Authors (including presenting author) :
Chan L(1)(2)(3) (Presenting Author), Yu EYT(4), Wan EYF(1)(5), Wong SYS(6), Chao DVK(7), Ko WWK(7), Chen CXR(7), Chan PPL(1)(2), Bilney EVM(1), Lee ES(8), Ng WL(9), Lam CLK(1)
Affiliation :
(1)Department of Family Medicine and Primary Care, The University of Hong Kong, Hong Kong, China.
(2)The Bau Institute of Medical and Health Sciences Education, The University of Hong Kong, Hong Kong, China.
(3)Department of Family Medicine and Primary Care, The University of Hong Kong-Shenzhen Hospital, Shenzhen, Guangdong, China.
(4)Primary Healthcare Commission, Health Bureau, Hong Kong, China.
(5)Department of Pharmacology and Pharmacy, The University of Hong Kong, Hong Kong, China.
(6)The Jockey Club School of Public Health and Primary Care, The Chinese University of Hong Kong, Hong Kong, China.
(7)Hong Kong Hospital Authority, Hong Kong, China.
(8)Lee Kong Chian School of Medicine, Nanyang Technology University, Singapore.
(9)Department of Primary Care Medicine, Faculty of Medicine, Universiti Malaya, Kuala Lumpur, Malaysia.
Introduction :
In Hong Kong (HK), type 2 diabetes is a growing public health concern, with prevalence rates of 8.5% in the general, and 14.6% in the high-risk populations respectively. Its early asymptomatic nature often delays detection, leaving many cases undiagnosed and potentially underestimating the true prevalence of type 2 diabetes in HK. In 2020, approximately 36.6% of individuals with type 2 diabetes in HK were undiagnosed - double the rate observed in the United Kingdom. Late diagnosis exacerbates the risk of developing diabetes-related complications and mortality. Early detection through effective screening is essential to mitigate these risks, reduce healthcare burdens, and improve patient outcomes.
In HK, type 2 diabetes screening is largely conducted in the private sector, requiring patients to bear the costs. Public primary care clinics (i.e., General Out-Patient Clinics (GOPCs)), provide limited low-cost screening predominantly for individuals with hypertension, a known risk factor for type 2 diabetes. This approach leads to diagnostic delays, particularly among socioeconomically disadvantaged populations who are unable to afford private screening and are hence inadequately screened. To address this gap, the Chronic Disease Co-Care (CDCC) Pilot Scheme was introduced, offering subsidised screening and management for chronic diseases, such as type 2 diabetes, through private healthcare providers in collaboration with District Health Centres or District Health Centre Expresses. However, the scheme relies heavily on proactive health-seeking behaviours by eligible individuals, which poses a barrier to widespread participation. A more accessible public screening approach is urgently needed to reduce the burden of undiagnosed type 2 diabetes and its associated complications.
The choice of screening test also influences the high rate of undiagnosed cases. Fasting plasma glucose (FPG) is the most commonly used test in HK due to its low cost, while the oral glucose tolerance test (OGTT) serves as a confirmatory diagnostic tool. However, FPG has limited sensitivity (50%) at the standard cut-off (≥7.0 mmol/L). In contrast, venous HbA1c (vHbA1c) is a more stable diagnostic measure that does not require fasting and is comparable to FPG in predicting diabetic retinopathy. Yet, logistical barriers in public clinics limit its implementation during routine visits. Point-of-care (POC) HbA1c assays, which use capillary blood (cHbA1c), offer a promising alternative. These assays provide rapid and accurate results without requiring venepuncture. Despite its potential, POC-cHbA1c testing in HK’s primary care clinics is currently restricted to monitoring glycaemic control in patients already diagnosed with diabetes.
Objectives :
Our study aimed to evaluate a two-step active opportunistic screening strategy in primary care that combined risk factor assessment with HbA1c testing to improve the detection of type 2 diabetes among at-risk individuals.
The primary objectives were to compare POC-cHbA1c (intervention) with vHbA1c testing (control) in terms of: 1) the uptake rate of HbA1c testing; and 2) the proportion of type 2 diabetes cases detected. Secondary objectives included determining: 1) the number-needed-to-screen (NNS) using POC-cHbA1c to identify one more type 2 diabetes case compared to vHbA1c testing; 2) the difference in uptake rates of confirmatory OGTT following abnormal HbA1c results (i.e., HbA1c ≥5.6%) between the intervention and control clinics; 3) the difference in the proportion of pre-diabetes (i.e., impaired fasting glucose [IFG] or impaired glucose tolerance [IGT]) detected between the intervention and control clinics; and 4) the difference in the proportion of type 2 diabetes plus pre-diabetes combined detected between the intervention and control clinics.
By addressing the above objectives, our study sought to establish the effectiveness of POC-cHbA1c testing as a more patient-centred and accessible approach to timely type 2 diabetes detection in primary care settings.
