Approach to patients with diabetes and obesity in primary care

Aims The aim of this study is to analyse the effect of pharmacological and non-pharmacological treatment on weight control in patients with diabetes and obesity. Design Epidemiological, descriptive, cross-sectional study. Site Primary care. In 11 health centres in Málaga and Cádiz during April and October 2022. Participants 281 patients over 18 years old with type 2 diabetes and obesity are included. Main measurements Socio-demographics, clinical, treatment and lifestyle habits variables were obtained from medical records and personal interview. Descriptive statistics were obtained for continuous variables. Statistical tests were performed based on the nature of the variables. Results Variables like marital status, level of education and occupation, and smoking habit, shows differences regarding the sex (p < 0.05). 82.3% of those who received education lost weight, compared to 67.5% of lost weight who received no health education (p = 0.004). GLP1 and SGLT2 were more commonly prescribed for women (p = 0.048), and SGLT2 more commonly prescribed for men (p = 0.047). Patients taking GLP1, SGLT2 or both, regardless of sex, weight loss during the study period was −3.1 kg (SE: 0.60), while the loss of those who took other medications was −1.33 kg (SE: 0.62). The mean difference was 1.75 kg (p = 0.046). Conclusions In terms of weight loss, obese diabetics who took GLP1, SGLT2 or both were 2.5 times more likely to lose weight than those who did not. Healthy lifestyle choices are key to weight loss in obese diabetic patients.


Introduction
Type 2 diabetes is a metabolic disease characterised by hyperglycaemia and the development of microvascular and macrovascular complications.In addition, a large proportion of those affected are obese. 1The age and sex-adjusted incidence in Spain is 11.6/1000 people per year.The most common is type 2 diabetes, which is associated with obesity in 45% of cases (SIDIAP study). 2,3he first therapeutic measure recommended by national and international guidelines is to make lifestyle changes, including a healthy diet, weight control, physical exercise, not smoking, and emotional well-being, all of which are essential in diabetes management. 4Therapeutic measures to treat diabetes also include pharmacological interventions to control blood glucose levels and help obese people lose weight.Molecules with proven efficacy should also be recommended. 5 recent study looked at the impact a healthy lifestyle had on people with diabetes.The results were low and did not specifically address clinical situations such as obesity which require increased enforcement of these recommendations. 6Pharmacological interventions by healthcare professionals (CAPTURE study) also suggest that there is room for improvement in the use of drugs with cardiovascular and weight-loss benefits. 7he aim of this study is to analyse the effect of pharmacological and non-pharmacological treatment on weight control in patients with diabetes and obesity.

Settings
An epidemiological, descriptive, cross-sectional, multicentre, regional study was conducted under routine clinical practice conditions.
Twenty physicians from 11 health centres in the provinces of Malaga and Cadiz participated.The inclusion criteria were: over 18 years of age; diagnosis of type 2 diabetes and obesity (BMI > 30), and having visited their primary care physician at least once in the last five years.All participants provided written informed consent.
Patients were recruited by consecutive sampling of obese type 2 diabetics who visited a doctor for any reason between April and October 2022.Data were collected from their medical records and personal interviews.In order to carry out the assessment, we required a blood test and weight control of the patient that had been taken within the last 6 months, and at least another for a period of no less than 1 year and no more than the previous 5 years to be able to make the comparison.
Patients attended a single visit which included an interview to obtain information not recorded in their medical history and to assess their current weight, lifestyle habits, exercise, treatment adherence, current pharmacological treatment, and its compliance with current guidelines.Patients were informed that their data were confidential.
To enable data collection and statistical analysis, clinical and socio-demographic variables were included in a specially designed online questionnaire that respected the anonymity and confidentiality of the individuals.
To calculate the sample size, we estimated population of approximately 30,000 people, in which 2671 people were diagnosed with diabetes (prevalence 8.903%) and in which 1200 people were estimated to have a BMI > 30.The required sample size was 280 participants (GRANMO 7.12 calculator, Institut Municipal d'Investigació Mèdica, Barcelona, Spain), 95% confidence level, 5% precision, 10% expected loss.

