Open access Original research Prevalence and coprevalence of chronic comorbid conditions in patients with type 2 diabetes in Catalonia: a population-based cross-sectional study Manel Mata-Cases,1,2 Josep Franch-Nadal,1,2 Jordi Real,1,2 Marta Cedenilla,3 Didac Mauricio2,4 To cite: Mata-Cases M, Abstract Franch-Nadal J, Real J, et al. Strengths and limitations of this study Objectives To evaluate the prevalence and coprevalence Prevalence and coprevalence of of several chronic conditions in patients with type 2 chronic comorbid conditions in ►► Population-based design. diabetes in a Mediterranean region. patients with type 2 diabetes in ►► Use of a primary care registry with a large number of Design A cross-sectional study. Catalonia: a population-based subjects managed under real world conditions. cross-sectional study. BMJ Open Setting Two hundred and eighty-six primary care teams ►► Unlike other population-based studies, the glycated- 2019;9:e031281. doi:10.1136/ of the Catalonian Health Institute (Catalonia, Spain). haemoglobin levels and the glomerular filtration rate bmjopen-2019-031281 Participants We included patients aged ≥18 years were available for approximately 90% of cases. with a diagnosis of type 2 diabetes by 31 December, ►► Prepublication history and ►► Diagnoses were based on data recorded in the 2016, who were registered in the Information System for database. Therefore, a misclassification cannot be additional material for this paper are available online. To the Development of Research in primary care (SIDIAP) excluded. view these files, please visit database. We excluded patients with a diagnosis of type 1 ►► For some outcomes (eg, albuminuria), the propor- the journal online (http://​dx.​doi.​ diabetes, gestational diabetes mellitus and any other type tion of missing data was high. Some laboratory data org/​10.​1136/​bmjopen-​2019-​ of diabetes. might have been available because the patients 031281). Primary and secondary outcome measures We were at higher risk of developing particular condi- collected data on diabetes-related comorbidities (ie, tions; therefore, our study may have overestimated Received 25 April 2019 chronic complications, associated cardiovascular risk Revised 29 August 2019 the prevalence of diagnoses based on those labora- factors and treatment complications). Diagnoses were tory values. Accepted 17 September 2019 based on the International Classification of Diseases, 10th Revision codes recorded in the database or, for some entities, on the cut-off points for a particular test result or a specific treatment indicated for that entity. The presence Introduction © Author(s) (or their and stage of chronic kidney disease (CKD) were based Multimorbidity is common,1 2 and diabetes employer(s)) 2019. Re-use permitted under CC BY-NC. No on the glomerular filtration rate, the CKD Epidemiology is the most common chronic comorbid commercial re-use. See rights Collaboration creatinine equation and the urine albumin- condition among patients with multimor- and permissions. Published by to-creatinine ratio. bidity after hypertension.3 The proportion BMJ. Results A total of 373 185 patients were analysed. 82% 1 of patients exhibited ≥2 comorbidities and 31% exhibited of patients with diabetes who exhibit two or DAP-Cat Group. Unitat de Suport a la Recerca Barcelona ≥4 comorbidities. The most frequent comorbidities more chronic conditions exceeds 90% in Ciutat, Institut Universitari were hypertension (72%), hyperlipidaemia (60%), studies conducted in the USA4–6 and Spain,7 d'Investigació en Atenció obesity (45%), CKD (33%), chronic renal failure (CRF) reaches 75% in a study conducted in the UK8 Primària Jordi Gol (IDIAP Jordi (28%) and cardiovascular disease (23%). The most and is over 50% in a study conducted in the Gol), Barcelona, Spain 2 frequently coprevalent pairs of chronic conditions were Gulf countries.9 When evaluated individu- CIBER of Diabetes and Associated Metabolic Diseases the combination of hypertension with hyperlipidaemia ally, the most common comorbidities are (CIBERDEM), Instituto de Salud (45%), obesity (35%), CKD (28%), CRF (25%) or hypertension, overweight/obesity, hyperlip- Carlos III (ISCIII), Barcelona, cardiovascular disease (19%), as well as the combination of idaemia, chronic kidney disease and cardio- Spain hyperlipidaemia with obesity (28%), CKD (21%), CRF (18%) vascular disease.4 5 When evaluated in terms 3 Medical Affairs, Merk Sharp & or cardiovascular disease (15%); other common pairs of of the coprevalence of two or more chronic Dohme Spain, Madrid, Spain comorbidities were obesity/CKD, obesity/CRF, hypertension/ 4 Department of Endocrinology conditions, the most common doublets or retinopathy, hypertension/albuminuria, hypertension/urinary & Nutrition, Hospital de la Santa tract infection, CVD/CRF and CVD/CKD, which were each triplets of comorbidities are concordant Creu i Sant Pau, Barcelona, present in more than 10% of patients. diseases (ie, diseases that are more directly Spain related to the pathophysiological risk profile Conclusion Patients with type 2 diabetes have a high Correspondence to frequency of coprevalence of metabolic risk factors, and/or management of diabetes).4 10–13 Dr Didac Mauricio; cardiovascular disease and CKD and thus require an The presence of certain comorbidities ​didacmauricio@​gmail.