Risk models and scores for type 2 diabetes: systematic review

Hepatitis C risk score as a tool to identify individuals with HCV infection: a demonstration and cross-sectional epidemiological study in Egypt

OBJECTIVES

Hepatitis C virus (HCV) infection poses a global health challenge. By the end of 2021, the WHO estimated that less than a quarter of global HCV infections had been diagnosed. There is a need for a public health tool that can facilitate the identification of people with HCV infection and link them to testing and treatment, and that can be customised for each country.

METHODS

We derived and validated a risk score to identify people with HCV in Egypt and demonstrated its utility. Using data from the 2008 and 2014 Egypt Demographic and Health Surveys, two risk scores were constructed through multivariable logistic regression analysis. A range of diagnostic metrics was then calculated to evaluate the performance of these scores.

RESULTS

The 2008 and 2014 risk scores exhibited similar dependencies on sex, age and type of place of residence. Both risk scores demonstrated high and similar areas under the curve of 0.77 (95% CI: 0.76 to 0.78) and 0.78 (95% CI: 0.77 to 0.80), respectively. For the 2008 risk score, sensitivity was 73.7% (95% CI: 71.5% to 75.9%), specificity was 68.5% (95% CI: 67.5% to 69.4%), positive predictive value (PPV) was 27.8% (95% CI: 26.4% to 29.2%) and negative predictive value (NPV) was 94.1% (95% CI: 93.5% to 94.6%). For the 2014 risk score, sensitivity was 64.0% (95% CI: 61.5% to 66.6%), specificity was 78.2% (95% CI: 77.5% to 78.9%), PPV was 22.2% (95% CI: 20.9% to 23.5%) and NPV was 95.7% (95% CI: 95.4% to 96.1%). Each score was validated by applying it to a different survey database than the one used to derive it.

CONCLUSIONS

Implementation of HCV risk scores is an effective strategy to identify carriers of HCV infection and to link them to testing and treatment at low cost to national programmes.

A dynamic probabilistic model of the onset and interaction of cardio-metabolic comorbidities on an ageing adult population

Comorbidity is widespread in the ageing population, implying multiple and complex medical needs for individuals and a public health burden. Determining risk factors and predicting comorbidity development can help identify at-risk subjects and design prevention strategies. Using socio-demographic and clinical data from approximately 11,000 subjects monitored over 11 years in the English Longitudinal Study of Ageing, we develop a dynamic Bayesian network (DBN) to model the onset and interaction of three cardio-metabolic comorbidities, namely type 2 diabetes (T2D), hypertension, and heart problems. The DBN allows us to identify risk factors for developing each morbidity, simulate ageing progression over time, and stratify the population based on the risk of outcome occurrence. By applying hierarchical agglomerative clustering to the simulated, dynamic risk of experiencing morbidities, we identified patients with similar risk patterns and the variables contributing to their discrimination. The network reveals a direct joint effect of biomarkers and lifestyle on outcomes over time, such as the impact of fasting glucose, HbA1c, and BMI on T2D development. Mediated cross-relationships between comorbidities also emerge, showcasing the interconnected nature of these health issues. The model presents good calibration and discrimination ability, particularly in predicting the onset of T2D (iAUC-ROC = 0.828, iAUC-PR = 0.294) and survival (iAUC-ROC = 0.827, iAUC-PR = 0.311). Stratification analysis unveils two distinct clusters for all comorbidities, effectively discriminated by variables like HbA1c for T2D and age at baseline for heart problems. The developed DBN constitutes an effective, highly-explainable predictive risk tool for simulating and stratifying the dynamic risk of developing cardio-metabolic comorbidities. Its use could help identify the effects of risk factors and develop health policies that prevent the occurrence of comorbidities.

Prediction of Diabetes and Prediabetes among the Saudi Population Using a Non-Invasive Tool (AUSDRISK)

Background and Objectives: Screening for type 2 diabetes mellitus (DM2) aims to identify asymptomatic individuals who may be at a higher risk, allowing proactive interventions. The objective of this study was to predict the incidence of DM2 and prediabetes in the Saudi population over the next five years. Materials and Methods: The study was conducted in the Aseer region through August 2023 using a cross-sectional survey for data collection. A multistage stratified random sampling technique was adopted, and data were collected through face-to-face interviews using the validated Arabic version of the Australian Type 2 Diabetes Risk Assessment Tool (AUSDRISK). Results: In total, 652 individuals were included in the study. Their mean age was 32.0 ± 12.0 years; 53.8% were male, 89.6% were from urban areas, and 55.8% were single. There were statistically significant differences between males and females in AUSDRISK items, including age, history of high blood glucose, use of medications for high blood pressure, smoking, physical activity, and measurements of waist circumference (p < 0.05). Based on AUSDRISK scores, 46.2% of the included participants were predicted to develop impaired glucose tolerance within the coming five years (65.8% among females vs. 23.6%), and 21.9% were predicted to develop DM2 (35.6% among males vs. 6.0% among females); this difference was statistically significant (p = 0.0001). Conclusions: Urgent public health action is required to prevent the increasing epidemic of DM2 in Saudi Arabia.

Development and validation of a risk score to predict unplanned hospital readmissions in ICU survivors: A data linkage study

BACKGROUND

Intensive Care Unit (ICU) follow-up clinics are growing in popularity internationally; however, there is limited evidence as to which patients would benefit most from a referral to this service.

OBJECTIVES

The objective of this study was to develop and validate a model to predict which ICU survivors are most likely to experience an unplanned hospital readmission or death in the year after hospital discharge and derive a risk score capable of identifying high-risk patients who may benefit from referral to follow-up services.

METHODS

A multicentre, retrospective observational cohort study using linked administrative data from eight ICUs was conducted in the state of New South Wales, Australia. A logistic regression model was developed for the composite outcome of death or unplanned readmission in the 12 months after discharge from the index hospitalisation.

RESULTS

12,862 ICU survivors were included in the study, of which 5940 (46.2%) patients experienced unplanned readmission or death. Strong predictors of readmission or death included the presence of a pre-existing mental health disorder (odds ratio [OR]: 1.52, 95% confidence interval [CI]: 1.40-1.65), severity of critical illness (OR: 1.57, 95% CI: 1.39-1.76), and two or more physical comorbidities (OR: 2.39, 95% CI: 2.14-2.68). The prediction model demonstrated reasonable discrimination (area under the receiver operating characteristic curve: 0.68, 95% CI: 0.67-0.69) and overall performance (scaled Brier score: 0.10). The risk score was capable of stratifying patients into three distinct risk groups-high (64.05% readmitted or died), medium (45.77% readmitted or died), and low (29.30% readmitted or died).

CONCLUSIONS

Unplanned readmission or death is common amongst survivors of critical illness. The risk score presented here allows patients to be stratified by risk level, enabling targeted referral to preventative follow-up services.

Risk assessment of diabetes using the Indian Diabetes Risk Score among older adults: Secondary analysis from the Longitudinal Ageing Study in India

BACKGROUND

The Indian Diabetes Risk Score (IDRS) is a simple tool to assess the probability of an individual having type 2 diabetes (T2DM) but its applicability in community-dwelling older adults is lacking. This study aimed to estimate the risk of T2DM and its determinants among older adults without prior diabetes (DM) using the IDRS, while also assessing its sensitivity and specificity in individuals with a history of diabetes.

METHODS

We analyzed cross-sectional data from the Longitudinal Ageing Study in India (LASI) wave-1 (2017-18). IDRS was calculated amongst individuals aged ≥45 years considering waist circumference, physical activity, age and family history of DM. Risk was categorized as high (≥60), moderate (30-50), and low (<30).

RESULTS

Among 64541 individuals, 7.27 % (95 % CI: 6.78, 7.80) were at low risk, 61.80 % (95 % CI: 60.99, 62.61) at moderate risk, and 30.93 % (95 % CI: 30.19, 31.67) at high risk for T2DM. Adjusted analysis showed higher risk of T2DM among men, widowed/divorced, urban residents, minority religions, overweight, obese, and individuals with hypertension. ROC curve yielded an AUC of 0.67 (95 % CI: 0.66, 0.67, P < 0.001). The IDRS cutoff ≥50 had 73.69 % sensitivity and 51.40 % specificity for T2DM detection.

CONCLUSION

More than 9 in 10 older adults in India without history of DM have high-moderate risk of T2DM when assessed with the IDRS risk-prediction tool. However, the low specificity and moderate sensitivity of IDRS in existing DM cases constraints its practical utility as a decision tool for screening.

Predictors of glycemic worsening in the next year in adults with screen-detected type 2 diabetes

Background and Aims

Identifying simple markers of risk for worsening glucose can allow care providers to target therapeutic interventions according to risk of worsening glycemic control. We aimed to determine which routine clinical measures herald near-term glycemic worsening in early type 2 diabetes(T2D).

Methods

The Early Diabetes Intervention Program (EDIP) was a clinical trial in individuals with screendetected T2D [HbA1C 6.3+0.63%(45+5mmol/mol)]. During the trial some participants experienced worsening fasting blood glucose (FBG). We investigated the time course of FBG, HbA1c, weight, and other clinical factors to determine which might herald glycemic worsening over the next year.

Results

Progressors (62/219, 28.5%) had higher FBG than non-progressors at baseline [118 vs 130mg/dL (6.6 vs 7.2 mmol/L), p=<0.001]. FBG was stable except in the year of progression, when progressors exhibited a large 1-year rise [mean change 14.2mg/dL(0.79 mmol/L)]. Current FBG and antecedent year change in FBG were associated with progression(p<0.01), although the magnitude of change was too small to be of clinical utility (0.19 mg/dL; 0.01 mmol/L). Current or antecedent year change in HbA1c, weight, TG or HDL were not associated with progression. In the year of glycemic worsening, rising glucose was strongly associated with a concurrent increase in weight (p<0.001).

Conclusions

Elevated FBG but not HbA1c identified individuals at risk for imminent glycemic worsening; the subsequent large rise in glucose was associated with a short-term increase in weight. Glucose and weight surveillance provide actionable information for those caring for patients with early diabetes.

