Canadian Journal of Diabetes最新文献

{"title":"Semaglutide Use With Automated Insulin Delivery in Adults With Type 1 Diabetes: Qualitative Analyses and Patient-reported Outcomes From a Randomized Controlled Trial","authors":"Melissa-Rosina Pasqua MD, PhD ,&nbsp;Joelle Doumat BSc ,&nbsp;Michael A. Tsoukas MD ,&nbsp;Ahmad Haidar PhD","doi":"10.1016/j.jcjd.2025.10.175","DOIUrl":"10.1016/j.jcjd.2025.10.175","url":null,"abstract":"<div><h3>Objectives</h3><div>Interest in incretin therapy for type 1 diabetes is increasing, but data are lacking regarding patient-reported outcomes and personal experience among those using this drug.</div></div><div><h3>Methods</h3><div>This work is an analysis from a double-blind, randomized, crossover trial assessing semaglutide vs placebo, with automated insulin delivery in adults with type 1 diabetes, after 15-week interventions. The following questionnaires were used after each intervention: the Diabetes Distress Scale; the Hypoglycemia Fear Survey; the Diabetes Treatment Satisfaction Questionnaire; Insulin-delivery Systems: Perceptions, Ideas, Reflections and Expectations (INSPIRE); and the Diabetes Bowel Symptom Questionnaire (DBSQ). Semistructured interviews were performed at the end of the trial. Interviews were recorded, transcribed, and coded by research personnel for themes using an inductive–deductive approach.</div></div><div><h3>Results</h3><div>Twenty-three participants completed their questionnaires, whereas 24 performed interviews. For semaglutide vs placebo, only the DBSQ showed differing scores with increased symptom frequency and severity with semaglutide. Within the interviews, 42% of patients expressed interest in semaglutide use outside the study, with their reasons being lower insulin requirements, weight loss, and improved glycemic management. Many of these qualities, including complication risk reduction, were qualities of the ideal adjunctive therapy as per participants. Nausea and fear of vomiting were barriers to accurate preprandial determination of upcoming carbohydrate intake and thus preprandial bolus. Synergy was noted between the drug and automated insulin delivery by the participants.</div></div><div><h3>Conclusions</h3><div>Semaglutide is of interest to those with type 1 diabetes. Safe and accurate bolus practice by patients in the context of nausea should be reviewed during dose titration. Questionnaires did not capture differences between semaglutide and placebo outside of increased gastrointestinal side effects.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 38-46.e5"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145427170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Effects of Continuous Glucose Monitoring on Patient-reported Outcomes in Adults With Non–Insulin-dependent Type 2 Diabetes: A Randomized Controlled Trial","authors":"Inga Olu-Jordan BA ,&nbsp;Pratima Singh PharmD, MPH ,&nbsp;Jiawei Ryan Zhang ,&nbsp;Roseanne O. Yeung MD, MSc ,&nbsp;Donna P. Manca MD, MCIS ,&nbsp;Tawnya Perry RN ,&nbsp;Gulelala Rahim MD ,&nbsp;Evan Hagen MD ,&nbsp;Darren Lau MD, PhD, FRCPC","doi":"10.1016/j.jcjd.2025.10.174","DOIUrl":"10.1016/j.jcjd.2025.10.174","url":null,"abstract":"<div><h3>Objectives</h3><div>Our aim in this study was to assess the effect of a 6-week continuous glucose monitor (CGM) intervention with telemonitoring-enabled virtual diabetes educator visits on patient-reported outcomes (PROs) in adults with type 2 diabetes (T2D) not using insulin.</div></div><div><h3>Methods</h3><div>Individuals with glycated hemoglobin (A1C) &gt;7.0% (n=105) were computer randomized in an open-label parallel-group design (<span><span>clinicaltrials.gov</span><svg><path></path></svg></span> NCT05319496) to receive either 6 weeks of upfront CGM (weeks 0 to 6) (n=45) or enhanced usual care (n=41), both with virtual diabetes educator visits. The following outcomes were measured at baseline (week 0), 6 weeks, and 12 weeks using the Problem Areas in Diabetes (PAID), the Diabetes Empowerment Scale---Short Form, the EuroQol-5D Visual Analogue Scale, the International Physical Activity Questionnaire, and the UK Diabetes and Diet Questionnaire. Between-group differences in change scores for each outcome at 6 and 12 weeks were assessed using t tests.