​com integrated management approach. in patients with diabetes may interfere with Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 1 Open access the achievement of therapeutic goals. Thus, the co-oc- information extracted from the e-CAP. Pharmacological currence of obesity is associated with poorer glycaemic treatments are automatically added to the SIDIAP from and blood pressure control11 14 and a higher likelihood the pharmacy-invoicing data provided by the Catalonian of emergency department visits and readmissions11; the National Health Service (CatSalut). co-occurrence of congestive heart failure correlates with The study population consisted of patients aged at the lack of achievement of cholesterol management.14 least 18 years with a diagnosis of type 2 diabetes mellitus Comorbidities in patients with diabetes also negatively (T2DM; International Classification of Diseases, 10th Revi- impact quality of life and the ability to self-manage the sion (ICD-10) codes E11, E11.0 to E11.9, E14 and E14.0 disease, and they increase healthcare utilisation.12 15–17 to E14.9) by 31 December, 2016. We excluded patients Certain comorbidities, such as macrovascular disease, with a diagnosis of type 1 diabetes, gestational diabetes chronic respiratory disease and chronic kidney disease, mellitus and any other type of diabetes. are associated with an increased risk of mortality.18 Never- In accordance with the Spanish regulations on obser- theless, the appropriate management of multimorbidity vational studies, this retrospective study conducted using in patients with diabetes can lead to more effective disease anonymised data did not require informed consent to be management and, in turn, improve patient outcomes.10 obtained from the patients. We did not involve patients or Although some studies have evaluated the prevalence the public in our study. and coprevalence of chronic comorbid conditions in patients with diabetes, most were conducted in the USA, Patient and public involvement and few studies have been conducted in Europe.3 7 8 19 20 Of Patients were not involved in the design or conduct of the these studies, two used non-representative convenience study. We plan to disseminate the results through peer-re- samples,19 20 and one was conducted in a Mediterranean viewed publications. region (Basque Country, Spain)7; however, this latter study does not include information on important comor- Comorbidities and other patient and illness-related bidities such as obesity, hyperlipidaemia and chronic characteristics kidney disease. Additionally, information on drug therapy We collected data on comorbidities on 31 December, among these patients is scarce.21 2016, which was considered the index date. Based on a In the present study, we aimed to evaluate the preva- previous study,4 the following diabetes-related comorbid- lence and coprevalence of several chronic conditions in ities (ie, chronic complications, associated cardiovascular patients with type 2 diabetes in a Mediterranean region risk factors and treatment adverse effects or contraindi- (Catalonia, Spain) using the Information System for the cations) were recorded: congestive heart failure, chronic Development of Research in primary care (SIDIAP) data- kidney disease, retinopathy, neuropathy, cardiovascular base. We also aimed to specifically study the prevalence disease (myocardial infarction, ischaemic heart disease, and coprevalence of cardiovascular diseases and chronic cerebrovascular disease and peripheral arterial disease), kidney disease (distinguishing between chronic kidney urinary tract infection, genital mycotic infection, hypogly- disease and chronic renal failure) in this population. caemia, pancreatitis, liver disease, hypertension, hyper- Finally, we evaluated the use of oral antihyperglycaemic lipidaemia, overweight/obesity, microalbuminuria and drugs in these patients. macroalbuminuria/proteinuria. Diagnoses were based on ICD-10 codes recorded in the database or, for some entities, on cut-off values for particular test results or a Patients and methods specific treatment indicated for that entity. Specific diag- Study design, data sources and selection criteria nosis codes and/or definitions appear in online supple- This cross-sectional study used the SIDIAP database, mentary table 1. which has been described in detail elsewhere.22 23 Health- In addition to comorbidities, the following informa- care is organised in Spain in the framework of a National tion was recorded: age; gender; smoking status; time Health System that is mainly financed by general taxes since the onset of T2DM; date of the T2DM diagnosis; and based on the principles of universality, free access the most recent values (closest to the index date) for and equity. People are assigned to a family physician and systolic and diastolic blood pressure, total cholesterol, a nurse practitioner (basic unit of care) who are working low density lipoprotein (LDL) cholesterol, high density together in geographically organised primary health- lipoprotein (HDL) cholesterol, triglycerides, non-HDL care teams. The Catalonian Health Institute is the main cholesterol, weight, height, urine albumin-to-creatinine provider of primary healthcare services in Catalonia, ratio (UACR), and glycated haemoglobin (HbA1c); the managing 286 primary care teams, including 74% of the highest serum creatinine lab value (worst clinical situ- total population. Primary care professionals of the Cata- ation) recorded in 2016; the use of antidiabetic agents lonian Health Institute (approximately 15 000) use the coded with the Anatomical Therapeutic and Chemical same computerised medical record programme (e-CAP) Classification system; date of initiation of antidiabetic that was created and managed by the institution itself. drugs; use of antihypertensive drugs and lipid-lowering The SIDIAP database was created in 2010 for research agents; use of cardiovascular drugs and use of other purposes and contains anonymous, longitudinal, patient treatments. The presence and stage of chronic kidney 2 Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 Open access disease were based on the estimated glomerular filtra- males (33% and 29% of women and men, respectively, tion rate calculated using the Chronic Kidney Disease exhibited four or more comorbidities). Patients aged ≥75 Epidemiology Collaboration (CKD-EPI)24 equation and years also showed a greater number of comorbidities the UACR. Microalbuminuria was defined as a UACR than younger patients (41% of patients aged ≥75 years of 30 to 299 mg/g, and macroalbuminuria was defined had four or more comorbidities compared with 24% of as a UACR ≥300 mg/g. Chronic kidney disease (CKD) patients aged <75 years). was defined as an estimated glomerular filtration rate The most frequent comorbidities were hypertension <60 mL/min/1.73 m2 calculated by using the CKD-EPI (72%), hyperlipidaemia (60%), obesity (45%), chronic equation and/or urine UACR ≥30 mg/g, while chronic kidney disease (33%), chronic renal failure (28%) and renal failure (CRF) was defined as an estimated glomer- cardiovascular disease (23%) (figure 1). The pattern of ular filtration rate <60 mL/min/1.73 m2 calculated using comorbidities was similar between the two age groups, but the CKD-EPI equation. The inclusion of the CKD defi- patients aged ≥75 years showed an increased frequency of nition was based on the recommendation of the Kidney hypertension, chronic kidney disease and chronic renal Disease: Improving Global Outcomes guidelines as a failure and, to a lesser extent, urinary tract infection and more comprehensive method to establish CKD staging. congestive heart failure; in contrast, the frequency of obesity was lower in this subgroup of patients (figure 2). Statistical analysis Males and females showed a similar pattern of comorbid- Continuous variables are presented as the means and SD, ities, with females showing a greater frequency of urinary and categorical outcomes variables are presented as abso- tract infection, obesity, chronic renal failure, hyper- lute and relative frequencies. tension and neuropathy, while males showed a greater We estimated the prevalence of individual conditions frequency of cardiovascular disease (figure 3). and the coprevalence of doublets of comorbid condi- tions, and these values are presented as the propor- Coprevalence of chronic comorbidities tion of patients exhibiting the condition or doublet of The most frequently coprevalent pairs of chronic condi- conditions. tions were the combinations of hypertension with hyper- As preplanned, the presented prevalence and copreva- lipidaemia (45%), obesity (35%), chronic kidney disease lence results were stratified by sex and age. We estimated (28%), chronic renal failure (25%) or cardiovascular the proportion of patients who had an HbA1c level <7.0% disease (19%), as well as the combinations of hyperlipi- in groups stratified by the presence of comorbidities in daemia with obesity (28%), chronic kidney disease (21%), target organs (ie, cardiovascular disease, heart failure, chronic kidney disease, neuropathy and retinopathy) chronic renal failure (18%) or cardiovascular disease and by the number of concurrent comorbidities in target (15%). Other common doublets of comorbidities were organs to explore the potential relationship between the obesity and chronic kidney disease, obesity and chronic presence of comorbidities and the degree of diabetes renal failure, hypertension and retinopathy, hypertension control. and albuminuria, hypertension and urinary tract infec- All statistical analyses were performed using Stata 15 tion, cardiovascular disease and chronic renal failure (Stata Corp, College Station, Texas, USA). and cardiovascular disease and chronic kidney disease, which were each present in greater than 10% of patients (figure 4). This pattern of comorbidities was similar in patients aged ≥75 years but, overall, with a higher preva- Results lence than the values described above (online supplemen- Patients’ disposition and characteristics tary table 2). The pattern of the prevalence of doublets By 31 December, 2016, the SIDIAP database contained of chronic conditions was similar between males and records from 7 251 277 patients; 373 185 (5.1%) met females, except for a higher frequency of the combina- the inclusion criteria and were included in the analyses tion of obesity with chronic kidney disease and hyperten- (online supplementary figure 1). The mean (SD) age of sion with urinary tract infections in females, which were the patients was 70.3 (12.1) years (37% were aged ≥75 each present in 16% of patients (online supplementary years), and the population contained a slightly higher percentage of men (55%) (table 1). Forty-five per cent of tables 3,4). the patients were obese, and 56% showed an HbA1c level lower than 7% (table 1). The most common treatments Cardiovascular disease and chronic renal failure received were non-insulin antidiabetic drugs, either as The proportions of patients presenting cardiovascular monotherapy (37%) or in combination (23%). disease and chronic renal failure alone or in combination are shown in figure 5. Both entities were present in 10% Prevalence of chronic comorbidities of the patients in the overall sample and 16% of patients The mean (SD) number of comorbidities was 2.8 (1.4), aged ≥75 years (figure 5). The distribution of chronic with 82% of patients exhibiting two or more comorbid- renal failure stages in patients with diabetes and cardio- ities and 31% exhibiting four or more comorbidities. vascular disease in the total population and in groups Females showed a greater number of comorbidities than stratified by age is presented in figure 6. Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 3 Open access Table 1 Patient characteristics Overall n=373 185 Gender Age # Evaluable Male Female <75 years ≥75 years Characteristic patients Result n=204 707 n=168 478 n=234 811 n=138 374 Sex (females), % 373 185 45.1 – – 39.0 55.6 Age (years), mean (SD) 373 185 70.3 (12.1) 68.2 (12.0) 72.5 (12.4) 62.7 (9.0) 82.6 (5.1) Current smokers, % 369 140 14.3 20.1 7.3 19.9 4.8 Body mass index (kg/m2) 321 739            Mean (SD) 30.1 (5.2) 29.4 (4.6) 30.8 (5.8) 30.7 (5.4) 28.9 (4.7)  <18.5 kg/m2, % 0.3 0.2 0.4 0.2 0.4 2  18.5–<25 kg/m , % 14.4 14.7 14.1 11.5 19.6  25–<30 kg/m2, % 40.4 45.0 34.7 38.3 43.9  30–35 kg/m2, % 29.3 29.0 29.7 31.3 26.0 2  >35 kg/m , % 15.6 11.1 21.1 18.7 10.1  Obesity (≥30 kg/m2), % 44.9 40.1 51.0 50.0 36.1 Duration of diabetes 373 185 8.7(4.5;12.8) 8.3(4.4;12.5) 9.2(4.7;13.1) 7.4(3.8;11.8) 11.2(6.3;14.3) (years), mean (IQR) HbA1c 330 014  %, Mean (SD) mmol/mol, 7.1 (1.3) 7.1 (1.4) 7.1 (1.3) 7.2 (1.4) 7,0 (1.2) mean (SD) 54.0 (12.1) 54.0 (12.4) 54.0 (11.7) 55.0 (13.0) 53.0 (10.4)  <7% (53 mmol/mol), % 55.5 55.0 56.1 53.3 59.1  <6% (42 mmol/mol), % 13.8 14.2 13.5 13.1 15.2  6–<6.5% (42–<48 mmol/ 20.9 20.5 21.4 19.9 22.5 mol), %  6.5–<7% (48–<53 mmol/ 20.7 20.3 21.2 20.3 21.4 mol), %  7–<8% (53–<64 mmol/ 24.9 24.9 25.0 25.1 24.7 mol), %  8–<9% (64–<75 mmol/ 10.6 10.6 10.5 11.0 9.8 mol), %  >9% (>75 mmol/mol), % 9.0 9.5 8.5 10.7 6.4 Blood pressure (mm Hg) 353 331            Systolic blood pressure, 133 (13.6) 133 (13.2) 133 (14.0)  133 (13.0) 134 (14.4) mean (SD)  Diastolic blood pressure, 75.0 (9.7) 75.4 (9.7) 74.4 (9.6)  77.1 (9.1) 71.4 (9.6) mean (SD)  <140/90 mm Hg, % 71.6 71.5 71.7 72.9 69.3 Lipid profile            Total cholesterol (mg/dL), 335 522 182 (39.9) 175 (39.2) 191 (39.1) 185 (40.3) 177 (38.6) mean (SD)  LDL-cholesterol (mg/dL), 315 578 103 (32.5) 99.1 (31.7) 108 (32.8) 105 (32.8) 99.3 (31.6) mean (SD)  LDL-cholesterol 315 578 80.6 83.9 76.8 78.2 84.5 <130 mg/dL, % (all)  LDL-cholesterol 74 353 68.0 72.4 59.5 68.5 67.5 <100 mg/dL, % (only SP)  LDL-cholesterol <70 mg/ 74 353 25.3 28.4 19.2 25.7 24.8 dL, (%) (only SP)  HDL-cholesterol (mg/dL), 326 354 48.8 (12.9) 45.6 (11.9) 52.5 (13.1) 47.9 (12.7) 50.2 (13.2) mean (SD)  Triglycerides (mg/dL), 328 291 159 (107) 161 (119) 158 (89.4) 169 (121) 142 (74.4) mean (SD) Chronic renal failure (eGFR 336 198 28.0 24.5 32.2 14.5 50.3 <60 mL/min), % Continued 4 Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 Open access Table 1 Continued Overall n=373 185 Gender Age # Evaluable Male Female <75 years ≥75 years Characteristic patients Result n=204 707 n=168 478 n=234 811 n=138 374 Severe chronic renal failure 336 198 3.9 3.4 4.6 1.7 7.6 (eGFR <30 mL/min), % End stage renal failure 336 198 0.7 0.7 0.6 0.5 1.0 (eGFR <15 mL/min), % Albuminuria, % 230 819 14.2 15.9 12.0 12.4 16.1 Therapeutic step, % 373 185            No drugs 18.4 17.9 19.0 17.8 19.4  NIAD monotherapy 37.1 36.7 37.6 36.7 37.9  NIADs in combination 23.2 25.1 20.9 25.3 19.7  Insulin 6.0 5.6 6.5 4.5 8.5  Insulin±NIAD 15.3 14.7 16.0 15.7 14.5 eGFR, estimated glomerular filtration rate; HbA1c, glycated haemoglobin; HDL, high-density lipoprotein; IQR, interquartile range; LDL, low-density lipoprotein; NIAD, non-insulin antidiabetic drug. Figure 1 Prevalence of comorbidities in patients with type Figure 2 Prevalence of comorbidities in patients with type 2 diabetes in the total population. The proportions of patients 2 diabetes stratified by age. The proportions of patients with a particular comorbidity, regardless of the presence of with a particular comorbidity, regardless of the presence of other comorbidities are shown. (a) Obesity is defined as a other comorbidities are shown. (a) Obesity is defined as a body mass index ≥30 kg/m2. (b) Chronic kidney disease is body mass index ≥30 kg/m2. (b) Chronic kidney disease is defined as an estimated glomerular filtration rate <60 mL/ defined as an estimated glomerular filtration rate <60 mL/ min/1.73 m2 calculated using the CKD-EPI equation and/or min/1.73 m2 calculated using the CKD-EPI equation and/or urine albumin-to-creatinine ratio ≥30 mg/g. (c) Chronic renal urine albumin-to-creatinine ratio ≥30 mg/g. (c) Chronic renal failure is defined as an estimated glomerular filtration rate failure is defined as an estimated glomerular filtration rate <60 mL/min/1.73 m2 calculated using the CKD-EPI equation. <60 mL/min/1.73 m2 calculated using the CKD-EPI equation. (d) Cardiovascular disease included myocardial infarction, (d) Cardiovascular disease included myocardial infarction, ischaemic heart disease, cerebrovascular disease and ischaemic heart disease, cerebrovascular disease and peripheral arterial disease. (e) Albuminuria is defined as a peripheral arterial disease. (e) Albuminuria is defined as a urine albumin-to-creatinine ratio ≥30 mg/g. CKD-EPI,Chronic urine albumin-to-creatinine ratio ≥30 mg/g. CKD-EPI,Chronic Kidney Disease Epidemiology Collaboration. Kidney Disease Epidemiology Collaboration. Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 5 Open access Figure 4 Coprevalence of medical conditions among Figure 3 Prevalence of comorbidities in patients with type patients with type 2 diabetes. All values are percentages 2 diabetes stratified by gender. The proportions of patients and have been rounded off to the nearest whole number, with a particular comorbidity, regardless of the presence of except for values less than 1.0. CKD, chronic kidney disease; other comorbidities are shown. (a) Obesity is defined as a CRF, chronic renal failure; CVD, cardiovascular disease; UTI, body mass index ≥30 kg/m2. (b) Chronic kidney disease is urinary tract infection. defined as an estimated glomerular filtration rate <60 mL/ min/1.73 m2 calculated using the CKD-EPI equation and/or urine albumin-to-creatinine ratio ≥30 mg/g. (c) Chronic renal failure is defined as an estimated glomerular filtration rate <60 mL/min/1.73 m2 calculated using the CKD-EPI equation. (d) Cardiovascular disease included myocardial infarction, ischaemic heart disease, cerebrovascular disease and peripheral arterial disease. (e) Albuminuria is defined as a urine albumin-to-creatinine ratio ≥30 mg/g. CKD-EPI,Chronic Kidney Disease Epidemiology Collaboration. Comorbidities and glycaemic control Overall, 56% of the patients showed good glycaemic control. Glycaemic control was higher among patients who did not exhibit a comorbidity in a target organ, showing the greatest difference in patients without reti- nopathy compared with patients with retinopathy (58% vs 42%, respectively), who were considered to have good glycaemic control. The proportion of patients exhib- iting good glycaemic control decreased as the number of comorbidities in target organs increased (59%, 56%, 52% and 44% for patients with 0, 1, 2 and at least three comor- bidities in target organs, respectively). Figure 5 Frequency of cardiovascular disease, chronic renal failure and both medical conditions concomitantly in Discussion patients with type 2 diabetes. Note: Patients who presented Patients with type 2 diabetes have a high prevalence of both comorbidities, cardiovascular disease and chronic renal chronic comorbidities, particularly comorbidities related failure, are also included in the individual categories. CVD included myocardial infarction, ischaemic heart disease, to diabetes, such as hypertension, hyperlipidaemia, cerebrovascular disease and peripheral arterial disease; CRF obesity, chronic kidney disease, chronic renal failure and is defined as an estimated glomerular filtration rate <60 mL/ cardiovascular disease. These comorbidities also show a min/1.73 m2 calculated using the CKD-EPI equation. CKD, high coprevalence in these patients. The prevalence and chronic kidney disease; CKD-EPI, Chronic Kidney Disease coprevalence of chronic comorbidities in patients with Epidemiology Collaboration; CRF, chronic renal failure; CVD, type 2 diabetes increase with age. cardiovasculardisease. 6 Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 Open access hyperlipidaemia with obesity, cardiovascular disease and chronic kidney disease. For the individual comorbidities, the pattern reported in our study is similar to patterns described in previous studies in the USA using informa- tion from primary care physicians,4 from a health infor- matics company11 or from the databases of the Veteran Administration and Medicare.13 The pattern is also similar to the pattern reported in a population-based study conducted in Ontario (Canada)12 and a study conducted by the diabetes clinic of a university hospital in Belgium.19 The pattern observed in the subgroup of patients aged ≥75 years is also similar to the overall sample, but the frequency of coprevalence of chronic kidney disease with other cardiovascular risk factors or cardiovascular disease increases substantially; thus, in this subgroup of patients, chronic kidney disease was coprevalent with hypertension in 47% of patients, with hyperlipidaemia in 32%, with obesity in 20% and with cardiovascular disease in 19%. These results are difficult to put into perspective with find- ings from other studies since the occurrence of chronic Figure 6 Distribution of chronic renal failure stages in kidney disease was usually based on the presence of a patients with type 2 diabetes and cardiovascular disease. diagnosis. However, in our study, chronic kidney disease eGFR, estimated glomerular filtration rate. was based on the actual laboratory values for creatinine or the urine albumin-to-creatinine ratio. For instance, in A high proportion (31%) of patients had four or more the study conducted by Steinman et al using the Veteran comorbidities, increasing to 41% in patients aged ≥75 Administration and Medicare databases, chronic kidney years. Consistent results were reported by Iglay et al, who disease (recorded in their study as chronic renal insuffi- found that 49% of patients with T2DM aged 75 years ciency) appears underrepresented, with less than 20% of or older exhibited four or more chronic comorbidities the patients ≥75 years old showing chronic renal failure.13 in a study using a similar design with a larger database In our study, 50% of patients aged ≥75 years presented in the USA.4 In contrast to their results, we observed a this condition. The same issue and similar findings are higher prevalence of comorbidities among women; applicable to the study by Lin et al, who observed a prev- however, in our study, women were 4 years older than alence of