Development and validation of a machine learning model for prediction of type 2 diabetes in patients with mental illness

BACKGROUND

Type 2 diabetes (T2D) is approximately twice as common among individuals with mental illness compared with the background population, but may be prevented by early intervention on lifestyle, diet, or pharmacologically. Such prevention relies on identification of those at elevated risk (prediction). The aim of this study was to develop and validate a machine learning model for prediction of T2D among patients with mental illness.

METHODS

The study was based on routine clinical data from electronic health records from the psychiatric services of the Central Denmark Region. A total of 74,880 patients with 1.59 million psychiatric service contacts were included in the analyses. We created 1343 potential predictors from 51 source variables, covering patient-level information on demographics, diagnoses, pharmacological treatment, and laboratory results. T2D was operationalised as HbA1c ≥48 mmol/mol, fasting plasma glucose ≥7.0 mmol/mol, oral glucose tolerance test ≥11.1 mmol/mol or random plasma glucose ≥11.1 mmol/mol. Two machine learning models (XGBoost and regularised logistic regression) were trained to predict T2D based on 85% of the included contacts. The predictive performance of the best performing model was tested on the remaining 15% of the contacts.

RESULTS

The XGBoost model detected patients at high risk 2.7 years before T2D, achieving an area under the receiver operating characteristic curve of 0.84. Of the 996 patients developing T2D in the test set, the model issued at least one positive prediction for 305 (31%).

CONCLUSION

A machine learning model can accurately predict development of T2D among patients with mental illness based on routine clinical data from electronic health records. A decision support system based on such a model may inform measures to prevent development of T2D in this high-risk population.

Application of deep neural survival networks to the development of risk prediction models for diabetes mellitus, hypertension, and dyslipidemia

OBJECTIVES

: Although numerous risk prediction models have been proposed, few such models have been developed using neural network-based survival analysis. We developed risk prediction models for three cardiovascular disease risk factors (diabetes mellitus, hypertension, and dyslipidemia) among a working-age population in Japan using DeepSurv, a deep feed-forward neural network.

METHODS

: Data were obtained from the Japan Epidemiology Collaboration on Occupational Health Study. A total of 51 258, 44 197, and 31 452 individuals were included in the development of risk models for diabetes mellitus, hypertension, and dyslipidemia, respectively; two-thirds of whom were used to develop prediction models, and the rest were used to validate the models. We compared the performances of DeepSurv-based models with those of prediction models based on the Cox proportional hazards model.

RESULTS

: The area under the receiver-operating characteristic curve was 0.878 [95% confidence interval (CI) = 0.864-0.892] for diabetes mellitus, 0.835 (95% CI = 0.826-0.845) for hypertension, and 0.826 (95% CI = 0.817-0.835) for dyslipidemia. Compared with the Cox proportional hazards-based models, the DeepSurv-based models had better reclassification performance [diabetes mellitus: net reclassification improvement (NRI) = 0.474, P  ≤ 0.001; hypertension: NRI = 0.194, P  ≤ 0.001; dyslipidemia: NRI = 0.397, P  ≤ 0.001] and discrimination performance [diabetes mellitus: integrated discrimination improvement (IDI) = 0.013, P  ≤ 0.001; hypertension: IDI = 0.007, P  ≤ 0.001; and dyslipidemia: IDI = 0.043, P  ≤ 0.001].

CONCLUSION

: This study suggests that DeepSurv has the potential to improve the performance of risk prediction models for cardiovascular disease risk factors.

Perceptions of diabetes risk and prevention in Nairobi, Kenya: A qualitative and theory of change development study

BACKGROUND

Type 2 diabetes is increasing in Kenya, especially in urban settings, and prevention interventions based on local evidence and context are urgently needed. Therefore, this study aimed to explore diabetes risk and co-create a diabetes prevention theory of change in two socioeconomically distinct communities to inform future diabetes prevention interventions.

METHODS

In-depth interviews were conducted with middle-aged residents in two communities in Nairobi (one low-income (n = 15), one middle-income (n = 14)), and thematically analysed. The theory of change for diabetes prevention was informed by analysis of the in-depth interviews and the Behaviour Change Wheel framework, and reviewed by a sub-set (n = 13) of interviewees.

RESULTS

The key factors that influenced diabetes preventive practices in both communities included knowledge and skills for diabetes prevention, understanding of the benefits/consequences of (un)healthy lifestyle, social influences (e.g., upbringing, societal perceptions), and environmental contexts (e.g., access to (un)healthy foods and physical activity facilities). The proposed strategies for diabetes prevention included: increasing knowledge and understanding about diabetes risk and preventive measures particularly in the low-income community; supporting lifestyle modification (e.g., upskilling, goal setting, action planning) in both communities; identifying people at high risk of diabetes through screening in both communities; and creating social and physical environments for lifestyle modification (e.g., positive social influences on healthy living, access to healthy foods and physical activity infrastructure) particularly in the low-income community. Residents from both communities agreed that the strategies were broadly feasible for diabetes prevention but proposed the addition of door-to-door campaigns and community theatre for health education. However, residents from the low-income community were concerned about the lack of government prioritisation for implementing population-level interventions, e.g., improving access to healthy foods and physical activity facilities/infrastructure.

CONCLUSION

Diabetes prevention initiatives in Kenya should involve multicomponent interventions for lifestyle modification including increasing education and upskilling at individual level; promoting social and physical environments that support healthy living at population level; and are particularly needed in low-income communities.

Association between cardiovascular disease risk and incident type 2 diabetes mellitus in individuals with prediabetes: A retrospective cohort study

AIMS

In this study, we aimed to evaluate the impact of overall cardiovascular disease (CVD) risk on the development of incident T2DM in patients with prediabetes.

METHODS

We retrospectively enrolled 5,908 subjects with prediabetes who underwent health check-ups at the Asan Medical Center. CVD risk was estimated using the Framingham Risk Score (FRS). We compared moderate- to high-risk groups with low-risk controls based on the FRS. Cox proportional hazards regressions were conducted to estimate the time-to-develop incident T2DM.

RESULTS

Among the 5908 subjects with prediabetes, 3031 (51.8 %) were identified to have either moderate or high CVD risk scores. During a median follow-up of 5.2 years, 278 (9.2 %) patients from the moderate- to high-risk group and 171 (5.9 %) from the low-risk group were diagnosed with T2DM. The covariate-adjusted hazard ratio for the incident T2DM was 1.30 (95 % CI, 1.06-1.60, p = 0.011) in the moderate- to high-risk group compared to the low-risk controls.

CONCLUSION

Among patients with prediabetes, those with high CVD risk were more likely to develop incident T2DM, as determined by the FRS. CVD risk factors should be properly evaluated and managed in individuals with prediabetes to reduce the risk of both incident T2DM and associated cardiovascular complications.

Mean age and body mass index at type 2 diabetes diagnosis: Pooled analysis of 56 health surveys across income groups and world regions

BACKGROUND

Screening for type 2 diabetes mellitus (T2DM) targets people aged 35+ years and those with overweight/obesity. With mounting evidence on young-onset T2DM and T2DM patients with lean phenotypes, it is worth revising the screening criteria to include younger and leaner adults. We quantified the mean age and body mass index (BMI; kg/m2 ) at T2DM diagnosis in 56 countries.

METHODS

Descriptive cross-sectional analysis of WHO STEPS surveys. We analysed adults (25-69 years) with new T2DM diagnosis (not necessarily T2DM onset) as per fasting plasma glucose ≥126 mg/dL measured during the survey. For people with new T2DM diagnosis, we summarized the mean age and the proportion of each five-year age group; also, we summarized the mean BMI and the proportion of mutually exclusive BMI categories.

RESULTS

There were 8695 new T2DM patients. Overall, the mean age at T2DM diagnosis was 45.1 years in men and 45.0 years in women; and the mean BMI at T2DM diagnosis was 25.2 in men and 26.9 in women. Overall, in men, 10.3% were 25-29 years and 8.5% were 30-34 years old; in women, 8.6% and 12.5% were 25-29 years and 30-34 years old, respectively. 48.5% of men and 37.3% of women were in the normal BMI category.

CONCLUSIONS

A non-negligible proportion of new T2DM patients were younger than 35 years. Many new T2DM patients were in the normal weight range. Guidelines for T2DM screening may consider revising the age and BMI criteria to incorporate young and lean adults.

Validation of prevalent diabetes risk scores based on non-invasively measured predictors in Ghanaian migrant and non-migrant populations - The RODAM study

Background

Non-invasive diabetes risk models are a cost-effective tool in large-scale population screening to identify those who need confirmation tests, especially in resource-limited settings.

Aims

This study aimed to evaluate the ability of six non-invasive risk models (Cambridge, FINDRISC, Kuwaiti, Omani, Rotterdam, and SUNSET model) to identify screen-detected diabetes (defined by HbA1c) among Ghanaian migrants and non-migrants.

Study design

A multicentered cross-sectional study.

Methods

This analysis included 4843 Ghanaian migrants and non-migrants from the Research on Obesity and Diabetes among African Migrants (RODAM) Study. Model performance was assessed using the area under the receiver operating characteristic curves (AUC), Hosmer-Lemeshow statistics, and calibration plots.

Results

All six models had acceptable discrimination (0.70 ≤ AUC <0.80) for screen-detected diabetes in the overall/combined population. Model performance did not significantly differ except for the Cambridge model, which outperformed Rotterdam and Omani models. Calibration was poor, with a consistent trend toward risk overestimation for screen-detected diabetes, but this was substantially attenuated by recalibration through adjustment of the original model intercept.

Conclusion

Though acceptable discrimination was observed, the original models were poorly calibrated among populations of African ancestry. Recalibration of these models among populations of African ancestry is needed before use.

Implementation status of prediction models for type 2 diabetes

Although a prediction model is expected to identify individuals who are at a high risk of type 2 diabetes, the implementation status of prediction models has not been well examined. Our review indicates that the implementation of predictive models in practice remains low despite the increase in models being developed.

