Estimating utility data from clinical indicators for patients with stable angina

Abstract

This study estimated a model from which data routinely collected in clinical trials of angina patients can be mapped to a utility scale and used to estimate quality-adjusted life years (QALYs). Patients with stable angina attending four cardiac out-patient clinics in the UK were included in the study. Data collected included information on patients' health-related quality of life (HRQL) using the EQ-5D, and severity of angina symptoms using two cardiac-specific measures [Breathlessness Grade and Canadian Cardiovascular Society (CCS) classification of angina]. Regression analysis was used to predict EQ-5D index values from the data. Data were obtained from 510 patients. For CCS grades, mean EQ-5D scores ranged from 0.36 (95% confidence interval 0.25-0.48) for grade 4 to 0.81 (0.77-0.85) for grade 0, and for breathlessness grades, EQ-5D scores ranged from 0.31 (0.06-0.55) for grade 0 to 0.84 (0.79-0.88) for grade 5. The final model used data on CCS grades, breathlessness grades, and patients' current medications to predict EQ-5D scores. The model had an R2 value of 0.37, and predictions for less severe angina were considered more reliable than the estimates for severe angina. In the absence of utility data collected as part of a clinical trial it is possible to map HRQL utility data from samples of patients with similar characteristics to those in the original trial. The uncertainty surrounding the estimates should be considered when using the results to estimate QALYs for purposes of economic evaluation.

Ivabradine in combination with beta-blocker improves symptoms and quality of life in patients with stable angina pectoris: results from the ADDITIONS study

“…An advantage of using generic measures of quality of life over disease-specific instruments is therefore that they can also be used for socioeconomic evaluation [43,45]. Of course, it might make sense to combine generic and disease-specific instruments, but this seemed to be beyond the scope of the study presented here [26].…”

“…It correlated with the measures of disease severity like CCS grade or commonly measured outcome variables [42,43,45].…”

Predicting EQ-5D Values Using the SGRQ

“…Although previous studies have mapped disease-specific measures to the EQ-5D [3][4][5][6][7][8][9], the extent to which these algorithms are validated varies and is often limited. This study used a rigorous approach to the development and validation of a prediction model using a data set in which both the EQ-5D and the SGRQ were routinely administered in COPD patients.…”

A taxonomy of health utility assessment methods and the role for uncertainty analysis

“…For ex ample, a map ping func tion was iden ti fied in a study that used the EQ-5D to pre dict utili ties from an gi na clin i cal mea sures. The cor re la tion be tween the pre dict ed and observed util i ty scores was 37, and the function was less re li able among se vere an gi na pa tients [11]. These ex am ples high light the de gree of un cer tain ty that may be as so ci ated with util i ty scores gained through in direct and map ping meth ods.…”

“…Ta ble 1). These fall into three main cat e gories: (a) direct val u a tion meth ods that di rect ly val ue all health states used in the anal y sis [1, 2, 3, 4]; (b) in di rect val u a tion meth ods that value a lim it ed num ber of health states di rectly and use a scor ing al go rithm to pre dict all oth er pos si ble health states that were not di rect ly val ued [5,6,7,8,9,10]; and (c) map ping meth ods that, in stead of valu ing health states di rect ly, ob serve/es tab lish a rela tion ship be tween an al ready ex ist ing utili ty mea sure and an oth er clin i cal or qual i ty of life (QOL) mea sure, and ex trap o late the re la tion ship to oth er stud ies [11].…”