Original ArticleThe standard gamble showed better construct validity than the time trade-off
Introduction
Preference-based instruments such as the standard gamble (SG), time trade-off (TTO), and rating scales are uniquely valuable for obtaining people's values for health states, and thus facilitating economic evaluation of health care interventions [1], [2]. Preference scores express the desirability that individuals exhibit for a specific health state and allow integration of impaired quality and quantity of life. In addition, preference-based instruments are increasingly used as global measures of health-related quality of life (HRQL), but evidence about their construct validity is still scarce [3], [4], [5].
There is a debate about the most appropriate choice of preference instrument. Because the SG meets theoretical econometric assumptions, including that of choice under uncertainty, some health economists favor it as the gold standard for measuring preferences [2]. Other investigators argue that the TTO is more consistent with individual preferences than the SG, less prone to bias [6], [7] and easier for patients to understand [8], [9], [10].
Current evidence suggests that the results of preference ratings from the SG are not interchangeable with the TTO. The SG yields systematically higher scores than the TTO [7]. Several studies showed that correlations between the SG and TTO are too low to allow for corrections or equations that would transform one score into another [8], [9], [11], [12], [13], [14], [15], [16].
To decide between instruments, investigators should have support for the validity of the instruments they use (i.e., the results represent what they intend to measure). Only a few studies compared the relative construct validity (i.e., correlation of preference scores with scores of other measures) of the SG and TTO [11], [13]. The aim of this study was to compare the cross-sectional construct validity of the SG and TTO in patients with Irritable Bowel Syndrome (IBS). IBS represents the most common disorder in gastroenterology practice [17] affecting over 15% of the general population [18], [19] and leading to substantial impairment in HRQL [20].
Section snippets
Study design and patients
For this analysis, we used data from a study that aimed at developing a disease-specific HRQL instrument for patients with IBS [21]. In brief, this was a cross-sectional study with one comprehensive assessment at either McMaster Health Sciences Centre, Hamilton, Canada or the patient's home. We identified patients through medical records of the gastroenterology clinic at McMaster Health Sciences Centre and through 10 local gastroenterologists.
We included patients between 18 and 75 years of age
Results
We enrolled 100 patients with IBS, of whom 96 completed all HRQL instruments. Table 1 shows their characteristics and the baseline scores of the utility and HRQL instruments. Most of the study participants were women around 40 years of age; almost 25% finished 8th grade or less. The SG scores were higher, with a smaller SD, compared to the TTO. Patients demonstrated widely varying scores on the validation measures indicating that the study population represented a broad sample of IBS patients.
Discussion
This comparison of the cross-sectional construct validity of the SG and TTO demonstrated higher correlations with generic and specific HRQL questionnaires as well as SF-36 derived preferences for the SG. The results suggest that in patients with IBS, the SG better reflects HRQL and patient preferences compared to the TTO.
Strengths of this study include the standardized administration of the utility instruments by a trained research assistant and the large number of validation measures. The
References (34)
- et al.
Limited reliability of marker states may explain lack of effect on measurement properties of utility instruments
J Clin Epidemiol
(2005) - et al.
Valuing health states: a comparison of methods
J Health Econ
(1996) - et al.
Variability among methods to assess patients' well-being and consequent effect on a cost-effectiveness analysis
J Clin Epidemiol
(1992) Diagnosis of irritable bowel syndrome
Gastroenterology
(2002)- et al.
Epidemiology of colonic symptoms and the irritable bowel syndrome
Gastroenterology
(1991) - et al.
The impact of irritable bowel syndrome on health-related quality of life
Gastroenterology
(2000) - et al.
Problems of HRQL assessment: how much is too much?
J Clin Epidemiol
(2001) - et al.
Deriving a preference-based single index from the UK SF-36 health survey
J Clin Epidemiol
(1998) - et al.
Measurement of agreement on health-related quality of life changes in response to respiratory rehabilitation by patients and physicians—a prospective study
Respir Med
(2004) Quality of life in epilepsy: comparison of four preference measures
Epilepsy Res
(1998)
Methodology for measuring health-state preferences–II: Scaling methods
J Clin Epidemiol
Measuring health state preferences and utilities: rating scale, time trade-off, and standard gamble techniques
Quality of life and pharmacoeconomics in clinical trials. Chapter 27
Methods for the economic evaluation of health care programmes
A clinical trial to evaluate the measurement properties of 2 direct preference instruments administered with and without hypothetical marker states
Med Decis Making
A randomized multicenter trial to evaluate simple utility elicitation techniques in patients with gastroesophageal reflux disease
Med Care
A new explanation for the difference between time trade-off utilities and standard gamble utilities
Health Econ
Acceptability of computerized Visual Analog Scale, time trade-off and standard gamble rating methods in patients and the public
Proc AMIA Symp
Cited by (27)
Cost-Effectiveness of Targeted Genetic Testing for Breast and Ovarian Cancer: A Systematic Review
2021, Value in HealthCitation Excerpt :In addition, the majority of studies combined utilities from a TTO approach with those obtained from Standard Gamble questionnaires. In addition to the usage of multiple sources in many studies combining these approaches may raise concerns because of the sometimes-diverging results.39 An additional challenge for researchers is in the choice of uptake rates of surgery, which have shown to be a sensitive factor in many studies.
Understanding Patient Acceptance of Risk with Treatment Options for Intermittent Claudication
2017, Annals of Vascular SurgeryCitation Excerpt :Time trade-off involves asking patients to quantify the number of years that they would be willing to sacrifice, to avoid worsening of symptoms or to try to obtain symptom improvement. Standard gamble has been shown to have better construct validity and reliability when compared with time trade-off18,19 and was more relevant for the purposes of this study, which aimed to understand the levels of risk that patients would accept with intervention. Microsoft Excel was used to collate the data and statistical analysis was done using SPSS.
Patient Utility Measurement for Managing Ureteral Stones: A Modified Standard Gamble Approach
2012, Value in Health Regional IssuesCitation Excerpt :Because the patient preference may vary with time, place, and ethnic group, it is worthwhile to measure patient preference over the choice of treatment modality with refined classifications to adapt the updated treatment modalities and preventive strategies of ureteral stone and identify its associated factors. Among three common direct eliciting methods—standard gamble [SG], time trade-off method, and visual analogue scale—for measuring the utility [4–12], each has strength and weakness in the aspects of practicality, reliability, and validity [4–14]. We prefer choice-based techniques (SG and time trade-off) to choice-less methods such as visual analogue scale [8,15].
Patients' perspective about risks and benefits of treatment for peripheral arterial disease
2011, Journal of Vascular and Interventional RadiologyCitation Excerpt :Responses to descriptive health status questions cannot reliably predict valuational QOL such as standard gamble utilities or time trade-off values (11,12). Standard gamble affords better construct validity and reliability of responses compared with time trade-off (16,17). Available treatment options for IC range from “stop smoking and keep walking” to angioplasty and stent placement to bypass surgery.
Common generic measures of health related quality of life in injured patients
2011, InjuryCitation Excerpt :Empiric methods used to determine utility in health economics include the Standard Gamble (SG), Time Tradeoff (TTO), and visual analogue scale (VAS). It is beyond the scope of this review to describe these instruments in depth, but it is worth noting that each metric has a theoretical basis for empiric use and has a body of literature establishing its validity and reliability.52,64 In contrast to these methods are indirect, preference-based questionnaires such as the HUI, Eq-5D, QWB-SA and SF-6D.67