Family Involvement to Stop the Conversion of Prediabetes to Diabetes

Prediabetes is a condition associated with an increased risk of developing diabetes, in which blood glucose levels are high but not high enough to be diagnosed as diabetes. The rapid increase in the prevalence of prediabetes is a major global health challenge. The incidence of prediabetes has increased to pandemic levels and can lead to serious consequences. Unfortunately, nearly 90% of prediabetic individuals are unaware of their ailment. A quarter of prediabetic individuals develop type 2 diabetes mellitus (T2DM) within 3-5 years. Although prediabetes is a reversible condition, the prevention of diabetes has received little attention. It is essential for prediabetic individuals to implement new health-improvement techniques. Focusing on family systems is one strategy to promote health, which is determined by health patterns that are often taught, established, and adjusted within family contexts. For disease prevention, a family-based approach may be beneficial. Family support is essential for the metabolic control of the disease. This study aimed to show several strategies for involving the patient's family members in preventing the conversion of prediabetes to T2DM and to emphasize that the patient's family members are a valuable resource to reduce the incidence of diabetes.

A scoping review of artificial intelligence-based methods for diabetes risk prediction

The increasing prevalence of type 2 diabetes mellitus (T2DM) and its associated health complications highlight the need to develop predictive models for early diagnosis and intervention. While many artificial intelligence (AI) models for T2DM risk prediction have emerged, a comprehensive review of their advancements and challenges is currently lacking. This scoping review maps out the existing literature on AI-based models for T2DM prediction, adhering to the PRISMA extension for Scoping Reviews guidelines. A systematic search of longitudinal studies was conducted across four databases, including PubMed, Scopus, IEEE-Xplore, and Google Scholar. Forty studies that met our inclusion criteria were reviewed. Classical machine learning (ML) models dominated these studies, with electronic health records (EHR) being the predominant data modality, followed by multi-omics, while medical imaging was the least utilized. Most studies employed unimodal AI models, with only ten adopting multimodal approaches. Both unimodal and multimodal models showed promising results, with the latter being superior. Almost all studies performed internal validation, but only five conducted external validation. Most studies utilized the area under the curve (AUC) for discrimination measures. Notably, only five studies provided insights into the calibration of their models. Half of the studies used interpretability methods to identify key risk predictors revealed by their models. Although a minority highlighted novel risk predictors, the majority reported commonly known ones. Our review provides valuable insights into the current state and limitations of AI-based models for T2DM prediction and highlights the challenges associated with their development and clinical integration.

Study protocol for development and validation of a single tool to assess risks of stroke, diabetes mellitus, myocardial infarction and dementia: DemNCD-Risk

INTRODUCTION

Current efforts to reduce dementia focus on prevention and risk reduction by targeting modifiable risk factors. As dementia and cardiometabolic non-communicable diseases (NCDs) share risk factors, a single risk-estimating tool for dementia and multiple NCDs could be cost-effective and facilitate concurrent assessments as compared with a conventional single approach. The aim of this study is to develop and validate a new risk tool that estimates an individual's risk of developing dementia and other NCDs including diabetes mellitus, stroke and myocardial infarction. Once validated, it could be used by the public and general practitioners.

METHODS AND ANALYSIS

Ten high-quality cohort studies from multiple countries were identified, which met eligibility criteria, including large representative samples, long-term follow-up, data on clinical diagnoses of dementia and NCDs, recognised modifiable risk factors for the four NCDs and mortality data. Pooled harmonised data from the cohorts will be used, with 65% randomly allocated for development of the predictive model and 35% for testing. Predictors include sociodemographic characteristics, general health risk factors and lifestyle/behavioural risk factors. A subdistribution hazard model will assess the risk factors' contribution to the outcome, adjusting for competing mortality risks. Point-based scoring algorithms will be built using predictor weights, internally validated and the discriminative ability and calibration of the model will be assessed for the outcomes. Sensitivity analyses will include recalculating risk scores using logistic regression.

ETHICS AND DISSEMINATION

Ethics approval is provided by the University of New South Wales Human Research Ethics Committee (UNSW HREC; protocol numbers HC200515, HC3413). All data are deidentified and securely stored on servers at Neuroscience Research Australia. Study findings will be presented at conferences and published in peer-reviewed journals. The tool will be accessible as a public health resource. Knowledge translation and implementation work will explore strategies to apply the tool in clinical practice.

Associations of updated cardiovascular health metrics, including sleep health, with incident diabetes and cardiovascular events in older adults with prediabetes: A nationwide population-based cohort study

AIMS

To investigate the association of updated cardiovascular health (CVH) metrics, including sleep health, with the risk of diabetes and major adverse cardiovascular events (MACE) in older adults with prediabetes.

METHODS

A total of 7,948 older adults with prediabetes aged ≥ 65 years were included in this study. CVH was assessed using seven baseline metrics according to the modified American Heart Association recommendations.

RESULTS

During a median follow-up time of 11.9 years, 2,405 (30.3%) cases of diabetes and 2,039 (25.6%) MACE were recorded. Compared with the poor composite CVH metrics group, the multivariable-adjusted hazard ratios (HRs) in the intermediate and ideal composite CVH metrics groups were respectively 0.87 (95% confidence intervals [CI] = 0.78-0.96) and 0.72 (95% CI = 0.65-0.79) for diabetes events and 0.99 (95% CI = 0.88-1.11) and 0.88 (95% CI = 0.79-0.97) for MACE. The ideal composite CVH metrics group had a lower risk of diabetes and MACE in older adults aged 65-74 years, but not in those aged ≥ 75 years.

CONCLUSIONS

Ideal composite CVH metrics in older adults with prediabetes were associated with a lower risk of diabetes and MACE.

Risk prediction models for type 2 diabetes using either fasting plasma glucose or HbA1c in Chinese, Malay, and Indians: Results from three multi-ethnic Singapore cohorts

AIMS

To assess three well-established type 2 diabetes (T2D) risk prediction models based on fasting plasma glucose (FPG) in Chinese, Malays, and Indians, and to develop simplified risk models based on either FPG or HbA1c.

METHODS

We used a prospective multiethnic Singapore cohort to evaluate the established models and develop simplified models. 6,217 participants without T2D at baseline were included, with an average follow-up duration of 8.3 years. The simplified risk models were validated in two independent multiethnic Singapore cohorts (N = 12,720).

RESULTS

The established risk models had moderate-to-good discrimination (area under the receiver operating characteristic curves, AUCs 0.762 - 0.828) but a lack of fit (P-values < 0.05). Simplified risk models that included fewer predictors (age, BMI, systolic blood pressure, triglycerides, and HbA1c or FPG) showed good discrimination in all cohorts (AUCs ≥ 0.810), and sufficiently captured differences between the ethnic groups. While recalibration improved fit the simplified models in validation cohorts, there remained evidence of miscalibration in Chinese (p ≤ 0.012).

CONCLUSIONS

Simplified risk models including HbA1c or FPG had good discrimination in predicting incidence of T2D in three major Asian ethnic groups. Risk functions with HbA1c performed as well as those with FPG.

Defeating diabetes in the desert: A community-based mHealth diabetes screening intervention in Jodhpur Rajasthan

Background

There is a paucity of demonstrated models for mHealth-based diabetes screening and coordinated care in India, especially in western Rajasthan, which is the part of Thar desert.

Materials and Methods

JSPH collaboratively developed and implemented an easy-to-use, noninvasive, mobile phone-based screening interview, to identify adults at high risk for diabetes. The high risk for diabetes was defined using multiple clinical and epidemiologic criteria, all based on the evidence for India and globally. Since participants above 35 years or older were only considered in the screening, the application was designed to categorize the participants as high and low risk.

Results

Out of 4000 screened participants, the percentage of males and females were 51% and 50%, respectively. Participants found to be at high risk and low risk were n = 3600 (90%) and 400 (10%). The mean age of high- and low-risk participants was 52.2 (+12.8) and 36.2 (+4.2), respectively. Of the 3600 high-risk individuals who have been given a follow-up interview, 90.50% of high-risk individuals obtained diabetes testing, and of these, 65.67% had a written report showing they test positive for diabetes or prediabetes, requiring ongoing clinical care.

Conclusions

JSPH mHealth application provided a novel noninvasive way to better identify those at high diabetes risk in the community and demonstrated how to optimize the use of mobile health methods in diabetes prevention and care services.

Machine learning for predicting diabetes risk in western China adults

OBJECTIVE

Diabetes mellitus is a global epidemic disease. Long-time exposure of patients to hyperglycemia can lead to various type of chronic tissue damage. Early diagnosis of and screening for diabetes are crucial to population health.

METHODS

We collected the national physical examination data in Xinjiang, China, in 2020 (a total of more than 4 million people). Three types of physical examination indices were analyzed: questionnaire, routine physical examination and laboratory values. Integrated learning, deep learning and logistic regression methods were used to establish a risk model for type-2 diabetes mellitus. In addition, to improve the convenience and flexibility of the model, a diabetes risk score card was established based on logistic regression to assess the risk of the population.

RESULTS

An XGBoost-based risk prediction model outperformed the other five risk assessment algorithms. The AUC of the model was 0.9122. Based on the feature importance ranking map, we found that hypertension, fasting blood glucose, age, coronary heart disease, ethnicity, parental diabetes mellitus, triglycerides, waist circumference, total cholesterol, and body mass index were the most important features of the risk prediction model for type-2 diabetes.

CONCLUSIONS

This study established a diabetes risk assessment model based on multiple ethnicities, a large sample and many indices, and classified the diabetes risk of the population, thus providing a new forecast tool for the screening of patients and providing information on diabetes prevention for healthy populations.

Data collaboration analysis in predicting diabetes from a small amount of health checkup data

Recent studies showed that machine learning models such as gradient-boosting decision tree (GBDT) can predict diabetes with high accuracy from big data. In this study, we asked whether highly accurate prediction of diabetes is possible even from small data by expanding the amount of data through data collaboration (DC) analysis, a modern framework for integrating and analyzing data accumulated at multiple institutions while ensuring confidentiality. To this end, we focused on data from two institutions: health checkup data of 1502 citizens accumulated in Tsukuba City and health history data of 1399 patients collected at the University of Tsukuba Hospital. When using only the health checkup data, the ROC-AUC and Recall for logistic regression (LR) were 0.858 ± 0.014 and 0.970 ± 0.019, respectively, while those for GBDT were 0.856 ± 0.014 and 0.983 ± 0.016, respectively. When using also the health history data through DC analysis, these values for LR improved to 0.875 ± 0.013 and 0.993 ± 0.009, respectively, while those for GBDT deteriorated because of the low compatibility with a method used for confidential data sharing (although DC analysis brought improvements). Even in a situation where health checkup data of only 324 citizens are available, the ROC-AUC and Recall for LR were 0.767 ± 0.025 and 0.867 ± 0.04, respectively, thanks to DC analysis, indicating an 11% and 12% improvement. Thus, we concluded that the answer to the above question was "Yes" for LR but "No" for GBDT for the data set tested in this study.