</div></div><div><h3>Results</h3><div>Of the 105 participants (mean age 57 years, 51% male, A1C 8.1%) studied, 86 completed the trial. CGM participants exhibited a greater reduction in total PAID scores at 12 weeks vs baseline when compared with those not using CGM (p=0.03), with similar between-group differences observed for PAID subscales of emotional distress (p=0.02) and food-related problems (p=0.004). CGM participants also had larger improvements in diabetes empowerment at 6 weeks (p=0.009). We did not detect any differences in the number of antihyperglycemic medication classes, physical activity, and diet.</div></div><div><h3>Conclusions</h3><div>In adults with T2D not on insulin, CGM with virtual educator visits produced improvements in emotional and food-related distress and in diabetes-related empowerment at 12 weeks.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 30-37.e2"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145403150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Safety and Effectiveness of Subcutaneous Insulin for Management of Mild to Moderate Diabetic Ketoacidosis in Nonpregnant Individuals: A Retrospective Cohort Study at a Tertiary Care Centre","authors":"Judy K. Qiang MD, MSc ,&nbsp;Alex Kiss PhD ,&nbsp;Wanqing Yu MD ,&nbsp;Andre Carlos Kajdacsy-Balla Amaral MD ,&nbsp;Dominick Shelton MD, MSc ,&nbsp;Lowyl Notario BScN, MSc ,&nbsp;Ilana Halperin MD, MSc","doi":"10.1016/j.jcjd.2025.10.173","DOIUrl":"10.1016/j.jcjd.2025.10.173","url":null,"abstract":"<div><h3>Objectives</h3><div>Subcutaneous (SC) insulin has been studied as an alternative to intravenous (IV) insulin to reduce resource consumption in the management of diabetic ketoacidosis (DKA). However, feasibility and safety of an SC protocol have not been demonstrated in a Canadian context.</div></div><div><h3>Methods</h3><div>In this retrospective cohort study we examined the association between IV and SC insulin treatment for DKA and time to anion gap (AG) closure and length of stay (LOS) in hospital at a large Canadian tertiary care centre in nonpregnant adults with mild to moderate DKA. Rates of hypoglycemia, hypokalemia, and AG reopening requiring intervention were compared between treatment groups.</div></div><div><h3>Results</h3><div>Compared with the SC-treated group, the IV group had a shorter time to AG closure (median time to AG closure 15.6 hours in the IV group vs 24.0 hours in the SC group, adjusted hazard ratio 0.65, 95% confidence interval [CI] 0.45 to 0.92, p=0.02). IV-treated patients had much higher rates of hypoglycemia (1.6% in SC treated vs 19.6% in IV treated). SC-treated patients had a much lower rate of hypokalemia compared with IV-treated patients (adjusted odds ratio 0.35, 95% CI 0.15 to 0.80, p=0.01). However, the groups had similar LOSs in hospital and rates of AG reopening requiring intervention.</div></div><div><h3>Conclusions</h3><div>These results suggest SC insulin is safe. Although it may take 8.4 hours longer to close the AG with SC insulin, there is less hypoglycemia and hypokalemia and no difference in LOS in hospital.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 23-29.e6"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145370604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-world Impact of Type 2 Diabetes Mellitus in Employer-sponsored Benefit Plans","authors":"Godfrey Mau BScPhm,&nbsp;Julie Hviid Hahn-Pedersen MSc","doi":"10.1016/j.jcjd.2025.10.170","DOIUrl":"10.1016/j.jcjd.2025.10.170","url":null,"abstract":"<div><h3>Objective</h3><div>Our aim in this study was to quantify the health-benefit costs associated with type 2 diabetes mellitus (T2D) among employees aged 40 to 59 years in Canadian employer-sponsored benefit plans.</div></div><div><h3>Methods</h3><div>We analyzed a self-insured health benefit plan database of &gt;100,000 employees during 2019 to 2023 in Canada. Employees with T2D or no diabetes were identified, and claims were examined across prescription drug, extended health care (EHC), short-term disability (STD), and long-term disability (LTD).