MedPath

Mapping Diabetes in Quebec: Validating Medico-administrative Algorithms for Type 1 Diabetes, Type 2 Diabetes and LADA

Not yet recruiting
Conditions
Diabetes Mellitus, Type 1
Diabetes;Adult Onset
Diabetes, Autoimmune
Diabete Type 2
Interventions
Other: no intervention
Registration Number
NCT06573905
Lead Sponsor
Universite du Quebec en Outaouais
Brief Summary

The goal of this observational study is to validate medico-administrative algorithms that classify diabetes phenotypes (Type 1, Type 2, and Latent Autoimmune Diabetes in Adults - LADA) in a population-based cohort in Quebec, including children, adolescents, and young adults up to 40 years old with diagnosed diabetes. The main questions it aims to answer are:

Can these algorithms accurately distinguish between Type 1, Type 2, and LADA across different age groups? What is the prevalence and incidence of each diabetes phenotype in Quebec? Participants will have their medical and administrative data analyzed, including data on medication usage and healthcare visits, to validate the accuracy of the algorithms. The study will involve comparing these algorithm-based classifications with clinical diagnoses or self-reported data to ensure reliability.

Detailed Description

The goal of this observational study is to validate the effectiveness of medico-administrative algorithms developed to classify diabetes phenotypes, specifically Type 1, Type 2, and Latent Autoimmune Diabetes in Adults (LADA), in a population-based cohort in Quebec. The study focuses on children, adolescents, and young adults up to 40 years old who have been diagnosed with diabetes.

The main questions it aims to answer are:

Can these algorithms accurately differentiate between Type 1, Type 2, and LADA across various age groups? What are the prevalence and incidence rates of these diabetes phenotypes in the Quebec population? Participants, who are already diagnosed with one of the three diabetes types and receiving standard medical care, will have their data collected from existing medical and administrative records. This data includes information on medication usage, healthcare visits, and self-reported health outcomes.

The study will involve a retrospective analysis where the classifications made by the algorithms will be compared with clinical diagnoses and self-reported data to determine the accuracy and reliability of the algorithms. This validation process is crucial for improving diabetes management and public health strategies by ensuring that these algorithms can be reliably used in broader epidemiological studies.

Recruitment & Eligibility

Status
NOT_YET_RECRUITING
Sex
All
Target Recruitment
17271
Inclusion Criteria
  • Individuals diagnosed with Type 1, Type 2, or Latent Autoimmune Diabetes in Adults (LADA) based on clinical or self-reported data.
  • Participants diagnosed between 1997 and 2024.
  • Residents of Quebec with available medico-administrative records from 1997 to 2024.
Exclusion Criteria
  • Non-residents of Quebec during the study period.

Study & Design

Study Type
OBSERVATIONAL
Study Design
Not specified
Arm && Interventions
GroupInterventionDescription
other phenotypesno interventionThis group contains participants diagnosed with diabetes-related phenotypes other than Type 1, Type 2, or LADA, as well as those with rarer forms of the disease (based on clinical data). The validation aims to determine the algorithm's effectiveness in correctly identifying and classifying these less common phenotypes, which is critical for ensuring comprehensive and accurate diabetes classification.
type 2 diabetesno interventionThis group includes participants diagnosed with Type 2 diabetes based on clinical data. The validation process focuses on assessing the algorithm's accuracy in identifying individuals with Type 2 diabetes, ensuring correct classification and minimizing the risk of misclassification as other diabetes phenotypes or non-diabetic.
Latent autoimmune diabete in adultsno interventionThis group consists of participants diagnosed with Latent Autoimmune Diabetes in Adults (LADA) according to self-reported data. The validation process for this group focuses on assessing the algorithm's ability to accurately identify individuals with LADA, which is often challenging due to its characteristics that overlap with both Type 1 and Type 2 diabetes. Accurate classification of LADA is crucial for improving treatment strategies and understanding its epidemiology.
type 1 diabetesno interventionThis group comprises participants diagnosed with Type 1 diabetes according to self-reported data. The primary goal of comparing this group with medico-administrative records is to validate the algorithm's ability to accurately classify individuals with Type 1 diabetes, ensuring that they are correctly identified as such without being misclassified into other categories.
Primary Outcome Measures
NameTimeMethod
Diagnostic Accuracy Measures (Percentages)Retrospective data from 1997 to 2024

