Purpose:  Routine diabetes care has grown more complex as guidelines now emphasize intensive control of glucose, blood pressure, and cholesterol levels.  Cost-effectiveness analyses(CEA) of diabetes treatments have had incomplete data on preferences/utilities for many complication states and no data on treatment health states.  Our aim was to provide some of the first descriptions of utilities for intermediate microvascular complications, cardiovascular complications, and treatment states associated with diabetes. 

Methods:  We surveyed 456 adult patients with diabetes from clinics affiliated with a major academic medical center and a community hospital.  We elicited patients' utilities using time trade-off questions in face-to-face interviews.  We assessed utilities for intermediate and end-stage complication states and nine treatment states.  Our treatment states included a comprehensive diabetes treatment (polypharmacy) health state and a less complicated comprehensive treatment state incorporating the polypill, a hypothetical daily pill which combines multiple agents.   

Results: The mean age of subjects was 62 years (standard deviation(SD) 14), and 58% were men.  Patients were 40% African-American, 30% White, and 30% Latino.  The mean duration of diabetes was 10 years.  Mean complication utilities scores (SD) are as follows:

Angina	Mild Stroke	Major Stroke	Neuro-pathy	Ampu-tation	Retino-pathy	Blindness	Mild Kidney Disease	End-Stage Renal Disease
0.64 (0.30)	0.70 (0.30)	0.33 (0.30)	0.65 (0.33)	0.56 (0.34)	0.53 (0.35)	0.39 (0.34)	0.65 (0.33)	0.37 (0.33)

Intermediate complication states generated higher utilities than end-stage complications.

Below are the mean treatment preference scores:

Conven-tional Glucose Control	Inten-sive Glucose Control	Conven-tional BP Control	Inten-sive BP Control	Cholesterol medications	Aspirin	Poly-pharmacy	Polypill	Diet and Exercise
0.76 (0.30)	0.67 (0.33)	0.78 (0.28)	0.73 (0.30)	0.79 (0.28)	0.82 (0.27)	0.64 (0.33)	0.65 (0.33)	0.91 (0.20)

Conventional glucose and blood pressure control generated higher utilities than their intensive counterparts.  The polypharmacy state had the lowest utility of all treatment states, and the diet and exercise state had the highest.  The average utility for the polypill was no better than with polypharmacy.

Conclusions: Intermediate complications and conventional treatments generated higher utilities than their more severe/intensive counterparts.  Increasingly intensive treatments reduced utilities to the level of intermediate complication states.  All health states had significant variation between patients.  CEA of diabetes treatments need to incorporate utilities for previously ignored health states, acknowledge how treatment utilities vary with intensity, and address preference heterogeneity among patients.