Health utilities for chronic low back pain

BACKGROUND Chronic low back pain (LBP) is a common health problem, with a large potential for primary prevention. Health utilities (HU) reflect which proportion of their expected remaining life time individuals would hypothetically trade to be alleviated of a health condition of interest. A value of 0 means "prefer to die immediately", a value of 1 means "not willing to trade any life time". The aim of this cross-sectional study was to assess HU for LBP patients and for healthy participants and to examine whether HU for LBP are useful indicators to substantiate preventive and therapeutic decision making. METHODS Healthy participants (n = 126) and LBP patients (n = 32) were recruited mainly among the employees of a tertiary care hospital in Germany. Standardized LBP scenarios were presented to all participants and HU values were assessed using the time-trade-off method. RESULTS Median HU for LBP were 0.90 (IQR 0.31) for participants and 0.93 (IQR 0.10) for LBP patients. Measurements were consistent across illness severity ratings with HU and with a visual analogue scale (VAS); in the healthy sample the intraclass correlation coefficient (ICC) was 0.61 (95% CI 0.23-1.00, F(1125) = 190, p < .001), in the patient sample the ICC was 0.66 (95% CI = 0.24-1.00, F(1,31) = 62, p < .001). 8% of participants reported HU of 1. There was no statistically significant relation between HU and age, income, or gender. CONCLUSION On average, participants chose a 7 to 10% shorter life expectancy to avoid LBP, but almost 1 in 10 participants were not willing to trade any life years. The results indicate a certain stability of HU due to the comparability of HU ratings across patients and healthy participants, the measurement consistency when comparing VAS and HU ratings, and the lack of association between demographic variables and HU. This underlines the usefulness of HU for measuring illness severity in comparative health economics evaluations of preventive and therapeutic measures that address chronic LBP or other pain-characterized diseases. Future studies should focus on different LBP intensities and derive stratified HU that reflect the distribution of pain intensity in the population.