Glycated haemoglobin trajectories and one-year risk of potentially avoidable hospitalisations among adult patients with type 2 diabetes from specialist outpatient clinics of a tertiary hospital: a cohort study

• Main Authors: Htun, Htet Lin, Lian, Weixiang, Phua, Hwee Pin, Lim, Moses Yidong, Quek, Timothy Peng Lim, Chew, Daniel Ek Kwang, Lim, Wei Yen
• Other Authors: Lee Kong Chian School of Medicine (LKCMedicine)
• Format: Article
• Language: English
• Published: 2023
• Subjects: Science::Medicine; Glycated Hemoglobin A; Glycemic Control
• Online Access: https://hdl.handle.net/10356/172243
• Institution: Nanyang Technological University

Aim: To evaluate the association between trajectories of glycated haemoglobin (HbA1c) and potentially avoidable hospitalisations (PAH). Methods: We performed a cohort study in a tertiary hospital in Singapore among adult type 2 diabetes patients with ≥ 3 HbA1c tests over two years. Then, we followed up for one year after the last HbA1c reading to determine the PAH outcome. Glycaemic control was analysed by (1) HbA1c trajectories through group-based trajectory modelling, and (2) mean HbA1c. PAH was defined using the Agency of Healthcare Research and Quality criteria, categorising as overall, diabetes, acute, and chronic-composites. Results: A total of 14,923 patients (mean age: 62.9 ± 12.8 years; 55.2% men) were included. Four HbA1c trajectories were observed; low-stable (n = 9854, 66.0%), moderate-stable (n = 3125, 20.9%), high-decrease (n = 1017, 6.8%) and high-persistent (n = 927, 6.2%). Compared to the low-stable trajectory, one-year risk ratio (RR) and 95% confidence interval (CI), respectively for moderate-stable, high-decrease and high-persistent trajectories were as follows: (1) overall PAH: 1.15 (1.00–1.31), 1.53 (1.31–1.80), 1.96 (1.58–2.43); (2) diabetes PAH: 1.30 (1.04–1.64), 1.98 (1.55–2.53), 2.24 (1.59–3.15); (3) acute PAH: 1.14 (0.90–1.44), 1.29 (0.95–1.77), 1.75 (1.17–2.62); and (4) chronic PAH: 1.21 (1.02–1.43), 1.62 (1.34–1.97), 2.14 (1.67–2.75). Mean HbA1c was significantly associated with overall and chronic-composites of PAH whilst evidence of a non-linear relationship with diabetes-composite of PAH was noted. Conclusion: Patients with high-decrease trajectory had a risk lower than those with persistently-high HbA1c, highlighting that a greater risk of hospitalisation conferred by poor glycaemic control is potentially reversible. Determining HbA1c trajectories could help to identify the high-risk individuals for targeted and intensive management to improve care and reduce hospitalisations.