Patient assessments
Available data on the following variables were collected from the medical records: Socio-demographics: age, sex, duration of DM, level of education, employment status, marital status.Healthy lifestyle habits: Physical activity.In order to measure adherence to exercise, we asked patients the question: ''Do you do at least 30 min of moderate physical activity (e.g., brisk walking) five or more days per week?''.This allowed us to identify patients who were following the recommendations and those who were not. 8motional well-being.Three questions were asked regarding the emotional aspects of the IMEVID questionnaire. 9linical: weight, height, BMI, waist circumference, blood pressure, basal glycaemia, glycosylated haemoglobin, lipid profile, toxic habits (tobacco consumption), diabetes complications, and associated comorbidities.Patients were asked where they received follow-up care (primary/specialist, public/private) and whether they had received any type of health education.It is carried out by nurses in individual or group interventions, with indications are given to modify lifestyles (nutrition and exercise), through the supports that the health system provides us.The main dependent variable was weight loss, with a recommended weight loss of at least 5% for diabetes control. 5reatment used: type of hypoglycaemic treatment, polymedication (more than 6 drugs), treatment adherence using the Haynes---Sackett test, 10 and the adequacy of pharmacological treatment (according to the current recommendations of the guidelines and answered by the treating physician).Descriptive statistics were obtained for continuous variables: mean and standard deviation or standard error (SD/SE).For qualitative variables: absolute and relative frequencies.

Analysis
Statistical tests were performed based on the nature of the variables.Chi-squared test (Fisher's exact test when required) was used to examine the relationship between categorical variables.Student's t-tests or ANOVA were used for tests with quantitative variables when normality was met, otherwise U-Mann---Whitney and Kruskal---Wallis were used.A significance level of ˛ = 0.05 was used for all statistical tests.
Logistic regression analysis: Logistic regression models allow us to analyse the results in both explanatory and predictive terms.We can determine the strength of association by means of the odds ratios of the predictor variables with the effect studied independently and determine the predictive value of each of them or of the model as a whole. 11,12n our study, we analysed which model best fits the qualitative dependent variable (weight loss ---yes/no) using the logistic regression method based on data from other known quantitative or qualitative variables acting as explanatory variables.The results of the independent variables are usually presented as odds ratios (OR) with 95% confidence intervals (CI).
The study was approved by the Cadiz Research Ethics Committee (PEIBA code 1125-N-22, reg.: 74.22).Study flowchart: Epidemiological, descriptive, crosssectional, multicentre, regional study to analyse the effects of pharmacological and non-pharmacological treatment on weight control in patients with diabetes and obesity.

Socio-demographic and clinical variables, and healthy lifestyle habits
281 patients were included with a mean age of 67.34 years (SD ± 10.46 years) and a mean diabetes duration of 10.30 years (SD ± 7.29 years).55% of the patients were male, 73.3% were married, 43.8% had completed primary edu- The mean weight of the patients was 90.81 ± 15.32 kg.The mean height was 163.56 ± 9.73 cm.Waist circumference was 113.78 ± 10.44 cm.BMI was 33.85 ± 4.71.There were statistically significant differences (p < 0.005) between men and women in all variables, except waist circumference.The socio-demographic and clinical characteristics of the patients included in the study are shown in Table 1.Regarding the sex of the patients, we found statistically significant differences (p < 0.05) in the following variables: marital status, level of education and occupation, and smoking habit, especially in former smokers.
The clinical variables of the patients included in the study are analysed in Table 2. Regarding the patients' medical history (comorbidities), statistically significant differences were found only for neuropathies and ischaemic heart disease----more frequent in men----and anxiety/depression----more frequent in women.

Treatment variables
Most patients control their disease with monotherapy (32.7%) followed by dual and triple therapy (27.8% and 18.1%, respectively).Other patients (21.4%) haven't got pharmacological treatment, only a diet prescribed.54.4% of obese diabetics are treated with GLP1 (8.9%) or SGLT2 (27%) or both (18.5%).There were significant differences regarding patients' sex and the use of GLP1 and SGLT2, with GLP1 and SGLT2 more commonly prescribed for women (p = 0.048), and SGLT2 more commonly prescribed for men (p = 0.047).Fifty-eight percent of patients were polymedicated and 90% reported good adherence.68.7% of the physicians believed the treatment provided was in line with the current recommendations.
No statistically significant differences were found between the sex of the patient and disease progression, therapeutic education received, pattern of antidiabetic medication used, or polymedication (Table 1).
Regarding the types of treatment and some of the patients' clinical histories, there were statistically significant differences (Pearson's Chi-squared, p < 0.05) in the type of treatment in all patients analysed, as shown in Fig. 1.
Regarding the level of education, no statistically significant differences were found between the patients' weight losses, although a trend was observed, so that the differences between a university education and no education were statistically significant (p < 0.05).
With regard to the health education offered to these patients, it should be noted that 67.5% of those who received no education (n = 67) lost weight, while 82.3% of those who received education lost weight, and these differences were statistically significant (p = 0.004).
In patients taking appropriate medications (GLP1, SGLT2 or both), regardless of patient sex, weight loss during the study period was −3.1 kg (SE: 0.60), while those taking other medications experienced a weight loss of −1.33 kg (SE: 0.62).There were statistically significant differences (p = 0.046).The mean difference was 1.75 kg.
When comparing weight loss (yes/no) with taking or not taking GLP1 and SGLT2 drugs, statistically significant differences were found, p = 0.009 (Fisher's exact test).59.3% of patients who lost weight were prescribed GLP1, SGLT2 or both, while 40.7% were prescribed other treatments.

