Objective: To study the mortality risk in a long-term population-based prospective study among subjects with prediabetes (PreDM) at baseline, compared to those with persistent normal glucose tolerance (NGT), taking into account whether they had had non-persistent PreDM, persistent PreDM, or converted to diabetes, during the follow-up period.

Methods: Subjects without history of diabetes had oral glucose tolerance test (OGTT) performed at each visit. Those who attended the baseline visit (1995-6) and returned for assessment at least once in 2000-4, 2005-8 and 2010-2 were included. PreDM was defined as impaired fasting glucose (≥ 5.6 mmol/L) or impaired glucose tolerance (2hour post-OGTT glucose ≥ 7.8 and <11.1 mmol/L). They were considered non-persistent PreDM if they reversed to NGT at least once, persistent if they remained PreDM throughout and converted if they progressed to diabetes. The NGT group included those attending at least 3 visits and had NGT persistently. Deaths from the Death Registry were adjudicated by record review. Mortality risk of PreDM subjects was compared to the NGT group.

Findings: 1225 subjects, 529 with PreDM (non-persistent, 41.9%; persistent, 23.3%; converted, 34.8%), were included for analysis. Factors independently associated with PreDM at baseline were age, body mass index, dyslipidemia and physical inactivity, as analysed by logistic regression. After 21.8 (IQR: 21.4-22.3) years, there were 101 deaths. Age, smoking and PreDM were the independent predictors of death identified using backward multivariable Cox regression. The non-persistent, persistent and converted groups all had higher mortality risk compared to NGT (Hazard Ratio: 2.33, 2.47, 2.43, all p<0.05).

Conclusions: We concluded that PreDM increased long-term mortality, regardless of the subsequent mode of glycemic progression. Measures for smoking control and prevention of PreDM through promotion of healthy diet, regular exercise and weight control, are recommended.


Y. Woo: None. C. Lee: None. H. Fong: None. M. Yeung: None. K.S.L. Lam: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.