RESUMO
Diabetes mellitus is a disease with no cure that can cause complications and even death. Moreover, over time, it will lead to chronic complications. Predictive models have been used to identify people with a tendency to develop diabetes mellitus. At the same time, there is limited information regarding the chronic complications of patients with diabetes. Our study is aimed at creating a machine-learning model that will be able to identify the risk factors of a diabetic patient developing chronic complications such as amputations, myocardial infarction, stroke, nephropathy, and retinopathy. The design is a national nested case-control study with 63,776 patients and 215 predictors with four years of data. Using an XGBoost model, the prediction of chronic complications has an AUC of 84%, and the model has identified the risk factors for chronic complications in patients with diabetes. According to the analysis, the most crucial risk factors based on SHAP values (Shapley additive explanations) are continued management, metformin treatment, age between 68 and 104 years, nutrition consultation, and treatment adherence. But we highlight two exciting findings. The first is a reaffirmation that high blood pressure figures across patients with diabetes without hypertension become a significant risk factor at diastolic > 70 mmHg (OR: 1.095, 95% CI: 1.078-1.113) or systolic > 120 mmHg (OR: 1.147, 95% CI: 1.124-1.171). Furthermore, people with diabetes with a BMI > 32 (overall obesity) (OR: 0.816, 95% CI: 0.8-0.833) have a statistically significant protective factor, which the paradox of obesity may explain. In conclusion, the results we have obtained show that artificial intelligence is a powerful and feasible tool to use for this type of study. However, we suggest that more studies be conducted to verify and elaborate upon our findings.
Assuntos
Diabetes Mellitus , Hipertensão , Metformina , Humanos , Idoso , Idoso de 80 Anos ou mais , Pressão Sanguínea , Metformina/uso terapêutico , Estudos de Casos e Controles , Inteligência Artificial , Hipertensão/complicações , Hipertensão/tratamento farmacológico , Hipertensão/epidemiologia , Fatores de Risco , Obesidade/complicações , Aprendizado de MáquinaRESUMO
OBJECTIVE: To evaluate Mexico's national Integrated Management of Diabetes in Stages (Manejo Integral de la Diabetes por Etapas, MIDE) program using three types of indicators: process, structure, and impact. METHODS: A cross-sectional study was conducted using data for 97 452 people with diabetes (PWD) who participated in the MIDE patient empowerment program (PEP) at "MIDE modules" (standardized diabetes health care units) at Mexico's Institute for Social Security and Services for State Workers (Instituto de Seguridad y Servicios Sociales de los Trabajadores del Estado, ISSSTE) hospital clinics and family medical clinics nationwide between 2007 and 2014. The program promotes diabetes patient empowerment and self-care through outpatient consultations with a multidisciplinary health care team supported by continuous training. Baseline data were compared with results post-program for the following indicators: process (metabolic control (MetC), based on glycated hemoglobin (HbA1c), triglyceride (TG), and total cholesterol (TC) levels); structure (number of MIDE modules installed at ISSSTE clinics and number of patients/health personnel accredited as diabetes experts/awarded diplomas); and impact (average number of patient illness days (IDs) and hospitalization episodes (HEs) per PWD over a 12-month period). RESULTS: Over the seven-year study period, the proportion of patients with MetC (HbA1c < 7.0%, TG < 150 mg/dL, and TC < 200 mg/dL) increased significantly (from 35.4% to 60% (with a peak level of 62% in 2013); P < 0.001); average HbA1c, triglycerides, and total cholesterol per PWD dropped by 25%, 31%, and 11% respectively; average number of IDs and HEs per PWD over a 12-month period dropped by 38% and 41% respectively; a total of 140 MIDE modules were installed at ISSSTE clinics; and a total of 1 117 diplomas were awarded to 826 health professionals, and 2 613 PWD were accredited as "patient experts in diabetes." CONCLUSIONS: The MIDE PEP is feasible, usable, and acceptable to PWD. The program improves MetC; reduces the frequency of IDs and HEs; and facilitates patient participation, the involvement of health personnel, and shared decision-making.
RESUMO
ABSTRACT Objective To evaluate Mexico's national Integrated Management of Diabetes in Stages (Manejo Integral de la Diabetes por Etapas, MIDE) program using three types of indicators: process, structure, and impact. Methods A cross-sectional study was conducted using data for 97 452 people with diabetes (PWD) who participated in the MIDE patient empowerment program (PEP) at "MIDE modules" (standardized diabetes health care units) at Mexico's Institute for Social Security and Services for State Workers (Instituto de Seguridad y Servicios Sociales de los Trabajadores del Estado, ISSSTE) hospital clinics and family medical clinics nationwide between 2007 and 2014. The program promotes diabetes patient empowerment and self-care through outpatient consultations with a multidisciplinary health care team supported by continuous training. Baseline data were compared with results post-program for the following indicators: process (metabolic control (MetC), based on glycated hemoglobin (HbA1c), triglyceride (TG), and total cholesterol (TC) levels); structure (number of MIDE modules installed at ISSSTE clinics and number of patients/health personnel accredited as diabetes experts/awarded diplomas); and impact (average number of patient illness days (IDs) and hospitalization episodes (HEs) per PWD over a 12-month period). Results Over the seven-year study period, the proportion of patients with MetC (HbA1c < 7.0%, TG < 150 mg/dL, and TC < 200 mg/dL) increased significantly (from 35.4% to 60% (with a peak level of 62% in 2013); P < 0.001); average HbA1c, triglycerides, and total cholesterol per PWD dropped by 25%, 31%, and 11% respectively; average number of IDs and HEs per PWD over a 12-month period dropped by 38% and 41% respectively; a total of 140 MIDE modules were installed at ISSSTE clinics; and a total of 1 117 diplomas were awarded to 826 health professionals, and 2 613 PWD were accredited as "patient experts in diabetes." Conclusions The MIDE PEP is feasible, usable, and acceptable to PWD. The program improves MetC; reduces the frequency of IDs and HEs; and facilitates patient participation, the involvement of health personnel, and shared decision-making.