chronic kidney disease of 9% in patients aged men. Meanwhile, in the study by Iglay et al, women were 65 years or older.11 Using the same definition of chronic slightly younger than men. The most frequent comorbid- kidney disease in patients from a diabetic clinic of a ities were hypertension (72%), hyperlipidaemia (60%), hospital in Belgium, Hermans and Dath identified a 46% obesity (45%), chronic kidney disease (33%), chronic prevalence of this condition; unfortunately, in this latter renal failure (28%) and cardiovascular disease (23%). study, the results were not stratified by age, and only data Iglay et al reported a higher prevalence of hypertension for the most frequent coprevalent comorbidities were (82%), overweight/obesity (78%) and hyperlipidaemia reported.19 (77%), a lower prevalence of chronic kidney disease Other doublets of frequent coprevalent comorbidities (24%) and a similar prevalence of cardiovascular disease in our study were obesity and chronic kidney disease, (22%).4 Differences in the prevalence values between our cardiovascular disease and chronic kidney disease, hyper- results and the study by Iglay et al may be explained by tension and retinopathy and hypertension and urinary the differences in the characteristics of the populations tract infection. The coprevalence of obesity and chronic from the USA and Spain, for instance, in the prevalence kidney disease and of hypertension and retinopathy in of obesity (38% and 18% for the USA and Spain, respec- patients with diabetes is well supported by the literature. tively25). Other studies conducted in various settings have Hypertension is an important factor contributing to the provided similar results, except for the results for chronic occurrence and progression of diabetic retinopathy,26 27 kidney disease, with a high frequency of hypertension, and tight control of blood pressure reduces the risk of metabolic disorders and cardiovascular disease.12 19 20 complications of this condition.28 Obesity is a risk factor Chronic kidney disease appears underrepresented in the for developing chronic kidney disease, even in patients other studies because the authors did not include a defi- without metabolic disease.29 nition based on the actual laboratory values but instead Chronic kidney disease and diabetes are risk factors for relied on the recorded diagnosis. cardiovascular diseases; therefore, comorbidity among The most frequent pairs of coprevalent disorders in our these three clinical conditions is frequent.30 Impor- study were the combination of hypertension and hyperlip- tantly, the presence of chronic kidney disease and/or idaemia and the combinations of both hypertension and cardiovascular disease is important in the selection of Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 7 Open access antihyperglycaemic treatment.31 32 In clinical practice, a diabetes,11 12 15 16 and their combined effects are respon- substantial proportion of patients with diabetes received sible, to a great extent, for their mortality.40 medications or doses that should be avoided in patients with these comorbid conditions.33–35 In fact, new treat- Acknowledgements This study was possible through the commitment of physicians and nurses working in the Catalan Health Institute who provide care ment guidelines from the American Diabetes Associa- to patients with diabetes. CIBER of Diabetes and Associated Metabolic Diseases tion/European Association for the Study of Diabetes for (CIBERDEM) is an initiative from Instituto de Salud Carlos III, Madrid, Spain. The the management of type 2 diabetes consider the presence authors thank Antón Gómez, Karine Ferreira and Gonzalo Fernández (Merck Sharp & of cardiovascular disease and chronic renal failure as key Dohme Spain) for their input on different aspects of the study design and analysis, and Fernando Rico-Villademoros (COCIENTE S.L., Madrid, Spain) for providing elements for the selection of second-line treatment.36 medical writing assistance. This problem increases in very old patients; thus, 50% Contributors MM-C, JF-N and DM contributed to the study design. JR was of patients ≥75 years old had chronic renal failure, as involved in data management and statistical analyses. MM-C wrote the first draft evaluated with the estimated glomerular filtration rate, of the manuscript. MM-C, JF-N, JR, MC and DM contributed to the analysis and greater than 30% had cardiovascular disease, and approx- interpretation of the data, provided critical input during the development of the imately 16% exhibited both conditions in the present manuscript and approved the final version for submission. MM-C had full access to all data in the study and takes responsibility for the integrity of data and the study. Finally, fewer studies have examined the relation- accuracy of the data analysis. ship between hypertension and urinary tract infection in Funding This study was supported by the Institut Universitari d’Investigació en patients with diabetes. Patients with diabetes appear to be Atenció Primària Jordi Gol (IDIAP Jordi Gol), and partially funded and sponsored by at increased risk of urinary tract infection compared with Merck Sharp & Dohme Spain. healthy controls, and this comorbidity also appears to Competing interests MM-C has received advisory honoraria from Astra-Zeneca, be associated with renal damage.37 In a descriptive study Bayer, Boehringer Ingelheim, GSK, Lilly, MSD, NOVARTIS, NovoNordisk and Sanofi. conducted in 1000 patients with diabetes, one-quarter