</div></div><div><h3>Results</h3><div>In 2023, employees with T2D (n=3,575) had higher annual per-person total benefits cost of $6,488 vs $3,003 for those without diabetes (n=31,314). This was mainly due to higher drug claims ($3,471 vs $1,047) and LTD claims ($1,745 vs $983). A higher proportion of employees with T2D had a disability claim compared to those without (LTD 9.6% vs 5.5%, age-adjusted prevalence ratio 1.64; STD 8.6% vs 5.4%, age-adjusted prevalence ratio 1.63). The greatest difference in disability claims was for cardiovascular conditions (LTD 0.48% vs 0.06%, age-adjusted prevalence ratio 7.77; STD 0.48% vs 0.08%, age-adjusted prevalence ratio 7.65).</div></div><div><h3>Conclusions</h3><div>This study underscores the substantial economic impact of T2D across health benefits for employers, highlighting the opportunity for employers to help their employees to better manage their diabetes by ensuring access to innovative medications and disease management programs.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 15-22.e5"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145305044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lowering the Diagnostic Threshold for Gestational Diabetes: A Comparison of 2 Centres","authors":"Sanaz Azizi MD, MSc ,&nbsp;Agnieszka Majdan MD ,&nbsp;Rachel Bond MD ,&nbsp;Natasha Garfield MD ,&nbsp;Sara Meltzer MD ,&nbsp;Shaun Eintracht MD ,&nbsp;Kathryn Morrison PhD ,&nbsp;Julia Ma MPH ,&nbsp;Kaberi Dasgupta MD, MSc ,&nbsp;Tricia Peters MD, PhD","doi":"10.1016/j.jcjd.2025.10.176","DOIUrl":"10.1016/j.jcjd.2025.10.176","url":null,"abstract":"<div><h3>Objective</h3><div>In this study, we compared 2 gestational diabetes mellitus (GDM) diagnostic thresholds in relation to large-for-gestational-age (LGA) status and secondary adverse pregnancy outcomes.</div></div><div><h3>Methods</h3><div>This retrospective cohort study examined 840 pregnant women who underwent 2-step GDM screening at 2 hospital centres in Montréal that use distinct GDM diagnostic thresholds. At the second step of GDM screening, one centre used glucose thresholds suggested by the Diabetes Canada preferred approach and the other centre used the lower International Association of Diabetes and Pregnancy Study Groups thresholds. We defined mild hyperglycemia (MH) as having intermediate glucose values that were diagnostic of GDM at one centre but not the other (fasting plasma glucose [PG] 5.1 to 5.2 mmol/L, 1-hour PG 10 to 10.5 mmol/L, or 2-hour PG 8.5 to 8.9 mmol/L). We conducted multivariable linear and logistic regression analyses to evaluate outcomes for women with untreated or treated MH compared to those diagnosed with GDM by higher glucose thresholds within the same centre, and we also explored differences between centres.</div></div><div><h3>Results</h3><div>The odds of LGA offspring were 3-fold higher (adjusted odds ratio [OR] 3.01, 95% confidence interval [CI] 1.47 to 6.19) among women with untreated MH compared with women treated for GDM at the same centre. Also, the odds of macrosomia were over 2-fold higher when comparing treated MH with GDM. In addition, untreated compared with treated MH had lower odds of induction of labour (adjusted OR 0.28, 95% CI 0.11 to 0.70).</div></div><div><h3>Conclusion</h3><div>Failure to optimally treat MH during pregnancy is associated with fetal overgrowth and may affect obstetrical management.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 47-54.e1"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145454228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CanDEG: Launching the Canadian Diabetes Epidemiology Group","authors":"Baiju R. Shah MD, PhD ,&nbsp;Kimberley Hanson MBA ,&nbsp;Gillian L. Booth MD, MSc ,&nbsp;Sonia Butalia MD ,&nbsp;Ghazal S. Fazli PhD ,&nbsp;Calvin Ke MD, PhD ,&nbsp;Lorraine L. Lipscombe MD, MSc ,&nbsp;Meranda Nakhla MD ,&nbsp;Brandy Wicklow MD ,&nbsp;Roseanne O. Yeung MD ,&nbsp;Kaberi Dasgupta MDCM ,&nbsp;Canadian Diabetes Epidemiology Group","doi":"10.1016/j.jcjd.2025.10.177","DOIUrl":"10.1016/j.jcjd.2025.10.177","url":null,"abstract":"","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 