Improving T2DM Detection Using Spot cHbA1c Test

Overview

Objectives: To evaluate the effectiveness of active opportunistic screening using point-of-care capillary Hemoglobin-A1c (POC-cHbA1c) testing, compared to venous HbA1c (vHbA1c) testing, in improving detection of type 2 diabetes mellitus (T2DM) among at-risk primary care patients. Design: Pragmatic cluster randomized controlled trial. Setting: 8 public primary care clinics in Hong Kong. Participants: A minimum of 776 patients (97 per clinic) who have ≥1 risk factor for T2DM, but no known diagnosis of DM or DM screening in the past 12 months. Intervention: Participants at intervention clinics (n=4) will be offered free POC-cHbA1c testing on-site, immediately informed of test results and DM risk, and scheduled for confirmatory oral glucose tolerance test (OGTT) if screened positive (i.e. HbA1c≥5.6%). Participants at control clinics will be offered free vHbA1c testing scheduled on a separate day, informed of test results and DM risk via phone, and scheduled for confirmatory OGTT if screened positive. Main outcome measures: Primary outcomes are uptake rate of POC-cHbA1c versus vHba1c testing, and difference in proportion of T2DM detected between intervention and control groups. Secondary outcomes include number-needed-to-screen to detect one more T2DM case. Data analysis: Participants' characteristics and test uptake rates will be summarized by descriptive statistics. Difference in the proportion of T2DM detected between groups will be compared using Chi-squared test. Number-needed-to screen to identify one additional patient with DM will be calculated. Expected results: A greater proportion of T2DM patients will be detected by POC-cHbA1c than vHbA1c due to a higher screening test uptake rate among the studied population.

Aim: This pragmatic cluster randomized controlled trial aims to determine the effectiveness of active opportunistic screening using point-of-care capillary HbA1c (POC-cHbA1c) testing compared to venous HbA1c (vHbA1c) testing in improving T2DM detection among at-risk public primary care patients, to estimate the impact of such a screening strategy in the real-world setting. The ultimate goal would be to fill the current preventive care gap and reduce the burden of undiagnosed T2DM in Hong Kong. Objectives: Primary Objectives 1. To determine the difference in the proportion of T2DM detected between the intervention group (POC-cHbA1c testing) and control group (conventional vHbA1c testing). 2. To evaluate the uptake rate of the POC-cHbA1c test relative to the vHbA1c test among at-risk public primary care patients in Hong Kong. Secondary Objectives 1. To determine the proportion of subjects with high-risk HbA1c concentration (i.e. HbA1c≥5.6%) among the studied at-risk group. 2. To compare the difference in uptake rate of confirmatory OGTT test between intervention and control groups. 3. To calculate the number-needed-to-screen to detect one more case with T2DM using POC-cHbA1c testing compared to vHbA1c testing. 4. The proportion of patients who refuse to join the study (among all eligible subjects). Hypotheses: 1. A greater proportion of T2DM patients will be detected in the intervention group (POC-cHbA1c testing) than in the control group (vHbA1c). 2. Uptake rate of POC-cHbA1c testing will be higher than vHbA1c testing among the studied population. 3. Uptake rate of confirmatory OGTT among screened positive patients (i.e. HbA1c ≥5.6%) in the intervention group will be higher than for those in the control group. Data analysis: Descriptive statistics will be used to summarize patients' characteristics in the intervention (POC-cHbA1c testing) group and control (vHbA1c testing) group. The difference in characteristics between groups will be assessed by independent t-tests for continuous variables or Chi-square tests for categorical variables. The uptake rate of POC-cHbA1c testing and vHbA1c testing will be reported. The difference in the proportion of T2DM detected between groups will also be compared using a Chi-squared test. The number-needed-to-screen to identify one additional patient with DM will be calculated. To take cluster effect into account, a mixed effect logistic model will be used to evaluate the effect of the intervention on the uptake rate by treating the clinic as a random effect and adjusting for the patients' characteristics. All statistical analyses will be performed using Stata. All significance tests will be two-tailed and findings with a p-value less than 0.05 will be considered statistically significant. Potential problem: Missing data will be handled by using multiple imputation. The chained equation method will be used to impute each missing value five times, adjusted for all baseline covariates and outcomes. The same analysis method will be adopted for each of the five imputed datasets, and the results will be pooled using Rubin's rule. Complete case analysis will also be conducted to confirm the results by multiple imputation analysis.