sant AM, AB, MM, JF, BB, and DM contributed to the conception and design and Adrià de Besòs, Barcelona, Spain. 16Centre d’Atenció Primària Raval Sud, methodology of the study. AM and DM are the scientific coordinators of the CIBER of Diabetes and Associated Metabolic Diseases (CIBERDEM), Institut project. AM, AB, MM, JF, BB and DM designed the intervention. BV and ER Català de la Salut, Institut Universitari d’Investigacióen Atenció Primària Jordi contributed to the study supervision, and execution. AM, AB, MM, JF, BB, BV, Gol, Av. Drassanes, 17-21 08001 Barcelona, Spain. 17Institut Universitari ER and DM developed the draft manuscript. All authors read and approved d’Investigacióen Atenció Primària Jordi Gol, Barcelona, Spain. 18Universitat the final manuscript. Autònoma de Barcelona, Campus de la UAB, Plaça Cívica, 08193 Bellaterra, Spain. 19Department of Endocrinology & Nutrition, CIBER of Diabetes and Associated Metabolic Diseases (CIBERDEM), Health Sciences Research Ethics approval and consent to participate Institute and University Hospital de la Santa Creu i Sant Pau, Barcelona, The study will be conducted according to the Helsinki Declaration and Good Spain. 20Institut Universitari d’Investigació en Atenció Primària Jordi Gol, Gran Clinical Practice guidelines. The study protocol was approved by the Ethics Via de Les Corts Catalanes, 591 atico, 08007 Barcelona, Spain. 21Department Committee of IDIAP Jordi Gol, Institute of Research in Primary Health Care. of Endocrinology & Nutrition, Hospital de la Santa Creu i Sant Pau, Sant Trial Registration Number: P14/129. Quinti, 89, 08041 Barcelona, Spain. The study was registered in the clinicaltrials.gov registry under clinical trial registration number: NCT02663245; January 25, 2016. Received: 17 January 2017 Accepted: 29 January 2019 Reporting of this trial will adhere to the most relevant and up-to-date CONSORT statement [29] and its relevant extensions [30]. The results from this study will be published regardless of the outcomes. Confidentiality and anonymity of the data will be ensured according to Law References 15/1999 of data confidentiality, both in the implementation phase of the 1. van Dieren S, Beulens JW, van der Schouw YT, Grobbee DE, Neal B. The project and in the resulting presentations and/or publications. Individual global burden of diabetes and its complications: an emerging pandemic. data will be codified to ensure anonymity. Only researchers and monitors Eur J Cardiovasc Prev Rehabil. 2010;17(Suppl 1):S3–8. will have access to the data. 2. Lorber D. Importance of cardiovascular disease risk management in Summary of the informed consent procedure patients with type 2 diabetes mellitus. Diabetes Metab Syndr Obes. We used the final study protocol to prepare the patient information 2014;7:169–83. sheet (PIS) and the written informed consent form (ICF) in accordance 3. Mata-Cases M, Roura-Olmeda P, Berengué-Iglesias M, Birulés-Pons M, with the current Spanish applicable legislation. The PIS included all the Mundet-Tuduri X, Franch-Nadal J, et al. Fifteen years of continuous procedures, visits, risks and inconveniences for the participant in the improvement of quality care of type 2 diabetes mellitus in primary care in study. A properly qualified site investigator provided the information to Catalonia, Spain. Int J Clin Pract. 2012;66:289–98. the potential study participants during a personal interview prior to the 4. Vinagre I, Mata-Cases M, Hermosilla E, Morros R, Fina F, Rosell M, et al. subject’s inclusion and before performing any procedures of the study. Control of glycemia and cardiovascular risk factors in patients with type During the interview, all subjects had an adequate period of time to 2 diabetes in primary care in Catalonia (Spain). Diabetes Care. 2012;35: reflect before making the decision to participate in the study, and had 774–9. the opportunity to ask any questions or concerns about the study 5. Tricco AC, Ivers NM, Grimshaw JM, Moher D, Turner L, Galipeau J, et al. procedures. Once they agreed to participate, each study subject signed Effectiveness of quality improvement strategies on the management of and dated the ICF, and included information regarding the complete diabetes: a systematic review and meta-analysis. Lancet. 