Logistic regression analysis
Table 3 shows the statistics for the variables included in each model.

Dependent variable
Any weight loss (yes/no): The model classified patients who DO lose weight with an accuracy rate of 92.4%.It included the following variables: sex, health education, and medication (GLP1 + SGLT2).Thus: Women are 1.677 times more likely to lose weight than men.Patients who received health education were 2.416 times more likely to lose weight than those who did not.Subjects whose treatment included  GLP1 + SGLT2 were 2.495 times more likely to lose weight than those not taking these drugs.Female subjects who receive health education at their health centre and take GLP1 + SGLT2, are 83% more likely to lose weight, while those who are male, do not receive health education and do not take the medication are 32% more likely to lose weight.

Dependent variable
Weight loss greater than 5% (yes/no): The model classified patients who did NOT lose 5% or more weight with an accuracy rate of 100%.It included the following variables: health education and taking both medications (GLP1 + SGLT2).Thus: Patients who received health education were 2.634 times more likely to lose more than 5% more weight than those who did not.Subjects whose treatment included GLP1 + SGLT2 are 2.505 times more likely to lose more than 5% of their body weight than those not taking these drugs.

Discussion
The aim of the study was to determine the efficacy of the interventions carried out by a public service (Andalusian Health Service) in obese diabetic patients over a five-year period, in terms of their impact on weight and the follow-up by health professionals of the pharmacological recommendations made by the diabetes care process in Andalusia, 13 and by national and international guidelines. 5,13As such, the interest of this study stems from the lack of data assessing the effectiveness of healthcare professionals' actions on weight.
In our study, 76.2% had received health education, 39.1% were physically active, 87.5% did not smoke, 90% adhered to the prescribed treatment, and 84.7% considered their emotional well-being satisfactory.We find similar data if we compare our study with other results obtained in Spain, 14 where 88.3% had received individual or group therapeutic education, 45.8% were physically active, 87.6% were non-smokers, 62.8% considered their emotional well-being satisfactory and just over 20% of patients followed a Mediterranean diet.The NHANES III survey 15 in the USA assessed weight, exercise, tobacco use, and fruit and vegetable intake----the data found were similar to those of the Spanish study. 14Both studies were conducted in a diabetic population aged > 18 years in which less than 50% of the sample was obese.
It is noteworthy that only 27% of patients have managed to reduce their weight by at least 5% in the last five years, which is recommended for diabetes control.However, if we look at a weight loss of >2%, we see that 47% of patients have achieved this.
In terms of treatment, GLP1 and SGLT2 are more commonly prescribed in women and SGLT2 in men, and even more so in people over 75 years of age, possibly because of the higher prevalence of ischaemic heart disease and the relationship between heart failure and age.According to the recommendations of national and international guidelines, the adequacy of treatment in people with diabetes and obesity is 54.4%, while 68.7% of health professionals consider treatment to be adequate.Regarding the comorbidities associated with diabetes and obesity, the adequacy of treatment for cardiovascular disease according to guidelines is 71.2% a figure much higher than that found in the CAPTURE study, 7 which was 21.5%.
Treatment with guideline-recommended weight-loss drugs is adequate in just over half of patients, and in patients with associated cardiovascular disease this figure is 71.4%, much higher than in previous studies.
About the limitations of the study, the patients were selected from a population receiving care by means of non-stratified consecutive sampling, with the consequent selection bias.Researchers took great care in collecting the data, although we cannot rule out the possibility that respondents may have been less than truthful in answering the questions posed.It is a cross-sectional observational study, suitable for understanding changes in lifestyle and weight, pharmacological treatment, its adaptation to national and international guidelines, and the clinical characteristics of obese diabetic patients seen in primary care in Spain, but it is not applicable to other populations or health systems.

Conclusion
In terms of weight loss, obese diabetics who took GLP1, SGLT2 or both were 2.5 times more likely to lose weight than those who did not.Healthy lifestyle choices are key to weight loss in obese diabetic patients.
was performed with SPSS Statistics 26.0 software licensed to the University of Malaga (IBM Corp. Released 2019.IBM SPSS Statistics for Windows, Version 26.0.Armonk, NY: IBM Corp).
Patient s with type 2 diabetes and obesity <18 yearsHaving visited their primary care physician at least once in the last five years.

Figure 1
Figure 1 Bivariate analysis: clinical history and treatment.

Table 1
Obese diabetic patients: sociodemographic and clinical variables.

Table 2
Obese diabetic patients: clinical variables.

Table 3
Descriptive factors of the logistic regression model.