Optimizing strategies to identify high risk of developing type 2 diabetes

Introduction

The success of diabetes prevention based on early treatment depends on high-quality screening. This study compared the diagnostic properties of currently recommended screening strategies against alternative score-based rules to identify those at high risk of developing diabetes.

Methods

The study used data from ELSA-Brasil, a contemporary cohort followed up for a mean (standard deviation) of 7.4 (0.54) years, to develop risk functions with logistic regression to predict incident diabetes based on socioeconomic, lifestyle, clinical, and laboratory variables. We compared the predictive capacity of these functions against traditional pre-diabetes cutoffs of fasting plasma glucose (FPG), 2-h plasma glucose (2hPG), and glycated hemoglobin (HbA1c) alone or combined with recommended screening questionnaires.

Results

Presenting FPG > 100 mg/dl predicted 76.6% of future cases of diabetes in the cohort at the cost of labeling 40.6% of the sample as high risk. If FPG testing was performed only in those with a positive American Diabetes Association (ADA) questionnaire, labeling was reduced to 12.2%, but only 33% of future cases were identified. Scores using continuously expressed clinical and laboratory variables produced a better balance between detecting more cases and labeling fewer false positives. They consistently outperformed strategies based on categorical cutoffs. For example, a score composed of both clinical and laboratory data, calibrated to detect a risk of future diabetes ≥20%, predicted 54% of future diabetes cases, labeled only 15.3% as high risk, and, compared to the FPG ≥ 100 mg/dl strategy, nearly doubled the probability of future diabetes among screen positives.

Discussion

Currently recommended screening strategies are inferior to alternatives based on continuous clinical and laboratory variables.

Large scale application of the Finnish diabetes risk score in Latin American and Caribbean populations: a descriptive study

Background

The prevalence of type 2 diabetes (T2D) continues to increase in the Americas. Identifying people at risk for T2D is critical to the prevention of T2D complications, especially cardiovascular disease. This study gauges the ability to implement large population-based organized screening campaigns in 19 Latin American and Caribbean countries to detect people at risk for T2D using the Finnish Diabetes Risk Score (FINDRISC).

Methods

This cross-sectional descriptive analysis uses data collected in a sample of men and women 18 years of age or older who completed FINDRISC via eHealth during a Guinness World Record attempt campaign between October 25 and November 1, 2021. FINDRISC is a non-invasive screening tool based on age, body mass index, waist circumference, physical activity, daily intake of fruits and vegetables, history of hyperglycemia, history of antihypertensive drug treatment, and family history of T2D, assigning a score ranging from 0 to 26 points. A cut-off point of ≥ 12 points was considered as high risk for T2D.

Results

The final sample size consisted of 29,662 women (63%) and 17,605 men (27%). In total, 35% of subjects were at risk of T2D. The highest frequency rates (FINDRISC ≥ 12) were observed in Chile (39%), Central America (36.4%), and Peru (36.1%). Chile also had the highest proportion of people having a FINDRISC ≥15 points (25%), whereas the lowest was observed in Colombia (11.3%).

Conclusions

FINDRISC can be easily implemented via eHealth technology over social networks in Latin American and Caribbean populations to detect people with high risk for T2D. Primary healthcare strategies are needed to perform T2D organized screening to deliver early, accessible, culturally sensitive, and sustainable interventions to prevent sequelae of T2D, and reduce the clinical and economic burden of cardiometabolic-based chronic disease.

Measurement of Perceived Risk of Developing Diabetes Mellitus: A Systematic Literature Review

BACKGROUND

This systematic review describes approaches to measuring perceived risk of developing type 2 diabetes among individuals without diagnoses and describes the use of theories, models, and frameworks in studies assessing perceived risk. While a systematic review has synthesized perceived risk of complications among individuals with diabetes, no reviews have systematically assessed how perceived risk is measured among those without a diagnosis.

METHODS

Medline, PubMed, PsycINFO, and CINAHAL databases were searched for studies conducted through October 2022 with measures of perceived risk among adults ≥ 18 years without a diabetes diagnosis. Extracted data included study characteristics, measures, and health behavior theories, models, or frameworks used.

RESULTS

Eighty-six studies met inclusion criteria. Six examined perceived risk scales' psychometric properties. Eighty measured perceived risk using (1) a single item; (2) a composite score from multiple items or subconstructs; and (3) multiple subconstructs but no composite score. Studies used items measuring "comparative risk," "absolute or lifetime risk," and "perceived risk" without defining how each differed. Sixty-four studies used cross-sectional designs. Twenty-eight studies mentioned use of health behavior theories in study design or selection of measures.

DISCUSSION

There was heterogeneity in how studies operationalized perceived risk; only one third of studies referenced a theory, model, or framework as guiding design or scale and item selection. Use of perceived lifetime risk, absolute risk, or comparative risk limits comparisons across studies. Consideration of context, target population, and how data are utilized is important when selecting measures; we present a series of questions to ask when selecting measures for use in research and clinical settings. This review is the first to categorize how perceived risk is measured in the diabetes prevention domain; most literature focuses on perceived risk among those with diabetes diagnoses. Limitations include exclusion of non-English and gray literature and single reviewer screening and data extraction.

Comparison of five different risk scores to predict incident type 2 diabetes in the Swiss HIV cohort study

OBJECTIVE

People with HIV (PWH) have a higher risk of type 2 diabetes (T2D) than HIV-negative individuals. In the general population, diabetes risk scores are used to identify persons at risk of developing T2D, but little is known regarding their performance in PWH.

DESIGN

Assessment of the capacity of five diabetes risk scores to predict T2D in PWH.

METHODS

A prospective study including all Swiss HIV cohort study (SHCS) participants followed between 2009 and 2019. Five diabetes risk scores were assessed: FINDRISC versions 1 and 2, Balkau, Swiss Diabetes Association (SDA), and Kraege.

RESULTS

Three thousand eight hundred fifty-three T2D-free PWH (78.5% men, 39.9 ± 11.3 years) were included. After a median follow-up of 4.8 years (interquartile range 2.2-7.8), 62 participants (1.6%) developed T2D, corresponding to an incidence rate of 3.18 per 1000 person-years (95% confidence interval = 2.47-4.08). Participants who developed T2D were older (48.7 ± 12.4 vs. 39.8 ± 11.2 years), more likely to be obese (22.6% vs. 7.4%), abdominally obese (9.7% vs. 1.5%), and to have a family history of diabetes (32.3% vs. 19.1%) than those without T2D. The AUC for incident T2D ranged between 0.72 (Kraege 16) and 0.81 (SDA, FINDRISC2 and Balkau). Sensitivity ranged between 3.2% (Balkau) and 67.7% (FINDRISC1) and specificity between 80.9% (FINDRISC1) and 98.3% (Balkau). Positive predictive values of all scores were below 20%, while negative predictive values were above 98%.

CONCLUSION

Our study shows that the performance of conventional diabetes risk scores in PWH is promising, especially for Balkau and FINDRISC2, which showed good discriminatory power. These scores may help identify patients at a low risk of T2D in whom careful assessment of modifiable T2D risk factors can be spared.

Polygenic Risk of Prediabetes, Undiagnosed Diabetes, and Incident Type 2 Diabetes Stratified by Diabetes Risk Factors

Context

Early diagnosis of type 2 diabetes is crucial to reduce severe comorbidities and complications. Current screening recommendations for type 2 diabetes include traditional risk factors, primarily body mass index (BMI) and family history, however genetics also plays a key role in type 2 diabetes risk. It is important to understand whether genetic predisposition to type 2 diabetes modifies the effect of these traditional factors on type 2 diabetes risk.

Objective

This work aimed to investigate whether genetic risk of type 2 diabetes modifies associations between BMI and first-degree family history of diabetes with 1) prevalent prediabetes or undiagnosed diabetes; and 2) incident confirmed type 2 diabetes.

Methods

We included 431 658 individuals aged 40 to 69 years at baseline of multiethnic ancestry from the UK Biobank. We used a multiethnic polygenic risk score for type 2 diabetes (PRS_T2D) developed by Genomics PLC. Prediabetes or undiagnosed diabetes was defined as baseline glycated hemoglobin greater than or equal to 42 mmol/mol (6.0%), and incident type 2 diabetes was derived from medical records.

Results

At baseline, 43 472 participants had prediabetes or undiagnosed diabetes, and 17 259 developed type 2 diabetes over 15 years follow-up. Dose-response associations were observed for PRS_T2D with each outcome in each category of BMI or first-degree family history of diabetes. Those in the highest quintile of PRS_T2D with a normal BMI were at a similar risk as those in the middle quintile who were overweight. Participants who were in the highest quintile of PRS_T2D and did not have a first-degree family history of diabetes were at a similar risk as those with a family history who were in the middle category of PRS_T2D.

Conclusion

Genetic risk of type 2 diabetes remains strongly associated with risk of prediabetes, undiagnosed diabetes, and future type 2 diabetes within categories of nongenetic risk factors. This could have important implications for identifying individuals at risk of type 2 diabetes for prevention and early diagnosis programs.