2012;379(9833): name of the investigator and the date of the interview. All the 2252–61. procedures of informed consent acquirement were adequately reflected 6. Phillips LS, Branch WT, Cook CB, Doyle JP, El-Kebbi IM, Gallina DL, et al. in the patient’s clinical history. Clinical inertia. Ann Intern Med. 2001;135(9):825–34. 7. Conthe P, Mata M, Orozco D, Pajuelo F, Barreto CS, Fernández-Anaya S, et al. Degree of control and delayed intensification of antihyperglycaemic Consent for publication treatment in type 2 diabetes mellitus patients in primary care in Spain. Not applicable. Diabetes Res Clin Pract. 2011;91(1):108–14. 8. Mata-Cases M, Roura-Olmeda P, Pepió-Vilaubí JM, Benito-Badorrey B, Franch Competing interests Nadal J, Coll de Tuero G. Actitud del médico de atención primaria ante una The authors declare that do not have any conflict of interest. HbA1c >7%: Valoración de la inercia terapéutica en los pacientes con Molló et al. BMC Family Practice (2019) 20:25 Page 11 of 11 control deficiente de la evaluación Gedaps 2007 en Catalunya. Av Diabetol. 29. Campbell MK, Mollison J, Grimshaw JM. Cluster trials in implementation 2010;26 Suppl 2:9–33. research: estimation of intracluster correlation coefficients and sample size. 9. McEwen LN, Bilik D, Johnson SL, Halter JB, Karter AJ, Mangione CM, et al. Stat Med. 2001;20(3):391–9. Predictors and impact of intensification of antihyperglycemic therapy in 30. Schulz KF, Altman DG, Moher D, CONSORT Group. CONSORT 2010 type 2 diabetes: translating research into action for diabetes (TRIAD). statement: updated guidelines for reporting parallel group randomised Diabetes Care. 2009;32(6):971–6. trials. Ann Intern Med. 2010;11(32):726–32. 10. Grant R, Adams AS, Trinacty CM, Zhang F, Kleinman K, Soumerai SB, et al. 31. Montgomery P, Grant S, Hopewell S, Macdonald G, Moher D, Michie S, et al. Relationship between patient medication adherence and subsequent Protocol for CONSORT-SPI: an extension for social and psychological clinical inertia in type 2 diabetes glycemic management. Diabetes Care. interventions. Implement Sci. 2013;8:99. 2007;30(4):807–12. 32. Menéndez Torre E, Lafita Tejedor FJ, Artola Menéndez S, Millán Núñez- 11. Phillips LS, Ziemer DC, Doyle JP, Barnes CS, Kolm P, Branch WT, et al. An Cortés J, Alonso García Á, Puig Domingo M, et al. Recomendaciones para el endocrinologist-supported intervention aimed at providers improves tratamiento farmacológico de la hiperglucemia en la diabetes tipo 2. diabetes management in a primary care site: improving primary care of Endocrinol y Nutr [Internet]. 2011;58(3):112–20 Available from: http:// African Americans with diabetes (IPCAAD) 7. Diabetes Care. 2005;28(10): linkinghub.elsevier.com/retrieve/pii/S1575092211000246. 2352–60. 33. American Diabetes Association. Standards of Medical Care in Diabetes-- 12. Ziemer DC, Doyle JP, Barnes CS, Branch WT Jr, Cook CB, El-Kebbi IM, 2014. Diabetes Care. 2014; et al. An intervention to overcome clinical inertia and improve diabetes 34. Garber A, Abrahamson M, Barzilay J, Blonde L, Bloomgarden Z, Bush M, mellitus control in a primary care setting: improving primary Care of et al. AACE Comprehensive Diabetes Management Algorithm 2013. Endocr African Americans with Diabetes (IPCAAD) 8. Arch Intern Med. 2006; Pract:2013. 166(5):507–13. 