The primary outcome measure is the accuracy of the medico-administrative algorithms in correctly classifying participants into one of the following diabetes phenotypes: Type 1, Type 2, LADA, or Other Phenotypes, compared to clinical or self-reported diagnoses.

1.1. Diagnostic Accuracy Measures (Percentages)

* Sensitivity (Se)

* Specificity (Sp)

* Positive Predictive Value (PPV)

* Negative Predictive Value (NPV) All reported as proportions or percentages.

These indicators will not be aggregated into a single value, but will be presented separately to respect their distinct units of measurement.

Classification Counts (Number of Cases)Retrospective data from 1997 to 2024

The primary outcome measure is the accuracy of the medico-administrative algorithms in correctly classifying participants into one of the following diabetes phenotypes: Type 1, Type 2, LADA, or Other Phenotypes, compared to clinical or self-reported diagnoses.

1.2. Classification Counts (Number of Cases)

* True Positives (TP)

* True Negatives (TN)

* False Positives (FP)

* False Negatives (FN) All reported as counts of participants.

These indicators will not be aggregated into a single value, but will be presented separately to respect their distinct units of measurement.

Likelihood Ratios (Unitless)Retrospective data from 1997 to 2024

The primary outcome measure is the accuracy of the medico-administrative algorithms in correctly classifying participants into one of the following diabetes phenotypes: Type 1, Type 2, LADA, or Other Phenotypes, compared to clinical or self-reported diagnoses.

1.3. Likelihood Ratios (Unitless)

* Positive Likelihood Ratio (LR+)

* Negative Likelihood Ratio (LR-) Reported as unitless ratios.

These indicators will not be aggregated into a single value, but will be presented separately to respect their distinct units of measurement.

Secondary Outcome Measures
NameTimeMethod
Prevalence of Each Diabetes Phenotype (Proportion/Percentage)Retrospective data from 1997 to 2024

Prevalences of each diabetes phenotype (Type 1, Type 2, LADA, and Other Phenotypes) within the study population : Determines the proportion of individuals who have each specific diabetes phenotype (Type 1, Type 2, LADA, or Other Phenotypes) at a given point in time (Reported as a percentage or proportion).

Unit of Measure: Proportion or percentage of the study population.

These calculations will provide insights into the distribution and emergence of different diabetes phenotypes within the Quebec population from 1997 to 2024, allowing for a better understanding of disease patterns and informing public health strategies and resource allocation.

These indicators will not be aggregated into a single value, but will be presented separately to respect their distinct units of measurement.

Incidence of Each Diabetes PhenotypeRetrospective data from 1997 to 2024

Incidences of each diabetes phenotype (Type 1, Type 2, LADA, and Other Phenotypes) within the study population: Incidence (I): Calculates the rate at which new cases of each diabetes phenotype occur in the study population over the defined period (Reported as a rate or proportion).

Unit of Measure: Rate of new cases (e.g., per 1,000 person-years) or proportion (cases/total population).

These calculations will provide insights into the distribution and emergence of different diabetes phenotypes within the Quebec population from 1997 to 2024, allowing for a better understanding of disease patterns and informing public health strategies and resource allocation.

These indicators will not be aggregated into a single value, but will be presented separately to respect their distinct units of measurement.

Trial Locations

Locations (1)

Philippe Corsenac

🇨🇦

Montréal, Quebec, Canada

© Copyright 2025. All Rights Reserved by MedPath