Aim: To quantify and compare the performance of setting priority for intensive treatment by novel clustering-based subgroups to threshold-based classifications using systematic coronary risk evaluation (SCORE) and HbA1c levels.

Methods: We divided 2,935 patients from the Hoorn diabetes care system cohort into five clustering-based subgroups: severe insulin-deficient (SIDD) , severe insulin-resistant (SIRD) , mild obesity-related (MOD) mild (MD) , and mild with high HDL-cholesterol (MDH) diabetes, and four risk threshold-based subgroups. The United Kingdom Prospective Diabetes Study Outcomes Model was used to simulate lifetime health outcomes in the U.S. and U.K. settings. Gains from hypothetical treatment scenarios based on clinical guidelines were compared to “care-as-usual” and expressed in incremental quality-adjusted life-expectancy (QALE) and complication costs.

Results: The SIRD and MOD subgroup had the lowest absolute and age-sex-standardized QALE, respectively (7.90; 9.07) . Threshold-based classifications better discriminated between patients with higher and lower absolute and standardized QALE than clustering-based subgroups. For MOD, hypothetical interventions costing up to $1973 (95%CI $1444-$2603) and £463 (95%CI £345-£603) per year would be cost-effective at $100,000 and £20,000 per QALY thresholds in the U.S. and U.K., respectively. The MOD, SIDD and SIRD subgroups had the best potential cost-effectiveness alongside the subgroup with high HbA1c and high SCORE.

Conclusions: Intensified treatment could be cost-effective at higher-than-average costs for three out of five of the clustering-based subgroups and two out of four of the threshold-based classifications. Both classification methods support priority setting for intensive treatment, but the threshold-based method may better identify those who have the most potential to benefit from intensified treatment.


X.Li: None. A.Van giessen: None. J.Altunkaya: None. R.Slieker: None. J.Beulens: None. E.Pearson: Speaker's Bureau; Sanofi. P.J.M.Elders: None. T.Feenstra: Research Support; Dutch Healthcare Institute (ZIN) , European Union/IMI Rhapsody. J.Leal: None.


RHAPSODY (115881)

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