35. Vaccaro O, Franzini L, Miccoli R, Cavalot F, Ardigò D, Boemi M, MIND.IT 13. Fiscella K, Volpe E, Winters P, Brown M, Idris A, Harren T. A novel approach Study Group, et al. Feasibility and effectiveness in clinical practice of a to quality improvement in a safety-net practice: concurrent peer review multifactorial intervention for the reduction of cardiovascular risk in patients visits. J Natl Med Assoc. 2010;102(12):1231–6. with type 2 diabetes: the 2-year interim analysis of the MIND.IT study: a 14. Marrugat J, Vila J, Baena-Díez JM, Grau M, Sala J, Ramos R, et al. Relative cluster randomized trial. Diabetes Care. 2013;36(9):2566–72. validity of the 10-year cardiovascular risk estimate in a population cohort of 36. Slingerland AS, Herman WH, Redekop WK, Dijkstra RF, Jukema JW, Niessen the REGICOR study. Rev Esp Cardiol. 2011;64(5):385–94. LW. Stratified patient-centered care in type 2 diabetes: a cluster- 15. Marshall MN. Sampling for qualitative research. Family Pract. 1996;13(6):522–5. randomized, controlled clinical trial of effectiveness and cost-effectiveness. 16. Berenguera A, Fernández de Sanmamed MJ, Pons M, Pujol E, Rodríguez D, Diabetes Care. 2013;36:3054–61. Saura S. Escuchar, observar y comprender. Recuperando la narrativa en las Ciencias de la Salud. Aportaciones de la investigación cualitativa. Barcelona: Institut Universitari d'Investigació en Atenció Primària Jordi Gol (IDIAP J. Gol); 2014. 17. Pope C, Ziebland S, Mays N. Qualitative research in health care. Analysing qualitative data. BMJ. 2000;320(7227):114–6. 18. MacLean LM, Meyer M, Estable A. Improving accuracy of transcripts in qualitative research. Qual Health Res. 2004;14:113–23 19. Berenguera A, Mollo-Inesta A, Mata-Cases M, Franch-Nadal J, Bolibar B, Rubinat E, et al. Understanding the physical, social, and emotional experiences of people with uncontrolled type 2 diabetes: a qualitative study. Patient Prefer Adherence. 2016;10:2323–32. 20. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide. BMJ. 2014;348:g1687. 21. Brugada M, Carreras T, Cano JF, Estruch M, Garrido JM, Mendoza G, et al. Abordatge de la diabetis mellitus tipus 2. [En línia]. 2a ed. Barcelona: Institut Català de la Salut; 2012. Guies de pràctica clínica i material docent, núm. 15 [URL available at: http://ics.gencat.cat/web/.content/documents/assistencia/ gpc/GuiaDiabetis2015.pdf]. 22. Pautes per al tractament farmacològic de la diabetis mellitus tipus 2. Barcelona: Agència de Qualitat i Avaluació Sanitàries de Catalunya. Departament de Salut. Generalitat de Catalunya; 2013. (Programa d’Harmonització Farmacoterapèutica de Medicaments en l’Àmbit de l’Atenció Primària i Comunitària del Servei Català de la Salut; 1/2013). 23. Bolíbar B, Avilés F, Morros R, Garcia-Gil M, Hermosilla E, Ramos R, GrupoSIDIAP, et al. SIDIAP database: electronic clinical records in primary care as a source of information for epidemiologic research. Med Clin (Barc). 2012;138(14):617–21. 24. Puig A, Pena O, Romagera M, Duran E, Heras A, Solà M, et al. How to identify physical inactivity in primary care: validation of the Catalan and Spanish versions of 2 short questionnaries. Aten Prim. 2012;44(8):485–93. 25. Gomis R, Herrera-Pombo J, Calderón A, Rubio-Terrés C, Sarasa P. Validación del cuestionario “Diabetes treatment satisfaction questionnaire” (DTSQ) en la población española. Pharmacoeconomics. 2006;3(1):7–18. 26. Millan M. Quality-of-life questionnaire designed for diabetes mellitus (EsDQOL). Aten Prim. 2002;29(8):517–21. 27. López-Simarro F, Brotons C, Moral I, Cols-Sagarra C, Selva A, Aguado-Jodar A, et al. Inertia and treatment compliance in patients with type 2 diabetes in primary care. Med Clin (Barc). 2012;138(9):377–84. 28. Hibbard JH, Mahoney ER, Stockard J, Tusler M. Development and testing of a short form of the patient activation measure. Health Serv Res. 2005;40(6 Pt 1):1918–30.