Journal of Clinical Epidemiology 56 (2003) 848–855 Validity of a modified standard gamble elicited from parents of a hospital-based cohort of children L. Sung a,b, * , M.L. Greenberg a , N.L. Young b , M. McLimont b , S. Ingber b , J. Rubenstein b , J. Wong b , T. Samanta b , J.J. Doyle a , A.M. Stain a , B.M. Feldman b,c a Division of Hematology/Oncology, Hospital for Sick Children, Toronto, Ontario, M5G 1X8 b Population Health Sciences, Hospital for Sick Children, Toronto, Ontario, M5G 1X8 c Division of Rheumatology, Hospital for Sick Children, Toronto, Ontario, M5G 1X8 Accepted 19 May 2003 Abstract Objectives: To examine the validity of a modified standard gamble (Mod SG) (nondeath baseline) by comparing these scores to SG (death baseline), time trade off (TTO), visual analog scale (VAS), Health Utilities Index (HUI), and Child Health Questionnaire (CHQ). Method: Respondents were parents of in-patients with cancer receiving chemotherapy and parents of children without cancer attending outpatient clinics. Construct validity was determined by comparing a priori hypotheses to actual correlations between measures. Discriminant validity was examined by anticipating that in-patients with cancer would have lower HRQL than outpatients. Results: 85 families were included. Both Mod SG and SG were moderately correlated with TTO (r = 0.50 and r = 0.49; P  .01 for both). Both Mod SG and SG were moderately correlated with TTO (r = 0.47 and r = 0.05, P  0.002 for both). Conclusion: The Mod SG did not perform better than SG. Two nonoverlapping groups of HRQL measures were demonstrated.  2003 Elsevier Inc. All rights reserved. Keywords: Health-related quality of life; Standard gamble; Visual analog scale; Utility; Validity; Children 1. Background The most commonly accepted method for measurement of utility is the standard gamble (SG) [1]; the SG is consid- ered by some to be the gold standard for utility elicitation. Others have questioned the validity of the SG related to a systematic dislike for risk among respondents or “gambling effect” [2–4], and indeed, several studies have found poor correlation between SG and other measures of health-related quality of life (HRQL) in adult respondents [5–8]. This “gambling effect” may be particularly pronounced when parents are proxy respondents for their children, an area of research that has received relatively little attention. A recent study of children with musculoskeletal disorders found that SG utilities were not related to other measures of HRQL, such as the Health Utilities Index (HUI) and categoric and analog rating scales [9]. The SG utilities were found to be * Corresponding author. Tel.: 416-813-5977; fax: 416-813-5327. E-mail address: Lillian.sung@sickkids.ca (L. Sung). 0895-4356/03/$ – see front matter  2003 Elsevier Inc. All rights reserved. doi: 10.1016/S0895-4356(03)00160-4 very high, even when the other ratings indicated poor health. The authors postulated that the parents were unwilling to “gamble” with death as a possible outcome, and that this risk aversion affected the validity of the SG. If the poor performance of the SG is related to risk aver- sion, and an unwillingness to “gamble” with death as a possible consequence, then one way to improve the SG might be to use a nondeath bottom anchor. In this modified SG (Mod SG) the respondent would choose between re- maining in the current health state or taking a lottery between perfect health and an undesirable but nondeath state. The resultant utility would then need to be recalibrated to the tra- ditional death baseline scale to make the results comparable to results obtained using the traditional SG. This method is often referred to as a chained utility elicitation, and is commonly used in the elicitation of preferences. Our objective was to determine whether a Mod SG was a more valid measure of parent assessed HRQL compared to the SG. We hypothesized that the Mod SG, when compared to SG, should be more similar to other measures of HRQL, and should be better able to discriminate between in-patient