Regular mailing of personalized feedback reports improves glycemic control in diabetes: A randomized controlled trial^†

To the Editor

Feedback reports that address personal concerns are more effective in motivating health-related behavioral changes than generic information.1 In the present randomized quality improvement (QI) program, we examined the effects of mailing regular personalized follow-up reports on metabolic control and hospitalization in patients with diabetes receiving integrated care augmented by information technology (IT).2, 3 This QI program took place at a university-affiliated hospital with 1300 beds and 22 medical clinics, which managed a population of 1 million, predominantly Chinese, in Hong Kong. Apart from providing education and care coordination, the Diabetes Centre offers a twice-weekly structured diabetes comprehensive assessment (CA) service to all inpatients and outpatients upon referral. All clinical and laboratory data were anonymously entered into the web-based Joint Asia Diabetes Evaluation (JADE) portal with built-in modules for structured data collection, risk stratification, issue of personalized reports, and decision support. The JADE portal was developed and supported by the Asia Diabetes Foundation as a research tool to integrate different components of QI initiatives targeted at system, healthcare providers and patients.4

In a randomized study that examined the effects of a telephone-based peer support program, we reported the benefits of the IT-augmented JADE Program with periodic CA and group empowerment in improving risk factor control by reducing clinical inertia and non-adherence.5 After the CA, all patients attended their usual follow-up clinics 2–4 monthly for medical review. Despite an average consultation time of 5–10 min, most doctors arranged measurement of blood pressure (BP) and body weight, as well as laboratory tests available in the hospital-based Clinical Management System (CMS).

In 2013, we asked the question whether mailing periodic follow-up reports between CA visits would further improve outcomes. We consecutively recruited 1200 patients with diabetes (95.8% with type 2 diabetes) who underwent CA and randomized 600 to receive two JADE follow-up reports by mail (R+) and 600 to go without these reports (R−). A PhD student (JY) retrieved relevant results from the CMS and entered them into the JADE portal to create a personalized follow-up report for informing and empowering patients in the R+ group about their trends of attained treatment targets (HbA1c, BP, low-density lipoprotein cholesterol) and body weight, with individualized reminders and practical suggestions.3 After a median of 575 days (interquartile range [IQR] 519–646 days), HbA1c fell by −0.2% in the whole cohort, with greater reduction in the R+ (−0.24%) than R– group (−0.15%; P = 0.03; Table 1). Linear regression confirmed independent effect of follow-up reports on reducing HbA1c (β coefficient − 0.23; 95% confidence interval [CI] −0.39, −0.07; P = 0.004). In patients with chronic kidney disease (CKD), fewer patients in the R+ group had hospitalizations (42.3% vs 69.2%; P = 0.047; relative risk [RR] 0.50; 95% CI 0.28, 0.91; P = 0.021) and shorter length of stay (RR 0.49; 95% CI 0.23, 0.94; P = 0.023) than the R− group, after adjustment for confounders.

Given a relatively low baseline HbA1c of 7.4%, the effects of providing additional reports was modest, albeit significant, whereas their effects on hospitalization in high-risk patients requires further evaluation. In this IT era, apart from using structured processes to collect quality data, integration, analysis and communication of these data to promote informed decision making can further improve the quality of care.

Disclosure

The authors declare they have no conflict of interests.