The role of cytokines and T-bet, GATA3, ROR-γt, and FOXP3 transcription factors of T cell subsets in the natural clinical progression of Type 1 Diabetes

Th cells play an important role in pathogenesis of type 1 diabetes (T1D). Peripheral blood mononuclear cells were isolated from peripheral blood samples from newly diagnosed (ND), 1-year (1YD), and 5-year T1D (5YD) patients (n:8 of each group), 8 healthy controls (HC), and cultured for 24 h under unstimulated (US) and stimulated conditions. Cell ratios of Th1, Th2, Th17, Treg, and intracellular levels of IFN-γ, TNF-α, IL-10, TGF-β, IL-5, IL-13, IL-17, and IL-21 cytokines were evaluated using the flow cytometry. mRNA expressions of transcription factors T-bet, GATA3, ROR-γt, and FOXP3 of these cells were determined by real-time PCR. Reduced CD4⁺CD25^high cell ratios were detected in ND. CD4⁺CD25^high cells were found to be reduced in ND and 1YD compared to HC under IL-2-stimulated conditions. Intracellular IFN-γ and TNF-α levels were low in all patients under US and IL-12-stimulated conditions. IL-17A and IL-21 were found to be high in patients with IL-6-stimulated conditions. Expressions of IL-10 and TGF-β have been observed to be reduced in patients. Th1/Th2, Th17/Treg, and Th1/Treg ratios were higher in patient groups. FOXP3 and GATA3 mRNA expressions were found to be low in patients, while RORγt and T-bet mRNA levels were higher than HC. Th1, Th17, and Treg cells and their cytokines have been shown to be associated with type 1 diabetes.

Assessing racial bias in type 2 diabetes risk prediction algorithms

Risk prediction models for type 2 diabetes can be useful for the early detection of individuals at high risk. However, models may also bias clinical decision-making processes, for instance by differential risk miscalibration across racial groups. We investigated whether the Prediabetes Risk Test (PRT) issued by the National Diabetes Prevention Program, and two prognostic models, the Framingham Offspring Risk Score, and the ARIC Model, demonstrate racial bias between non-Hispanic Whites and non-Hispanic Blacks. We used National Health and Nutrition Examination Survey (NHANES) data, sampled in six independent two-year batches between 1999 and 2010. A total of 9,987 adults without a prior diagnosis of diabetes and with fasting blood samples available were included. We calculated race- and year-specific average predicted risks of type 2 diabetes according to the risk models. We compared the predicted risks with observed ones extracted from the US Diabetes Surveillance System across racial groups (summary calibration). All investigated models were found to be miscalibrated with regard to race, consistently across the survey years. The Framingham Offspring Risk Score overestimated type 2 diabetes risk for non-Hispanic Whites and underestimated risk for non-Hispanic Blacks. The PRT and the ARIC models overestimated risk for both races, but more so for non-Hispanic Whites. These landmark models overestimated the risk of type 2 diabetes for non-Hispanic Whites more severely than for non-Hispanic Blacks. This may result in a larger proportion of non-Hispanic Whites being prioritized for preventive interventions, but it also increases the risk of overdiagnosis and overtreatment in this group. On the other hand, a larger proportion of non-Hispanic Blacks may be potentially underprioritized and undertreated.

Utilizing fog computing and explainable deep learning techniques for gestational diabetes prediction

Gestational diabetes mellitus (GDM) is one of the pregnancy complications that poses a significant risk on mothers and babies as well. GDM usually diagnosed at 22–26 of gestation. However, the early prediction is desirable as it may contribute to decrease the risk. The continuous monitoring for mother’s vital signs helps in predicting any deterioration during pregnancy. The originality of this paper is to provide comprehensive framework for pregnancy women monitoring. The proposed Data Replacement and Prediction Framework consists of three layers which are: (i) IoT Layer, (ii) Fog Layer, and (iii) Cloud Layer. The first layer used IOT sensors to aggregate vital sings from pregnancies using invasive and noninvasive sensors. Then the vital signs transmitted to fog nodes to processed and finally stored in the cloud layer. The main contribution in this paper is located in the fog layer producing GDM module to implement two influential tasks which are: (i) Data Finding Methodology (DFM), and (ii) Explainable Prediction Algorithm (EPM) using DNN. First, the DFM is used to replace the unused data to free the cache space for the new incoming data items. The cache replacement is very important in the case of healthcare system as the incoming vital signs are frequent and must be replaced continuously. Second, the EPM is used to predict the incidence of GDM that may occur in the second trimester of the pregnancy. To evaluate our model, we extract data of 16,354 pregnancy women from medical information mart for intensive care (MIMIC III) benchmark dataset. For each woman, vital signs, demographic data and laboratory tests was aggregated. The results of the prediction model superior the state of the art (ACC = 0.957, AUC = 0.942). Regarding to explainability, we utilized Shapley additive explanation framework to provide local and global explanation for the developed models. Overall, the proposed framework is medically intuitive, allow the early prediction of GDM with cost effective solution.

External validation of the risk prediction model for early diabetic kidney disease in Taiwan population: a retrospective cohort study

OBJECTIVES

This study aims to independently and externally validate the Risk Prediction Model for Diabetic Kidney Disease (RPM-DKD) in patients with type 2 diabetes mellitus (T2DM).

DESIGN

This is a retrospective cohort study.

SETTING

Outpatient clinics at Lee's United Clinics, Taiwan, China.

PARTICIPANTS

A total of 2504 patients (average age 55.44 years, SD, 7.49 years) and 4455 patients (average age 57.88 years, SD, 8.80 years) were included for analysis in the DKD prediction and progression prediction cohorts, respectively.

EXPOSURE

The predicted risk for DKD and DKD progression for each patient were all calculated using the RPM-DKD.

PRIMARY AND SECONDARY OUTCOME MEASURES

The primary outcome measure was overall incidence of DKD. Secondary outcomes included DKD progression. The discrimination, calibration and precision of the RPM-DKD score were assessed.

RESULTS

The DKD prediction cohort and progression prediction cohort consisted of patients with 2504 and 4455 T2DM, respectively. The RPM-DKD examined in this study showed moderately discriminative ability with area under the curve ranged from 0.636 to 0.681 for the occurrence of DKD and 0.620 to 0.654 for the progression of DKD. The Hosmer-Lemeshow χ² test indicted the RPM-DKD was not well calibrated for predicting the occurrence of DKD and overestimated the progression of DKD. The precision for predicting the occurrence and progression of DKD were 43.2% and 42.2%, respectively.

CONCLUSIONS

On external validation, the RPM-DKD cannot accurately predict the risk of DKD occurrence and progression in patients with T2DM.

Establishment of dynamic nomogram and risk score models for T2DM: a retrospective cohort study in Beijing

BACKGROUND

Health interventions can delay or prevent the occurrence and development of diabetes. Dynamic nomogram and risk score (RS) models were developed to predict the probability of developing type 2 diabetes mellitus (T2DM) and identify high-risk groups.

METHODS

Participants (n = 44,852) from the Beijing Physical Examination Center were followed up for 11 years (2006-2017); the mean follow-up time was 4.06 ± 2.09 years. Multivariable Cox regression was conducted in the training cohort to identify risk factors associated with T2DM and develop dynamic nomogram and RS models using weighted estimators corresponding to each covariate derived from the fitted Cox regression coefficients and variance estimates, and then undergone internal validation and sensitivity analysis. The concordance index (C-index) was used to assess the accuracy and reliability of the model.

RESULTS

Of the 44,852 individuals at baseline, 2,912 were diagnosed with T2DM during the follow-up period, and the incidence density rate per 1,000 person-years was 16.00. Multivariate analysis indicated that male sex (P < 0.001), older age (P < 0.001), high body mass index (BMI, P < 0.05), high fasting plasma glucose (FPG, P < 0.001), hypertension (P = 0.015), dyslipidaemia (P < 0.001), and low serum creatinine (sCr, P < 0.05) at presentation were risk factors for T2DM. The dynamic nomogram achieved a high C-index of 0.909 in the training set and 0.905 in the validation set. A tenfold cross-validation estimated the area under the curve of the nomogram at 0.909 (95% confidence interval 0.897-0.920). Moreover, the dynamic nomogram and RS model exhibited acceptable discrimination and clinical usefulness in subgroup and sensitivity analyses.

CONCLUSIONS

The T2DM dynamic nomogram and RS models offer clinicians and others who conduct physical examinations, respectively, simple-to-use tools to assess the risk of developing T2DM in the urban Chinese current or retired employees.

Integrating the Biology of Cardiovascular Disease into the Epidemiology of Economic Decision Modelling via Mendelian Randomisation

Health economic analyses are essential for health services research, providing decision-makers and payers with evidence about the value of interventions relative to their opportunity cost. However, many health economic approaches are still limited, especially regarding the primary prevention of cardiovascular disease (CVD). In this article, we discuss some limitations to current health economic models and then outline an approach to address these via the incorporation of genomics into the design of health economic models for CVD. We propose that when a randomised clinical trial is not possible or practical, health economic models for primary prevention of CVD can be based on Mendelian randomisation analyses, a technique to assess causality in observational data. We discuss the advantages of this approach, such as integrating well-known disease biology into health economic models and how this may overcome current statistical approaches to assessing the benefits of interventions. We argue that this approach may provide the economic argument for integrating genomics into clinical practice and the efficient targeting of newer therapeutics, transforming our approach to the primary prevention of CVD, thereby moving from reactive to preventive healthcare. We end by discussing some limitations and potential pitfalls of this approach.

Gradient boosting decision tree becomes more reliable than logistic regression in predicting probability for diabetes with big data

We sought to verify the reliability of machine learning (ML) in developing diabetes prediction models by utilizing big data. To this end, we compared the reliability of gradient boosting decision tree (GBDT) and logistic regression (LR) models using data obtained from the Kokuho-database of the Osaka prefecture, Japan. To develop the models, we focused on 16 predictors from health checkup data from April 2013 to December 2014. A total of 277,651 eligible participants were studied. The prediction models were developed using a light gradient boosting machine (LightGBM), which is an effective GBDT implementation algorithm, and LR. Their reliabilities were measured based on expected calibration error (ECE), negative log-likelihood (Logloss), and reliability diagrams. Similarly, their classification accuracies were measured in the area under the curve (AUC). We further analyzed their reliabilities while changing the sample size for training. Among the 277,651 participants, 15,900 (7978 males and 7922 females) were newly diagnosed with diabetes within 3 years. LightGBM (LR) achieved an ECE of 0.0018 ± 0.00033 (0.0048 ± 0.00058), a Logloss of 0.167 ± 0.00062 (0.172 ± 0.00090), and an AUC of 0.844 ± 0.0025 (0.826 ± 0.0035). From sample size analysis, the reliability of LightGBM became higher than LR when the sample size increased more than [Formula: see text]. Thus, we confirmed that GBDT provides a more reliable model than that of LR in the development of diabetes prediction models using big data. ML could potentially produce a highly reliable diabetes prediction model, a helpful tool for improving lifestyle and preventing diabetes.