55-57"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145454139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"General Population Screening for Type 1 Diabetes in Canada---CanScreenT1D","authors":"Peter A. Senior BMedSci, MBBS, PhD ,&nbsp;Ashish Marwaha BMBCh, MA(Oxon), PhD ,&nbsp;Holly O. Witteman PhD ,&nbsp;Pranesh Chakraborty MD ,&nbsp;Robin Z. Hayeems ScM, PhD ,&nbsp;Diane K. Wherrett MD ,&nbsp;CanScreenT1D Consortium","doi":"10.1016/j.jcjd.2025.11.001","DOIUrl":"10.1016/j.jcjd.2025.11.001","url":null,"abstract":"","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 58-61"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145491157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Postpartum Diabetes Screening and Conversion Rates Among Women Diagnosed With Gestational Diabetes Mellitus Using Standard vs Modified Criteria During the COVID-19 Pandemic","authors":"Sonia Butalia MD, FRCPC, MSc ,&nbsp;Olesya Barrett PhD ,&nbsp;Anamaria Savu PhD ,&nbsp;Vichy Liyanage PhD ,&nbsp;Peter Senior MD ,&nbsp;Roseanne O. Yeung MD, FRCPC, MPH ,&nbsp;Padma Kaul PhD","doi":"10.1016/j.jcjd.2025.10.001","DOIUrl":"10.1016/j.jcjd.2025.10.001","url":null,"abstract":"<div><h3>Objective</h3><div>Our aim in this work was to examine postpartum diabetes screening and rates in women diagnosed with gestational diabetes mellitus (GDM) using standard vs modified criteria during the COVID-19 pandemic.</div></div><div><h3>Methods</h3><div>Women with GDM pregnancies between January 1, 2020, and December 31, 2021, in Alberta, Canada, were stratified by the GDM diagnosis criteria and followed for 18 months postpartum diabetes screening. Proportions of prediabetes and diabetes were compared between the standard vs modified GDM criteria groups at 6 and 18 months. Multivariable logistic regression analysis was used to examine differences in prediabetes and diabetes rates between the 2 GDM criteria groups after adjusting for baseline differences.</div></div><div><h3>Results</h3><div>Among 10,238 individuals with GDM, 780 were diagnosed using the modified criteria and 9,458 were diagnosed using the standard criteria. There was no difference in the proportion of individuals who underwent postpartum screening by 6 months (27.1% vs 28.9%, p=0.29) or by 18 months (43.1% vs 45.3%, p=0.24) among the modified and standard groups, respectively. Diabetes proportions were higher in women diagnosed with GDM using the modified criteria compared with those diagnosed using the standard criteria (27.0% vs 4.2%, p&lt;0.0001; adjusted odds ratio 8.18, 95% confidence interval 5.76 to 11.6). Proportions of prediabetes and diabetes at 18 months were 15.2% and 20.8% (p=0.014) and 29.8% and 6.1% (p&lt;0.0001) for modified and standard groups, respectively.</div></div><div><h3>Conclusions</h3><div>Regardless of the GDM diagnostic method, postpartum diabetes screening among women with GDM was suboptimal during the COVID-19 pandemic. The modified criteria for GDM identified a group of women who were at higher risk for conversion to diabetes.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 1-7"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145276947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"“It’s a Chronic, Vicious Cycle”: Diabetes Health-care Professionals’ Perceptions of the Impact of Knee Osteoarthritis on Type 2 Diabetes Management","authors":"Joy Du BSc, MD ,&nbsp;Esther J. Waugh PhD, BScPT, MSc ,&nbsp;Rebecca Voth MD ,&nbsp;Crystal MacKay PhD, BScPT ,&nbsp;Ian Stanaitis BMSc ,&nbsp;Gillian A. Hawker MSc, MD, FRCPC ,&nbsp;Lorraine L. Lipscombe MD, MSc, FRCPC ,&nbsp;Lauren K. King MD, PhD, FRCPC","doi":"10.1016/j.jcjd.2025.10.169","DOIUrl":"10.1016/j.jcjd.2025.10.169","url":null,"abstract":"<div><h3>Objectives</h3><div>Type 2 diabetes (T2D) and knee osteoarthritis (OA) frequently co-occur, and concomitant knee OA increases risk for diabetes complications. Despite this, OA is frequently undertreated. Diabetes health-care professionals’ (HPs’) perceptions of the impact of knee OA in people with T2D may impact how it is addressed in clinical practice. We aimed to understand how diabetes HPs perceive the impact of knee OA on diabetes management and outcomes.