Incorporating Objective Measures of Sedentary Behaviour Into the Detection and Control Methods of Type 2 Diabetes Mellitus in Office Employees: Development of a Mathematical Model for Clinical Practice.

Abstract

Abstract Introduction : Type 2 diabetes mellitus (DM2) is one of the main public health threats of the 21st century. Identifying and predicting DM2 is the first step to stop its progression, and new strategies with low-cost, non-invasive early detection systems must be urgently implemented. Sedentary behaviour (SB) is one of the risk factors leading to the current increase in the prevalence of DM2, so incorporating the SB pattern into the detection methods of DM2 is essential. Objective To develop a simple mathematical model for clinical practice that allows early identification of office employees with a diagnosis of DM2 or at risk of presenting it, based on objective measurements of the SB pattern, hours of sleep and anthropometric variables. Methods Cross-sectional study. Anthropometric variables (sex, age and body mass index, BMI), sleep time (hours) and the SB pattern (sedentary breaks and time spent in sedentary bouts with four different lengths) of two groups of office employees (adults with and without diabetes) were measured and compared using the ActivPAL3M device. Eighty-one participants had DM2 and 132 had normal glucose metabolism (NGM). The risk of having DM2 was modelled using a generalised linear model (GLM), selecting the variables that presented a significant correlation with DM2. Results The DM2 prediction model used five non-invasive clinical variables -sex, age, BMI, sleep time (hours) and sedentary breaks < 20 minutes (number/day) – related to the SB pattern. The validated model correctly classified 88.89% of the participants. The model correctly classified all the office employees with NGM and 77% of office employees with DM2. It also allowed, based on the anthropometric profile of the participant, the design of a preventive tool to modify the SB pattern of office employees with DM2. Conclusion Understanding SB patterns by means of mathematical models could be a simple application solution for the early identification of office employees with DM2 in clinical practice. Incorporating an algorithm that contains a mathematical expression in wearable devices for monitoring the SB pattern could promote the early detection and comprehensive control of DM2 in clinical practice.