External Validation of Finnish Diabetes Risk Score and Australian Diabetes Risk Assessment Tool Prediction Models to Identify People with Undiagnosed Type 2 Diabetes: A Cross-sectional Study in Iran

BACKGROUND

Noninvasive risk prediction models have been widely used in various settings to identify individuals with undiagnosed diabetes.

OBJECTIVES

We aimed to evaluate the discrimination, calibration, and clinical usefulness of the Finnish Diabetes Risk Score (FINDRISC) and Australian Diabetes Risk Assessment (AUSDRISK) to screen undiagnosed diabetes in Kerman, Iran.

METHODS

We analyzed data from 2014 to 2018 in the second round of the Kerman Coronary Artery Disease Risk Factors Study (KERCADRS), Iran. Participants aged 35 - 65 with no history of confirmed diabetes were eligible. The area under the receiver operating characteristic curve (AUROC) and decision curve analysis were applied to evaluate the discrimination power and clinical usefulness of the models, respectively. The calibration was assessed by the Hosmer-Lemeshow test and the calibration plots.

RESULTS

Out of 3262 participants, 145 (4.44%) had undiagnosed diabetes. The estimated AUROCs were 0.67 and 0.62 for the AUSDRISK and FINDRISC models, respectively (P < 0.001). The chi-square test results for FINDRISC and AUSDRISC were 7.90 and 16.47 for the original model and 3.69 and 14.61 for the recalibrated model, respectively. Based on the decision curves, useful threshold ranges for the original models of FINDRIS and AUSDRISK were 4% to 10% and 3% to 13%, respectively. Useful thresholds for the recalibrated models of FINDRISC and AUSDRISK were 4% to 8% and 4% to 9%, respectively.

CONCLUSIONS

The original AUSDRISK model performs better than FINDRISC in identifying patients with undiagnosed diabetes and could be used as a simple and noninvasive tool where access to laboratory facilities is costly or limited.

The role of NMR-based circulating metabolic biomarkers in development and risk prediction of new onset type 2 diabetes

Associations of circulating metabolic biomarkers with type 2 diabetes (T2D) and their added value for risk prediction are uncertain among Chinese adults. A case-cohort study included 882 T2D cases diagnosed during 8-years' follow-up and a subcohort of 789 participants. NMR-metabolomic profiling quantified 225 plasma biomarkers in stored samples taken at recruitment into the study. Cox regression yielded adjusted hazard ratios (HRs) for T2D associated with individual biomarkers, with a set of biomarkers incorporated into an established T2D risk prediction model to assess improvement in discriminatory ability. Mean baseline BMI (SD) was higher in T2D cases than in the subcohort (25.7 [3.6] vs. 23.9 [3.6] kg/m²). Overall, 163 biomarkers were significantly and independently associated with T2D at false discovery rate (FDR) controlled p < 0.05, and 138 at FDR-controlled p < 0.01. Branched chain amino acids (BCAA), apolipoprotein B/apolipoprotein A1, triglycerides in VLDL and medium and small HDL particles, and VLDL particle size were strongly positively associated with T2D (HRs 1.74-2.36 per 1 SD, p < 0.001). HDL particle size, cholesterol concentration in larger HDL particles and docosahexaenoic acid levels were strongly inversely associated with T2D (HRs 0.43-0.48, p < 0.001). With additional adjustment for plasma glucose, most associations (n = 147 and n = 129 at p < 0.05 and p < 0.01, respectively) remained significant. HRs appeared more extreme among more centrally adipose participants for apolipoprotein B/apolipoprotein A1, BCAA, HDL particle size and docosahexaenoic acid (p for heterogeneity ≤ 0.05). Addition of 31 selected biomarkers to an established T2D risk prediction model modestly, but significantly, improved risk discrimination (c-statistic 0.86 to 0.91, p < 0.001). In relatively lean Chinese adults, diverse metabolic biomarkers are associated with future risk of T2D and can help improve established risk prediction models.

Performance of Stepwise Screening Methods in Identifying Individuals at High Risk of Type 2 Diabetes in an Iranian Population

BACKGROUND

Recent evidence recommended stepwise screening methods for identifying individuals at high risk of type 2 diabetes to be recruited in the lifestyle intervention programs for the prevention of the disease. This study aims to assess the performance of different stepwise screening methods that combine non-invasive measurements with lab-based measurements for identifying those with 5-years incident type 2 diabetes.

METHODS

3037 participants aged ≥30 years without diabetes at baseline in the Tehran Lipid and Glucose Study (TLGS) were followed. Thirty-two stepwise screening methods were developed by combining a non-invasive measurement (an anthropometric measurement (waist-to-height ratio, WtHR) or a score based on a non-invasive risk score [Australian Type 2 Diabetes Risk Assessment Tool, AUSDRISK]) with a lab-based measurement (different cut-offs of fasting plasma glucose [FPG] or predicted risk based on three lab-based prediction models [Saint Antonio, SA; Framingham Offspring Study, FOS; and the Atherosclerosis Risk in Communities, ARIC]). The validation, calibration, and usefulness of lab-based prediction models were assessed before developing the stepwise screening methods. Cut-offs were derived either based on previous studies or decision-curve analyses.

RESULTS

203 participants developed diabetes in 5 years. Lab-based risk prediction models had good discrimination power (area under the curves [AUCs]: 0.80-0.83), achieved acceptable calibration and net benefits after recalibration for population's characteristics and were useful in a wide range of risk thresholds (5%-21%). Different stepwise methods had sensitivity ranged 20%-68%, specificity 70%-98%, and positive predictive value (PPV) 14%-46%; they identified 3%-33% of the screened population eligible for preventive interventions.

CONCLUSION

Stepwise methods have acceptable performance in identifying those at high risk of incident type 2 diabetes.

Impact of applying a diabetes risk score in primary care on change in physical activity: a pragmatic cluster randomised trial

AIM

There is little evidence of the impact of diabetes risk scores on individual diabetes risk factors, motivation for behaviour changes and mental health. The aim of this study was to investigate the effect of applying a noninvasive diabetes risk score in primary care as component of routine health checks on physical activity and secondary outcomes.

METHODS

Cluster randomised trial, in which primary care physicians (PCPs), randomised (1:1) by minimisation, enrolled participants with statutory health insurance without known diabetes, ≥ 35 years of age with a body mass index ≥ 27.0 kg/m². The German Diabetes Risk Score was applied as add-on to the standard routine health check, conducted in the controls. Primary outcome was the difference in participants' physical activity (International Physical Activity Questionnaire) after 12 months. Secondary outcomes included body mass index, perceived health, anxiety, depression, and motivation for lifestyle change. Analysis was by intention-to-treat principle using mixed models.

RESULTS

36 PCPs were randomised; remaining 30 PCPs (intervention: n = 16; control: n = 14) recruited 315 participants (intervention: n = 153; controls: n = 162). A slight increase in physical activity was observed in the intervention group with an adjusted mean change of 388 (95% confidence interval: - 235; 1011) metabolic equivalents minutes per week. There were no relevant changes in secondary outcomes.

CONCLUSIONS

The application of a noninvasive diabetes risk score alone is not effective in promoting physical activity in primary care.

CLINICAL TRIAL REGISTRATION

ClinicalTrials.gov (NCT03234322, registration date: July 31, 2017).

Developing a simple and practical decision model to predict the risk of incident type 2 diabetes among the general population: The Di@bet.es Study

AIMS

To develop a simple multivariate predictor model of incident type 2 diabetes in general population.

METHODS

Participants were recruited from the Spanish Di@bet.es cohort study with 2570 subjects meeting all criteria to be included in the at-risk sample studied here. Information was collected using an interviewer-administered structured questionnaire, followed by physical and clinical examination. CHAID algorithm, which collects the information of individuals with and without type 2 diabetes, was used to develop a decision tree based type 2 diabetes prediction model.

RESULTS

156 individuals were identified as having developed type 2 diabetes (6.5% incidence). Fasting plasma glucose (FPG) at the beginning of the study was the main predictive variable for incident type 2 diabetes: FPG ≤ 92 mg/dL (ref.), 92-106 mg/dL (OR = 3.76, 95%CI = 2.36-6.00), > 106 mg/dL (OR = 13.21; 8.26-21.12). More than 25% of subjects starting follow-up with FPG levels > 106 mg/dL developed type 2 diabetes. When FPG <106 mg/dL, other variables (fasting triglycerides (FTGs), BMI or age) were needed. For levels ≤ 92 mg/dL, higher FTGs levels increased risk of incident type 2 diabetes (FTGs > 180 mg/dL, OR = 14.57; 4.89-43.40) compared with the group of FTGs ≤ 97 mg/dL (FTGs  = 97-180 mg/dL, OR = 3.12; 1.05-9.24). This model correctly classified 93.5% of individuals.

CONCLUSIONS

The type 2 diabetes prediction model is based on FTGs, FPG, age, gender, and BMI values. Utilizing commonly available clinical data and a simple blood test, a simple tree diagram helps identify subjects at risk of developing type 2 diabetes, even in apparently low risk subjects with normal FPG.

Integration of Risk Scores and Integration Capability in Electronic Patient Records

BACKGROUND

Digital availability of patient data is continuously improving with the increasing implementation of electronic patient records in physician practices. The emergence of digital health data defines new fields of application for data analytics applications, which in turn offer extensive options of using data. Common areas of data analytics applications include decision support, administration, and fraud detection. Risk scores play an important role in compiling algorithms that underlay tools for decision support.

OBJECTIVES

This study aims to identify the current state of risk score integration and integration capability in electronic patient records for cardiovascular disease and diabetes in German primary care practices.