He has received speaker honoraria from Astra-Zeneca, Bayer, Boehringer Ingelheim, GSK, Lilly, Menarini, MSD, Novartis, NovoNordisk and Sanofi. He has received had a urinary tract infection, and the factors associated research grants to the institution from Astra-Zeneca, GSK, Lilly, MSD, Novartis, with the presence of this condition were hypertension, NovoNordisk and Sanofi. He has received research grants from Institut Universitari insulin therapy and nephropathy38; our results are consis- d’Investigació en Atenció Primària Jordi Gol (IDIAP Jordi Gol)(Barcelona, Spain), tent with these findings. Instituto de Salud Carlos III (Madrid, Spain), Generalitat de Catalunya, Peris 2016- 2020 and The Strategic Plan for Health Research and Innovation (Barcelona, Spain). Our study has several limitations. We only included Josep Franch-Nadal has received advisory and or speaking fees from Astra-Zeneca, certain comorbidities; therefore, our results only reflect a Ascensia, Boehringer Ingelheim, GSK, Lilly, MSD, Novartis, NovoNordisk and Sanofi; snapshot of diabetes-related comorbidities. Since most of he has received research grants to the institution from Astra-Zeneca, GSK, Lilly, our diagnoses were based on data recorded in the data- MSD, Novartis, NovoNordisk, Sanofi and Boehringer. Jordi Real has no conflict of interest to declare. Marta Cedenilla is full-time employee of Merck Sharp & Dohme base, a misclassification cannot be excluded. However, Spain.Dídac Mauricio has received advisory and or speaking fees from Astra- the validity of the SIDIAP database for studying the prev- Zeneca, Ascensia, Boehringer Ingelheim, GSK, Lilly, MSD, Novartis, NovoNordisk and alence of cardiovascular risk factors has been confirmed Sanofi; he has received research grants to the institution from Astra-Zeneca, GSK, in a previous study.39 For some outcomes, the proportion Lilly, MSD, Novartis, NovoNordisk, Sanofi and Boehringer. of missing data was high, such as albuminuria with 38% Patient consent for publication Not required. missing data. The relevant data may have been available Ethics approval The study was approved by the Ethics Committee of the Primary for patients because the laboratory test was requested by Healthcare University Research Institute (IDIAP) Jordi Gol (Barcelona, Spain). their physicians due to the suspicion of their being at a Provenance and peer review Not commissioned; externally peer reviewed. greater risk of developing those conditions; therefore, Data availability statement Data are available upon reasonable request. our study may have overestimated the prevalence of diag- Open access This is an open access article distributed in accordance with the noses based on those laboratory values. The strengths of Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which our study include a population-based design, the use of permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is a primary care registry with a large number of subjects properly cited, appropriate credit is given, any changes made indicated, and the use managed under real world conditions and, unlike other is non-commercial. See: http://​creativecommons.​org/​licenses/​by-​nc/​4.​0/. population-based studies, including previous studies from our group analysing the SIDIAP database, the HbA1c levels and glomerular filtration rates were available for References 1 Buttorff C, Ruder T. Multiple chronic conditions in the United States. approximately 90% of patients. Santa Monica, CA: RAND Coorporation, 2017. 2 Heide vander I, Snoeijs S, Melchiorre MG, et al. Innovating care for people with multiple chronic conditions in Europe. NIVEL (Netherlands Institute for Health Services Research): Utrecht, 2015. 3 Hopman P, Heins MJ, Korevaar JC, et al. Health care utilization Conclusions of patients with multiple chronic diseases in the Netherlands: differences and underlying factors. Eur J Intern Med 2016;35:44–50. Consistent with previous studies, our findings show that 4 Iglay K, Hannachi H, Joseph Howie P, et al. Prevalence and co- patients with diabetes have a high frequency of coprev- prevalence of comorbidities among patients with type 2 diabetes alent metabolic risk factors, cardiovascular disease and mellitus. Curr Med Res Opin 2016;32:1243–52. 5 An J, Le QA, Dang T. Association between different types of chronic kidney disease. These findings again highlight comorbidity and disease burden in patients with diabetes. J Diabetes the importance of implementing an integrated approach 2019;11:65–74. 6 Lin P-J, Pope E, Zhou FL. Comorbidity type and health care costs in for managing all these conditions since they are respon- type 2 diabetes: a retrospective claims database analysis. Diabetes sible for a substantial proportion of the burden of Ther 2018;9:1907–18. 8 Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 Open access 7 Alonso-Morán E, Orueta JF, Fraile Esteban JI, Esteban JIF, et al. The 24 Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate prevalence of diabetes-related complications and multimorbidity in glomerular filtration rate. Ann Intern Med 2009;150:604–12. the population with type 2 diabetes mellitus in the Basque country. 25 Organisation for Economic Co-operation and Development. Obesity BMC Public Health 2014;14:1059. update 2017. Paris, France: OECD, 2017. 8 Nowakowska M, Zghebi SS, Ashcroft DM, et al. The comorbidity 26 Yamazaki D, Hitomi H, Nishiyama A. Hypertension with diabetes burden of type 2 diabetes mellitus: patterns, clusters and predictions mellitus complications. Hypertens Res 2018;41:147–56. from a large English primary care cohort. BMC Med 2019;17:145. 