</div></div><div><h3>Methods</h3><div>In this qualitative study we performed a secondary analysis of semistructured interviews with 18 diabetes HPs (primary care providers, endocrinologists, and diabetes educators) in Ontario, Canada. Transcripts were inductively coded and thematically analyzed.</div></div><div><h3>Results</h3><div>We developed 3 themes: 1) Patients commonly raise OA-related concerns during diabetes appointments; 2) Impact of OA on diabetes management; and 3) Conscious disconnect between perceived patient and HP priorities. Diabetes HPs recognized that knee OA commonly co-occurred in their patients. Most HPs perceived that OA has deleterious effects on diabetes management through physical inactivity, as well as other mechanisms. Despite observing OA’s impact on their patients, most participants did not address OA due to the focussed structure of diabetes appointments, “single-problem” appointments, and culture of siloed care.</div></div><div><h3>Conclusions</h3><div>Diabetes HPs recognized the high prevalence of knee OA in their patients and its deleterious effects on diabetes management, although OA management was usually not prioritized. This highlights a missed opportunity in optimizing care for people with T2D. Implementing strategies to promote OA care during diabetes visits may improve disease outcomes for both conditions.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"50 1","pages":"Pages 8-14.e2"},"PeriodicalIF":2.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145288071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting the Risk of COVID-19 Among Adult Patients With Diabetes: A Machine Learning Approach","authors":"Dean T. Eurich PhD ,&nbsp;Darren Lau MD ,&nbsp;Weiting Li MSc ,&nbsp;Olivia Weaver MSc ,&nbsp;Tanya Joon MSc ,&nbsp;Ming Ye PhD ,&nbsp;Finlay A. McAlister MD ,&nbsp;Padma Kaul PhD ,&nbsp;Salim Samanani MD","doi":"10.1016/j.jcjd.2025.09.001","DOIUrl":"10.1016/j.jcjd.2025.09.001","url":null,"abstract":"<div><h3>Objectives</h3><div>In this study our aim was to develop a machine learning model that could accurately predict the risk of acquiring COVID-19 in community-dwelling adults with type 1 and/or type 2 diabetes in Alberta, Canada.</div></div><div><h3>Methods</h3><div>This predictive supervised machine learning study included adults (≥18 years old) living in Alberta, Canada, between April 1, 2019, and March 31, 2021, with pre-existing diabetes (n=372,055, excluding 2,541 due to migration; final sample size 369,514). The outcome of interest was a positive severe acute respiratory syndrome–coronavirus-2 (SARS-CoV-2) polymerase chain reaction test result between March 1, 2020, and March 1, 2021. Model features were extracted from routinely collected Alberta administrative health data from March 1, 2015, to March 1, 2020. Fifteen algorithms were trained on 67% of the data and the top performer (Light Gradient Boost [LGBoost] model) was validated on the remaining 33%. The model was calibrated and model performance was assessed using area under the receiver-operating characteristic curve (AUROC), area under the precision recall curve (AUPRC), and threshold analyses.</div></div><div><h3>Results</h3><div>Among the 369,514 individuals with diabetes, 140,511 were tested, of whom 13,082 had a positive SARS-CoV-2 test. The LGBoost model incorporated 367 features with AUROC and AUPRC of 0.69 and 0.08, respectively. The model was well-calibrated for common risk thresholds (&lt;0.2 probability) with high specificity (≥0.98 at all thresholds); however, sensitivity and positive predictive values were low at all thresholds (≤0.08 and ≤0.18, respectively).</div></div><div><h3>Conclusions</h3><div>The LGBoost model lacked the sensitivity to be clinically useful in predicting SARS-CoV-2 infection in Albertans with diabetes. Alternative data sources may be required to improve future COVID-19 prediction models from the community.</div></div>","PeriodicalId":9565,"journal":{"name":"Canadian Journal of Diabetes","volume":"49 8","pages":"Pages 446-453.e10"},"PeriodicalIF":2.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145056521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}