METHODS

We developed an evaluation framework to determine the current state of risk score integration and future integration options for four cardiovascular disease risk scores (arriba, Pooled Cohort Equations, QRISK3, and Systematic Coronary Risk Evaluation) and two diabetes risk scores (Finnish Diabetes Risk Score and German Diabetes Risk Score). We then used this framework to evaluate the integration of risk scores in common practice software solutions by examining the software and inquiring the respective software contact person.

RESULTS

Our evaluation showed that the most widely integrated risk score is arriba, as recommended by German medical guidelines. Every software version in our sample provided either an interface to arriba or the option to implement one. Our assessment of integration capability revealed a more nuanced picture. Results on data availability were mixed. Each score contains at least one variable, which requires laboratory diagnostics. Our analysis of data standardization showed that only one score documented all variables in a standardized way.

CONCLUSION

Our assessment revealed that the current state of risk score integration in physician practice software is rather low. Integration capability currently faces some obstacles. Future research should develop a comprehensive framework that considers the reasonable integration of risk scores into practice workflows, disease prevention programs, and the awareness of physicians and patients.

Cost-effectiveness of community diabetes screening: Application of Akaike information criterion in rural communities of Nigeria

Background

The prevalence of diabetes mellitus (DM) is increasing globally, and this requires several approaches to screening. There are reports of alternative indices for prediction of DM, besides fasting blood glucose (FBG) level. This study, investigated the ability of combination of biochemical and anthropometric parameters and orodental disease indicators (ODIs) to generate models for DM prediction, using Akaike information criterion (AIC) to substantiate health economics of diabetes screening.

Methods

Four hundred and thirty-three subjects were enrolled in the study in Ndokwa communities, Delta State, Nigeria, and their glycaemic status was determined, using the CardioChek analyser^® and previous data from the Prediabetes and Cardiovascular Complications Study were also used. The cost of screening for diabetes (NGN 300 = $0.72) in a not-for-profit organization/hospital was used as basis to calculate the health economics of number of individuals with DM in 1,000 participants. Data on the subjects' anthropometric, biochemical and ODI parameters were used to generate different models, using R statistical software (version 4.0.0). The different models were evaluated for their AIC values. Lowest AIC was considered as best model. Microsoft Excel software (version 2020) was used in preliminary analysis.

Result

The cost of identifying <2 new subjects with hyperglycemia, in 1,000 people was ≥NGN 300,000 ($ 716). A total of 4,125 models were generated. AIC modeling indicates FBG test as the best model (AIC = 4), and the least being combination of random blood sugar + waist circumference + hip circumference (AIC ≈ 34). Models containing ODI parameters had AIC values >34, hence considered as not recommendable.

Conclusion

The cost of general screening for diabetes in rural communities may appear high and burdensome in terms of health economics. However, the use of prediction models involving AIC is of value in terms of cost-benefit and cost-effectiveness to the healthcare consumers, which favors health economics.

COVID-19 risk score as a public health tool to guide targeted testing: A demonstration study in Qatar

We developed a Coronavirus Disease 2019 (COVID-19) risk score to guide targeted RT-PCR testing in Qatar. The Qatar national COVID-19 testing database, encompassing a total of 2,688,232 RT-PCR tests conducted between February 5, 2020-January 27, 2021, was analyzed. Logistic regression analyses were implemented to derive the COVID-19 risk score, as a tool to identify those at highest risk of having the infection. Score cut-off was determined using the ROC curve based on maximum sum of sensitivity and specificity. The score’s performance diagnostics were assessed. Logistic regression analysis identified age, sex, and nationality as significant predictors of infection and were included in the risk score. The ROC curve was generated and the area under the curve was estimated at 0.63 (95% CI: 0.63–0.63). The score had a sensitivity of 59.4% (95% CI: 59.1%-59.7%), specificity of 61.1% (95% CI: 61.1%-61.2%), a positive predictive value of 10.9% (95% CI: 10.8%-10.9%), and a negative predictive value of 94.9% (94.9%-95.0%). The concept and utility of a COVID-19 risk score were demonstrated in Qatar. Such a public health tool can have considerable utility in optimizing testing and suppressing infection transmission, while maximizing efficiency and use of available resources.

Prediction of type 2 diabetes mellitus onset using logistic regression-based scorecards

Background

Type 2 diabetes (T2D) accounts for ~90% of all cases of diabetes, resulting in an estimated 6.7 million deaths in 2021, according to the International Diabetes Federation. Early detection of patients with high risk of developing T2D can reduce the incidence of the disease through a change in lifestyle, diet, or medication. Since populations of lower socio-demographic status are more susceptible to T2D and might have limited resources or access to sophisticated computational resources, there is a need for accurate yet accessible prediction models.

Methods

In this study, we analyzed data from 44,709 nondiabetic UK Biobank participants aged 40-69, predicting the risk of T2D onset within a selected time frame (mean of 7.3 years with an SD of 2.3 years). We started with 798 features that we identified as potential predictors for T2D onset. We first analyzed the data using gradient boosting decision trees, survival analysis, and logistic regression methods. We devised one nonlaboratory model accessible to the general population and one more precise yet simple model that utilizes laboratory tests. We simplified both models to an accessible scorecard form, tested the models on normoglycemic and prediabetes subcohorts, and compared the results to the results of the general cohort. We established the nonlaboratory model using the following covariates: sex, age, weight, height, waist size, hip circumference, waist-to-hip ratio, and body mass index. For the laboratory model, we used age and sex together with four common blood tests: high-density lipoprotein (HDL), gamma-glutamyl transferase, glycated hemoglobin, and triglycerides. As an external validation dataset, we used the electronic medical record database of Clalit Health Services.

Results

The nonlaboratory scorecard model achieved an area under the receiver operating curve (auROC) of 0.81 (95% confidence interval [CI] 0.77-0.84) and an odds ratio (OR) between the upper and fifth prevalence deciles of 17.2 (95% CI 5-66). Using this model, we classified three risk groups, a group with 1% (0.8-1%), 5% (3-6%), and the third group with a 9% (7-12%) risk of developing T2D. We further analyzed the contribution of the laboratory-based model and devised a blood test model based on age, sex, and the four common blood tests noted above. In this scorecard model, we included age, sex, glycated hemoglobin (HbA1c%), gamma glutamyl-transferase, triglycerides, and HDL cholesterol. Using this model, we achieved an auROC of 0.87 (95% CI 0.85-0.90) and a deciles' OR of ×48 (95% CI 12-109). Using this model, we classified the cohort into four risk groups with the following risks: 0.5% (0.4-7%); 3% (2-4%); 10% (8-12%); and a high-risk group of 23% (10-37%) of developing T2D. When applying the blood tests model using the external validation cohort (Clalit), we achieved an auROC of 0.75 (95% CI 0.74-0.75). We analyzed several additional comprehensive models, which included genotyping data and other environmental factors. We found that these models did not provide cost-efficient benefits over the four blood test model. The commonly used German Diabetes Risk Score (GDRS) and Finnish Diabetes Risk Score (FINDRISC) models, trained using our data, achieved an auROC of 0.73 (0.69-0.76) and 0.66 (0.62-0.70), respectively, inferior to the results achieved by the four blood test model and by the anthropometry models.

Conclusions

The four blood test and anthropometric models outperformed the commonly used nonlaboratory models, the FINDRISC and the GDRS. We suggest that our models be used as tools for decision-makers to assess populations at elevated T2D risk and thus improve medical strategies. These models might also provide a personal catalyst for changing lifestyle, diet, or medication modifications to lower the risk of T2D onset.

Funding

The funders had no role in study design, data collection, interpretation, or the decision to submit the work for publication.

Oral glucose tolerance testing as a complement to fasting plasma glucose in screening for type 2 diabetes: population-based cross-sectional analyses of 146 000 health examinations in Västerbotten, Sweden

OBJECTIVE

To assess the effect of adding an oral glucose tolerance test (OGTT) to fasting plasma glucose (FPG) in terms of detection of type 2 diabetes (T2D) and impaired glucose tolerance (IGT).

DESIGN

Retrospective analysis of serial cross-sectional screening study.

SETTING

Population-based health examinations within primary care in Västerbotten County, Sweden.

PARTICIPANTS

Individuals aged 40- 50 and 60 years with participation from 1985 to 2017. Those with previously diagnosed diabetes and FPG≥7 mmol/L were excluded.

PRIMARY AND SECONDARY OUTCOME MEASURES

Prevalence of hyperglycaemia on the OGTT (IGT and T2D defined as 2-hour postload capillary plasma glucose of 8.9-12.1 mmol/L and ≥12.2 mmol/L, respectively). Analyses were further stratified by age, sex and risk factor burden to identify groups at high or low risk of IGT and T2D on testing. The numbers needed to screen (NNS) to prevent one case of T2D through detection and treatment of IGT was estimated, combining prevalence numbers with average progression rates and intervention effects from previous meta-analyses.

RESULTS

The prevalence of IGT ranged from 0.9% (95% CI 0.7% to 1.1%) to 29.6% (95% CI 27.4% to 31.7%), and the prevalence of T2D ranged from 0.06% (95% CI 0.02% to 0.11%) to 7.0% (95% CI 5.9% to 8.3%), depending strongly on age, sex and risk factor burden. The estimated NNS to prevent one case of T2D through detection and lifestyle treatment of IGT ranged from 1332 among 40-year-old men without risk factors, to 39 among 60-year-old women with all risk factors combined.

CONCLUSIONS

The prevalence of hyperglycaemia on OGTT is highly dependent on age, sex and risk factor burden; OGTT should be applied selectively to high-risk groups to avoid unnecessary testing in the general population.

