Introduction: There is increasing emphasis on patient reported outcomes and quality of life (QOL) measures in assessment of patients with Crohn's disease (CD). Descriptive QOL measures often require time-consuming questionnaires comprised of multiple QOL domains. Utility assessment is an attractive alternative measure because it provides a single global measure of overall QOL. The paper standard gamble (PSG) is a 1-page questionnaire in which patients are offered a hypothetical pill that confers a specific risk of either perfect health or death and has been shown to be a simple and reliable measure of patient utility. In this study we compared patients in different CD health states with respect to their generic health measure (EQ5D), health rating (EQ5D visual analog scale (VAS)), and utility (PSG). Methods: We prospectively surveyed adults with CD seen for outpatient gastroenterology clinic evaluation. Patients provided informed consent and filled out both the EQ5D and PSG. A gastroenterologist blinded to questionnaire results reviewed the clinic visit and classified the patient's health state based on symptoms (symptomatic or asymptomatic) and objective assessment (active disease, inactive disease, or no testing if none had been done within 6 months of the clinic visit). Results: 219 CD patients (mean age 41.9 years; range 18-77) were recruited. Disease duration was 16.5 + 12.9 years and 57% had prior CD abdominal surgery. The EQ5D index score, EQ5D VAS and the PSG all differentiated between asymptomatic and symptomatic patients but did not differentiate between inactive and active disease states (Tables 1 and 2). When comparing the most extreme of the four health states, asymptomatic inactive (“remission”) versus the symptomatic active groups, all 3 scales differentiated between states (Table 3). The Spearman rank correlation between the utility and other measures was weak. Conclusion: While symptomatic patients had lower health ratings and generic health measures, their utilities were not significantly different from asymptomatic patients. Disease activity was not associated with changes in any measure. Symptomatic patients with objective active inflammation appear willing to take a 7.3% greater risk of death in exchange for perfect health compared to patients with no symptoms and no active disease. Given the limited correlation between utility and other measures, the PSG, a simple one question utility, can be an important tool for clinical trials and QOL studies.689_A Figure 1. Correlation of symptoms and disease activity with EQ5D and PSG689_B Figure 2. Comparison of asymptomatic inactive patients to symptomatic active patients
The performance of prediction models can be assessed using a variety of methods and metrics. Traditional measures for binary and survival outcomes include the Brier score to indicate overall model performance, the concordance (or c) statistic for discriminative ability (or area under the receiver operating characteristic [ROC] curve), and goodness-of-fit statistics for calibration.Several new measures have recently been proposed that can be seen as refinements of discrimination measures, including variants of the c statistic for survival, reclassification tables, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Moreover, decision-analytic measures have been proposed, including decision curves to plot the net benefit achieved by making decisions based on model predictions.We aimed to define the role of these relatively novel approaches in the evaluation of the performance of prediction models. For illustration, we present a case study of predicting the presence of residual tumor versus benign tissue in patients with testicular cancer (n = 544 for model development, n = 273 for external validation).We suggest that reporting discrimination and calibration will always be important for a prediction model. Decision-analytic measures should be reported if the predictive model is to be used for clinical decisions. Other measures of performance may be warranted in specific applications, such as reclassification metrics to gain insight into the value of adding a novel predictor to an established model.
Over the past decade, there have been a rising number of clinically used tests that combine 2 or more biochemical or molecular assays, demographics, and clinical information into an algorithm to generate diagnostic, prognostic, or predictive information for a specific disease. The concept of multianalyte analyses is relatively new in the field of laboratory medicine. Dating back to the 1980s, prenatal screening for fetal abnormalities, such as Down syndrome, by use of maternal biomarkers is among the pioneer tests that use algorithmic analyses for risk assessment. Yet, the number of multianalyte algorithms used clinically remains modest. The American Medical Association provides current procedural terminology (CPT)8 codes for 20 multianalyte assays with algorithmic analyses (MAAAs.) Among these, 9 consist of biochemical markers detected by immunoassay or mass spectrometry, with or without other clinical information; 11 use molecular genetics markers; and 1 is for generic use. In this Q&A, we refer to multivariable tests with risk scores as MAAAs, although not all of them have an associated MAAA CPT code.
Generally speaking, MAAAs aim at improving diagnostics for diseases in which single biomarkers have limited clinical validity. The Prostate Health Index (Beckman Coulter) and the 4Kscore® (GenPath) exemplify some of these strategies for prostate cancer detection. Likewise, multianalyte analyses such as the Risk of Ovarian Malignancy Algorithm (ROMA®) and OVA1® and its second-generation OVERATM (Vermillion Inc.) improve upon the suboptimal performance of the tumor marker CA125 in the differential diagnosis and likelihood of ovarian carcinoma in women presenting with a pelvic mass. While both have Food and Drug Administration (FDA) clearance, ROMA is a nonproprietary algorithm cleared on various commercial immunoassays (Fujirebio Diagnostics, Inc.; Abbott Laboratories; Roche Diagnostics). Early identification of acute kidney injury is another area in which an FDA-approved multianalyte test, Nephrocheck® (Astute Medical), …
'/%3e%3c/defs%3e%3cpath fill='%23012169' d='M0 0h10080v5040H0z'/%3e%3cpath stroke='%23fff' stroke-width='.6' d='m0 0 6 3m0-3L0 3' clip-path='url(%23b)' transform='scale(840)'/%3e%3cpath stroke='%23e4002b' stroke-width='.4' d='m0 0 6 3m0-3L0 3' clip-path='url(%23c)' transform='scale(840)'/%3e%3cpath stroke='%23fff' stroke-width='840' d='M2520 0v2520M0 1260h5040'/%3e%3cpath stroke='%23e4002b' stroke-width='504' d='M2520 0v2520M0 1260h5040'/%3e%3cg fill='%23fff'%3e%3cuse xlink:href='%23d' x='2520' y='3780'/%3e%3cuse xlink:href='%23a' x='7560' y='4200'/%3e%3cuse xlink:href='%23a' x='6300' y='2205'/%3e%3cuse xlink:href='%23a' x='7560' y='840'/%3e%3cuse xlink:href='%23a' x='8680' y='1869'/%3e%3cuse xlink:href='%23e' x='8064' y='2730'/%3e%3c/g%3e%3c/svg%3e)