RESUMEN Objetivo: Evaluar el programa nacional mexicano "Manejo Integral de la Diabetes por Etapas" (MIDE) usando tres tipos de indicadores: proceso, estructura e impacto. Métodos Se realizó un estudio transversal con los datos de 97 452 personas con diabetes que participaron entre el 2007 y el 2014 en el programa de empoderamiento de pacientes de MIDE que se llevó a cabo en los "módulos del MIDE" (unidades estandarizadas de atención de salud de diabetes) en el Instituto de Servicios y de Seguridad Sociales de los Trabajadores del Estado (ISSSTE) de México, además de en consultorios de hospitales y centros de medicina familiar de todo el país. El programa promueve el empoderamiento y el autocuidado en los pacientes con diabetes mediante consultas ambulatorias con un equipo de atención de salud multidisciplinario complementadas con talleres de capacitación continua. Se compararon los datos iniciales y los resultados al final del programa con respecto a los siguientes indicadores: proceso (control metabólico basado en análisis de los niveles de glucohemoglobina [HbA1c], triglicéridos y colesterol total); estructura (número de módulos del MIDE instalados en clínicas del ISSSTE y número de pacientes y personal de salud acreditado como experto en diabetes o que ha obtenido un certificado); e impacto (número promedio de días enfermos de los pacientes y número de hospitalizaciones por persona con diabetes durante un período de 12 meses). Resultados En el período de estudio de siete años, la proporción de pacientes con control metabólico (HbA1c < 7,0%, triglicéridos < 150 mg/dL y colesterol total < 200 mg/dL) aumentó significativamente (de 35,4% a 60%, con un nivel máximo de 62% en 2013; P < 0,001); el nivel promedio de HbA1c, triglicéridos y colesterol total por persona con diabetes descendió en 25%, 31% y 11% respectivamente; el número promedio de días enfermos y número de hospitalizaciones por persona con diabetes durante un período de 12 meses descendió en 38% y en 41% respectivamente; se instalaron un total de 140 módulos del MIDE en clínicas del ISSSTE, y se otorgaron un total de 1117 certificados a 826 profesionales de la salud y 2613 personas con diabetes fueron acreditados como "paciente experto en diabetes". Conclusiones El programa de empoderamiento de pacientes del MIDE es un programa factible y aceptable para utilizarse con las personas con diabetes. El programa mejora los valores del control metabólico; reduce la frecuencia de días enfermos y número de hospitalizaciones; y facilita la participación tanto de los pacientes como del personal de salud, así como la toma de decisiones compartida.
RESUMO Objetivo Avaliar o Programa de manejo integral de diabetes por etapas (MIDE) no México segundo três tipos de indicadores: processo, estrutura e impacto. Métodos Estudo transversal realizado com dados de 97.452 pessoas com diabetes (PD) que participaram do Programa MIDE de empoderamento do paciente em "módulos" designados (unidades de saúde que prestam atenção padronizada à diabetes) de ambulatórios hospitalares e centros de medicina da família do Instituto de Serviços e Seguridade Social dos Trabalhadores do Estado (ISSSTE) entre 2007 e 2014. O programa promove o empoderamento e o autocuidado dos pacientes com diabetes realizando consultas ambulatoriais com uma equipe de saúde multidisciplinar que recebe capacitação contínua. Os dados de referência foram comparados aos resultados pós-programa para os seguintes indicadores: processo (controle metabólico segundo os níveis de hemoglobina glicada [HbA1c], triglicerídeos [TG] e colesterol total [CT]); estrutura (número de módulos do MIDE instalados nos ambulatórios do ISSSTE e número de pacientes/pessoal de saúde credenciados como especialistas/diplomados em diabetes) e impacto (número médio de dias de doença do paciente [DD] e internações hospitalares (IH) por PD por um período de 12 meses). Resultados No período de estudo de sete anos, o percentual de pacientes com controle metabólico (HbA1c <7,0%, TG <150 mg/dl e CT <200 mg/dl) aumentou significativamente (de 35,4% a 60%, atingindo o ponto mais alto de 62% em 2013; P < 0,001). Houve redução dos níveis médios de HbA1c, TG e CT por PD de 25%, 31% e 11%, respectivamente. O número médio de DD e IH por PD em um período de 12 meses teve queda de 38% e 41%, respectivamente. Ao todo, foram instalados 140 módulos do Programa MIDE nos ambulatórios do ISSSTE, 1.117 diplomas foram concedidos a 826 profissionais da saúde e 2.613 PD foram credenciados como "pacientes especialistas em diabetes". Conclusões O programa MIDE de empoderamento do paciente é viável, utilizável e aceitável às PD. Promove melhor controle metabólico da diabetes, reduz o número de DD e IH e facilita a participação de pacientes, o envolvimento do pessoal da saúde e a tomada de decisão conjunta.