Effects of colchicine use on ischemic and hemorrhagic stroke risk in diabetic patients with and without gout

This study aimed to determine the effect of colchicine use on the risk of stroke among patients with diabetes mellitus (DM). We retrospectively enrolled patients with DM between 2000 and 2013 from the Longitudinal Health Insurance Database and divided them into a colchicine cohort (n = 8761) and noncolchicine cohort (n = 8761) by using propensity score matching (PSM). The event of interest was a stroke, including ischemic stroke and hemorrhagic stroke. The incidence of stroke was analyzed using multivariate Cox proportional hazards models between the colchicine cohort and the comparison cohort after adjustment for several confounding factors. The subdistribution hazard model was also performed for examination of the competing risk. The colchicine cohort had a significantly lower incidence of stroke [adjusted hazard ratios (aHR), 95% confidence intervals (95%CI)] (aHR = 0.61, 95%CI = 0.55–0.67), ischemic stroke (aHR = 0.59, 95%CI = 0.53–0.66), and hemorrhagic stroke (aHR = 0.66, 95%CI = 0.53–0.82) compared with the noncolchicine cohort. Drug analysis indicated that patients in the colchicine cohort who received colchicine of cumulative daily defined dose (cDDD) > 14 and duration > 28 days had a lower risk of stroke and ischemic stroke compared with nonusers. The colchicine cohort (cDDD > 150, duration > 360 days) also had a lower risk of stroke, ischemic stroke, and hemorrhagic stroke. The cumulative incidence of stroke, ischemic stroke, and hemorrhagic stroke in the colchicine cohort was significantly lower than that in the noncolchicine cohort (log-rank P < 0.001). However, the subdistribution hazard model reveal the colchicine was not associated with the hemorrhagic stroke in DM patients without gout (aHR = 0.69, 95%CI = 0.47–1.00). Colchicine use with cDDD > 14 and duration > 28 days was associated with lower risk of stroke and ischemic stroke, and colchicine use with cDDD > 150 and duration > 360 days played an auxiliary role in the prevention of stroke, ischemic stroke, and hemorrhagic stroke in patients with DM. The colchicine for the hemorrhagic stroke in DM patients without gout seem to be null effect.

Predictive value of circulating NMR metabolic biomarkers for type 2 diabetes risk in the UK Biobank study

BACKGROUND

Effective targeted prevention of type 2 diabetes (T2D) depends on accurate prediction of disease risk. We assessed the role of metabolomic profiling in improving T2D risk prediction beyond conventional risk factors.

METHODS

Nuclear magnetic resonance (NMR) metabolomic profiling was undertaken on baseline plasma samples in 65,684 UK Biobank participants without diabetes and not taking lipid-lowering medication. Among a subset of 50,519 participants with data available on all relevant co-variates (sociodemographic characteristics, parental history of diabetes, lifestyle-including dietary-factors, anthropometric measures and fasting time), Cox regression yielded adjusted hazard ratios for the associations of 143 individual metabolic biomarkers (including lipids, lipoproteins, fatty acids, amino acids, ketone bodies and other low molecular weight metabolic biomarkers) and 11 metabolic biomarker principal components (PCs) (accounting for 90% of the total variance in individual biomarkers) with incident T2D. These 11 PCs were added to established models for T2D risk prediction among the full study population, and measures of risk discrimination (c-statistic) and reclassification (continuous net reclassification improvement [NRI], integrated discrimination index [IDI]) were assessed.

RESULTS

During median 11.9 (IQR 11.1-12.6) years' follow-up, after accounting for multiple testing, 90 metabolic biomarkers showed independent associations with T2D risk among 50,519 participants (1211 incident T2D cases) and 76 showed associations after additional adjustment for HbA1c (false discovery rate controlled p < 0.01). Overall, 8 metabolic biomarker PCs were independently associated with T2D. Among the full study population of 65,684 participants, of whom 1719 developed T2D, addition of PCs to an established risk prediction model, including age, sex, parental history of diabetes, body mass index and HbA1c, improved T2D risk prediction as assessed by the c-statistic (increased from 0.802 [95% CI 0.791-0.812] to 0.830 [0.822-0.841]), continuous NRI (0.44 [0.38-0.49]) and relative (15.0% [10.5-20.4%]) and absolute (1.5 [1.0-1.9]) IDI. More modest improvements were observed when metabolic biomarker PCs were added to a more comprehensive established T2D risk prediction model additionally including waist circumference, blood pressure and plasma lipid concentrations (c-statistic, 0.829 [0.819-0.838] to 0.837 [0.831-0.848]; continuous NRI, 0.22 [0.17-0.28]; relative IDI, 6.3% [4.1-9.8%]; absolute IDI, 0.7 [0.4-1.1]).

CONCLUSIONS

When added to conventional risk factors, circulating NMR-based metabolic biomarkers modestly enhanced T2D risk prediction.

Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta-analysis

AIMS/INTRODUCTION

Recently, an increasing number of cohort studies have suggested using machine learning (ML) to predict type 2 diabetes mellitus. However, its predictive ability remains inconclusive. This meta-analysis evaluated the current ability of ML algorithms for predicting incident type 2 diabetes mellitus.

MATERIALS AND METHODS

We systematically searched longitudinal studies published from 1 January 1950 to 17 May 2020 using MEDLINE and EMBASE. Included studies had to compare ML's classification with the actual incidence of type 2 diabetes mellitus, and present data on the number of true positives, false positives, true negatives and false negatives. The dataset for these four values was pooled with a hierarchical summary receiver operating characteristic and a bivariate random effects model.

RESULTS

There were 12 eligible studies. The pooled sensitivity, specificity, positive likelihood ratio and negative likelihood ratio were 0.81 (95% confidence interval [CI] 0.67-0.90), 0.82 [95% CI 0.74-0.88], 4.55 [95% CI 3.07-6.75] and 0.23 [95% CI 0.13-0.42], respectively. The area under the summarized receiver operating characteristic curve was 0.88 (95% CI 0.85-0.91).

CONCLUSIONS

Current ML algorithms have sufficient ability to help clinicians determine whether individuals will develop type 2 diabetes mellitus in the future. However, persons should be cautious before changing their attitude toward future diabetes risk after learning the result of the diabetes prediction test using ML algorithms.

Machine learning for diabetes clinical decision support: a review

Type 2 diabetes has recently acquired the status of an epidemic silent killer, though it is non-communicable. There are two main reasons behind this perception of the disease. First, a gradual but exponential growth in the disease prevalence has been witnessed irrespective of age groups, geography or gender. Second, the disease dynamics are very complex in terms of multifactorial risks involved, initial asymptomatic period, different short-term and long-term complications posing serious health threat and related co-morbidities. Majority of its risk factors are lifestyle habits like physical inactivity, lack of exercise, high body mass index (BMI), poor diet, smoking except some inevitable ones like family history of diabetes, ethnic predisposition, ageing etc. Nowadays, machine learning (ML) is increasingly being applied for alleviation of diabetes health burden and many research works have been proposed in the literature to offer clinical decision support in different application areas as well. In this paper, we present a review of such efforts for the prevention and management of type 2 diabetes. Firstly, we present the medical gaps in diabetes knowledge base, guidelines and medical practice identified from relevant articles and highlight those that can be addressed by ML. Further, we review the ML research works in three different application areas namely—(1) risk assessment (statistical risk scores and ML-based risk models), (2) diagnosis (using non-invasive and invasive features), (3) prognosis (from normoglycemia/prior morbidity to incident diabetes and prognosis of incident diabetes to related complications). We discuss and summarize the shortcomings or gaps in the existing ML methodologies for diabetes to be addressed in future. This review provides the breadth of ML predictive modeling applications for diabetes while highlighting the medical and technological gaps as well as various aspects involved in ML-based diabetes clinical decision support.

Development and Validation of a Simple Risk Model for Predicting Metabolic Syndrome (MetS) in Midlife: A Cohort Study

Purpose

To develop and validate a simple risk model for predicting metabolic syndrome in midlife using a prospective cohort data.

Design

Prospective cohort study.

Participants

A total of 7626 members of the 1958 British birth cohort (individuals born in the first week of March 1958) participated in the biomedical survey at age 45 and have completed information on metabolic syndrome.

Methods

Variables utilised were obtained prospectively at birth, 7, 16, 23 and 45 years. Multivariable logistic regression was used to develop a total of ten (10) MetS risk prediction models taking the life course approach. Measures of discrimination and calibration were used to evaluate the performance of the models. A pragmatic criteria developed was used to select one model with the most potential to be useful. The internal validity (overfitting) of the selected model was assessed using bootstrap technique of Stata.

Main Outcome Measure

Metabolic syndrome was defined based on the NCEP-ATP III clinical criteria.

Results

There is high prevalence of MetS among the cohort members (19.6%), with males having higher risk as compared to females (22.8% vs 16.4%, P < 0.001). Individuals with MetS are more likely to have higher levels of HbA1c and low HDL-cholesterol. Similarly, regarding the individual components of MetS, male cohort members are more likely to have higher levels of glycaemia (HbA1c), BP and serum triglycerides. In contrast, female cohort members have lower levels of HDL-cholesterol and higher levels of waist circumference. Furthermore, a total of ten (10) MetS risk prediction models were developed taking the life course approach. Of these, one model with the most potential to be applied in practical setting was selected. The model has good accuracy (AUROC 0.91 (0.90, 0.92)), is well calibrated (Hosmer-Lemeshow 6.47 (0.595)) and has good internal validity.

Conclusion

Early life factors could be included in a risk model to predict MetS in midlife. The developed model has been shown to be accurate and has good internal validity. Therefore, interventions targeting socioeconomic inequality could help in the wider prevention of MetS. However, the validity of the developed model needs to be further established in an external population.

Single-nucleotide polymorphisms as important risk factors of diabetes among Middle East population

Diabetes is a chronic metabolic disorder that leads to the dysfunction of various tissues and organs, including eyes, kidneys, and cardiovascular system. According to the World Health Organization, diabetes prevalence is 8.8% globally among whom about 90% of cases are type 2 diabetes. There are not any significant clinical manifestations in the primary stages of diabetes. Therefore, screening can be an efficient way to reduce the diabetic complications. Over the recent decades, the prevalence of diabetes has increased alarmingly among the Middle East population, which has imposed exorbitant costs on the health care system in this region. Given that the genetic changes are among the important risk factors associated with predisposing people to diabetes, we examined the role of single-nucleotide polymorphisms (SNPs) in the pathogenesis of diabetes among Middle East population. In the present review, we assessed the molecular pathology of diabetes in the Middle East population that paves the way for introducing an efficient SNP-based diagnostic panel for diabetes screening among the Middle East population. Since, the Middle East has a population of 370 million people; the current review can be a reliable model for the introduction of SNP-based diagnostic panels in other populations and countries around the world.