27 Wat N, Wong RL, Wong IY. Associations between diabetic 9 Morgan SA, Ali MM, Channon AA, et al. Prevalence and correlates retinopathy and systemic risk factors. Hong Kong Med J of diabetes and its comorbidities in four Gulf cooperation Council 2016;22:589–99. countries: evidence from the world health survey plus. J Epidemiol 28 Matthews DR, Stratton IM, Aldington SJ, et al. Risks of progression Community Health 2019;73:630–6. of retinopathy and vision loss related to tight blood pressure 10 Piette JD, Kerr EA. The impact of comorbid chronic conditions on control in type 2 diabetes mellitus: UKPDS 69. Arch Ophthalmol diabetes care. Diabetes Care 2006;29:725–31. 2004;122:1631–40. 11 Lin P-J, Kent DM, Winn A, et al. Multiple chronic conditions in type 2 29 Nehus E. Obesity and chronic kidney disease. Curr Opin Pediatr diabetes mellitus: prevalence and consequences. Am J Manag Care 2018;30:241–6. 2015;21:e23–34. 30 Suckling R, Gallagher H, disease Ckidney. Chronic kidney 12 Gruneir A, Markle-Reid M, Fisher K, et al. Comorbidity burden and disease, diabetes mellitus and cardiovascular disease: risks and health services use in community-living older adults with diabetes commonalities. J Ren Care 2012;38 Suppl 1:4–11. mellitus: a retrospective cohort study. Can J Diabetes 2016;40:35–42. 31 American Diabetes Association. 8. Pharmacologic Approaches to 13 Steinman MA, Lee SJ, John Boscardin W, et al. Patterns Glycemic Treatment: Standards of Medical Care in Diabetes—2018. of multimorbidity in elderly veterans. J Am Geriatr Soc Diabetes Care 2018;41:S73–85. 2012;60:1872–80. 32 Menéndez SA. Actualización del algoritmo de hiperglucemia 2017. 14 Magnan EM, Palta M, Mahoney JE, et al. The relationship of Diabetes práctica 2017:57–60. individual comorbid chronic conditions to diabetes care quality. BMJ 33 Patel PA, Liang L, Khazanie P, et al. Antihyperglycemic medication Open Diabetes Res Care 2015;3:e000080. use among Medicare beneficiaries with heart failure, diabetes 15 Kerr EA, Heisler M, Krein SL, et al. Beyond comorbidity counts: mellitus, and chronic kidney disease. Circ Heart Fail 2016;9:e002638. how do comorbidity type and severity influence diabetes patients' 34 Clemens KK, Liu K, Shariff S, et al. Secular trends in treatment priorities and self-management? J Gen Intern Med antihyperglycaemic medication prescriptions in older adults with 2007;22:1635–40. diabetes and chronic kidney disease: 2004-2013. Diabetes Obes 16 Beverly EA, Wray LA, Chiu CJ, et al. Perceived challenges and Metab 2016;18:607–14. priorities in co-morbidity management of older patients with type 2 35 Huang H, Shetty S, Bauer E, et al. Concordance with prescribing diabetes. Diabet Med 2011;28:781–4. information dosage recommendations for dipeptidyl-peptidase-4 17 Terauchi Y, Ozaki A, Zhao X, et al. Humanistic and economic burden inhibitors among type 2 diabetes mellitus patients with of cardiovascular disease related comorbidities and hypoglycaemia moderate to severe chronic kidney disease. Curr Med Res Opin among patients with type 2 diabetes in Japan. Diabetes Res Clin 2018;34:1021–7. Pract 2019;149:115–25. 36 Davies MJ, D'Alessio DA, Fradkin J, et al. Management of 18 Yokomichi H, Nagai A, Hirata M, et al. Survival of macrovascular hyperglycemia in type 2 diabetes, 2018. A consensus report disease, chronic kidney disease, chronic respiratory disease, cancer by the American diabetes association (ADA) and the European and smoking in patients with type 2 diabetes: Biobank Japan cohort. association for the study of diabetes (EASD). Diabetes Care J Epidemiol 2017;27:S98–106. 2018;41:2669–701. 19 Hermans MP, Dath N. Prevalence and co-prevalence of comorbidities 37 Goswami R, Bal CS, Tejaswi S, et al. Prevalence of urinary tract in Belgian patients with type 2 diabetes mellitus: a transversal, infection and renal scars in patients with diabetes mellitus. Diabetes descriptive study. Acta Clin Belg 2018;73:68–74. Res Clin Pract 2001;53:181–6. 20 Teljeur C, Smith SM, Paul G, et al. Multimorbidity in a cohort of 38 Al-Rubeaan KA, Moharram O, Al-Naqeb D, et al. Prevalence of patients with type 2 diabetes. Eur J Gen Pract 2013;19:17–22. urinary tract infection and risk factors among Saudi patients with 21 Doos L, Roberts EO, Corp N, et al. Multi-Drug therapy in chronic diabetes. World J Urol 2013;31:573–8. condition multimorbidity: a systematic review. Fam Pract 39 Ramos R, Balló E, Marrugat J, et al. Validity for use in research 2014;31:654–63. on vascular diseases of the SIDIAP (information system for the 22 Bolíbar B, Fina Avilés F, Morros R, et al. [SIDIAP database: electronic development of research in primary care): the EMMA study. Rev Esp clinical records in primary care as a source of information for Cardiol 2012;65:29–37. epidemiologic research]. Med Clin 2012;138:617–21. 40 Global Burden of Metabolic Risk Factors for Chronic Diseases 23 Mata-Cases M, Franch-Nadal J, Real J, et al. Glycaemic control and Collaboration. Cardiovascular disease, chronic kidney disease, and antidiabetic treatment trends in primary care centres in patients with diabetes mortality burden of cardiometabolic risk factors from 1980 type 2 diabetes mellitus during 2007-2013 in Catalonia: a population- to 2010: a comparative risk assessment. Lancet Diabetes Endocrinol based study. BMJ Open 2016;6:e012463. 2014;2:634–47. Mata-Cases M, et al. BMJ Open 2019;9:e031281. doi:10.1136/bmjopen-2019-031281 9