Methodology :
A cluster randomised controlled trial was conducted from June 2022 to January 2024 to assess the effectiveness of a two-step active opportunistic screening strategy using POC-cHbA1c compared to vHbA1c testing in improving the detection of type 2 diabetes among public primary care patients with risk factors. The trial involved eight GOPCs from four Hospital Authority clusters in HK, with two clinics per cluster randomised to either the intervention (POC-cHbA1c) or control (vHbA1c) group. Randomisation at the clinic level was conducted using a random allocation sequence produced by statistical software.
A total of 852 at-risk patients were recruited through consecutive sampling during primary care consultations following a two-step screening process. In step one, patients attending the eight participating GOPCs underwent opportunistic screening for type 2 diabetes risk factors based on criteria from the World Health Organisation/International Diabetes Federation and the Hong Kong Reference Framework for Diabetes Care for Adults in Primary Care Settings. Eligible at-risk patients who met none of the exclusion criteria were informed of their type 2 diabetes risk and offered free HbA1c testing.
In step two, participants at intervention clinics were immediately offered free on-site POC-cHbA1c testing during the recruitment visit, with results available within 10 minutes. High risk for type 2 diabetes was defined as HbA1c levels ≥5.6%, a threshold with high sensitivity (96.1%) and negative predictive value (94.5%) in the local population. Those with HbA1c levels ≥5.6% were promptly informed of their elevated risk and invited to schedule a confirmatory OGTT within 2-4 weeks at the same clinic. At control clinics, eligible participants were offered free vHbA1c testing, scheduled for a separate visit within 1-2 weeks. Patients were informed of their results via telephone when their lab reports returned, and those with HbA1c levels ≥5.6% were also invited during the phone conversation to schedule a confirmatory OGTT within 2-4 weeks at the corresponding control clinic.
The primary outcome measures were: 1) HbA1c test uptake; and 2) the proportion of type 2 diabetes cases detected. Secondary outcomes included: 1) the NNS using POC-cHbA1c to detect one more type 2 diabetes case versus vHbA1c; 2) confirmatory OGTT uptake; 3) pre-diabetes detection; and 4) the combined detection rate of either type 2 diabetes or pre-diabetes. Multilevel logistic regression analyses were used to evaluate intervention effects while adjusting for patient characteristics and clinic-level clustering.
Result & Outcome :
Our study was the first to evaluate the real-world effectiveness of a two-step active opportunistic screening strategy using POC-cHbA1c testing compared to vHbA1c in detecting type 2 diabetes among at-risk public primary care patients in HK. The findings aligned with our hypotheses, showing that the intervention group achieved higher rates of type 2 diabetes detection, driven by greater uptake of POC-cHbA1c testing and confirmatory OGTT.
The uptake rate for POC-cHbA1c testing in the intervention group was 76.0%, significantly higher than the 37.5% for vHbA1c testing (OR=7.06, 95% CI [2.47–20.18], p< 0.001). This highlights the potential of POC-cHbA1c to improve screening participation, enabling earlier detection and intervention. Among the risk factors, hyperlipidaemia, obesity, and having a first-degree relative with diabetes, significantly enhanced patients' willingness to undergo HbA1c testing and subsequent OGTT, potentially due to increased awareness and motivation.
POC-cHbA1c detected a higher proportion of type 2 diabetes cases compared to vHbA1c (4.2% vs. 1.4%, p=0.016) and also identified more pre-diabetes cases (11.8% vs. 6.9%, p=0.015). The NNS was 61, indicating that one additional case of type 2 diabetes was identified for every 61 individuals screened using POC-cHbA1c compared to vHbA1c. Overall, POC-cHbA1c detected significantly more cases of either type 2 diabetes or pre-diabetes combined (OR=1.99, 95% CI [1.01–3.95], p=0.048).
POC-cHbA1c testing also significantly improved confirmatory OGTT uptake rates, with a 74% higher likelihood of participation compared to vHbA1c (OR=1.74, 95% CI [1.00–3.02], p=0.049). This suggests a cascading effect, where the convenience of POC-cHbA1c testing increased the likelihood of follow-up diagnostic testing. Interestingly, hypertensive patients were less likely to participate in HbA1c testing or OGTT. This may be attributed to the Risk Factor Assessment and Management Programme for Hypertension (RAMP-HT) offered in GOPCs, which includes annual low-cost type 2 diabetes screening. Understandably, such patients would be reluctant to have additional HbA1c testing and OGTT if they are already being screened regularly through RAMP-HT.
Overall, these findings demonstrate the potential of POC-cHbA1c testing to enhance type 2 diabetes screening and detection in primary care. Its higher uptake rates and greater type 2 diabetes detection compared to vHbA1c make it a valuable tool for expanding screening coverage, accelerating diagnosis, and facilitating timely treatment.
