Errors and Bias in Using Predictive Scoring Systems

Cowen, Jay; Kelley, Mark A.

doi:10.1016/s0749-0704(18)30144-1

Cited by 62 publications

(27 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Scoring systems are successfully employed in several fields of medicine in order to assist physicians in drawing appropriate complex therapeutic decisions [20]. The choice between NACT and PDS certainly represents an hot topic in the management of AEOC, deeply influencing the natural history of newly diagnosed disease [21,22].…”

Section: Discussionmentioning

confidence: 99%

“…Staging LPS score has emerged over the last years as a reliable tool to correctly identify women suitable for optimal PDS [10,16,18]. However, it has to be considered that all predictive systems need to be continuously updated to keep in pace with therapeutic innovations, and technical advancements introduced in clinical practice [20]. For these reasons, the relevance of the present study, which provides an updated version of the previously developed LPS-PI is not to be underestimated.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Definition of a dynamic laparoscopic model for the prediction of incomplete cytoreduction in advanced epithelial ovarian cancer: Proof of a concept

Petrillo

Vizzielli

Fanfani

et al. 2015

Gynecologic Oncology

101

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Definition of a dynamic laparoscopic model for the prediction of incomplete cytoreduction in advanced epithelial ovarian cancer: Proof of a concept

Petrillo

Vizzielli

Fanfani

et al. 2015

Gynecologic Oncology

101

View full text Add to dashboard Cite

“…[63][64][65][66] Physicians could combine these trends with other important prognostic intreatment. Although an analysis that included only 117 individuals who met our selection criteria seems small, formation that is not included in these estimates.…”

Section: Daily Prognosis: Patients At 70% To 99% Daymentioning

confidence: 99%

“…27 In addition, the prognostic impact of cirrhosis and of ICU and other sources of bias. [66][67][68] In addition, the values of patients and families will have important influences admission diagnosis was also based on analysis of a larger group of admissions with cirrhosis (n Å 537) on decisions to either increase or reduce therapy.…”

Section: Daily Prognosis: Patients At 70% To 99% Daymentioning

confidence: 99%

Intensive care unit admissions with cirrhosis: Risk-stratifying patient groups and predicting individual survival

et al. 1996

View full text Add to dashboard Cite

Prognosis for acutely ill patients with cirrhosis is in-hepatic portosystemic shunting 3-7 and other operative fluenced by the severity of hepatic abnormalities and by procedures. [8][9][10][11] These systems have also been useful for dysfunction of other organ systems. The purpose of this risk-stratifying groups of patients with cirrhosis, [12][13][14][15] in study was to examine the usefulness of the Acute Physi-evaluating therapy for complications of cirrhosis, [16][17][18] ology, Age, and Chronic Health Evaluation (APACHE III) and in assessing the efficacy of procedures such as prognostic system for risk-stratifying groups of inten-sclerotherapy. [19][20][21] Although useful for clinical trials, sive care unit (ICU) patients with cirrhosis and in pre-the Child-Pugh classification has limited predictive acdicting individual survival. We used data for 17,440 ICU curacy for individuals because of overlapping criteria, admissions at 40 American hospitals to select 117 of the and interobserver variation for subjective criteria. 3,22 537 patients with a history of cirrhosis who were ventiIn addition, important prognostic factors unrelated to lated on ICU day 1, a group known to have a high mortality rate. We then calculated each patient's probability of hepatic function, e.g., measures of renal function, 23,24 hospital death on ICU days 1 through 7, using seven arterial pressure, 25 and study entry time, 26 are not inpreviously validated multivariate equations. Hospital cluded. Because of these limitations, we examined the mortality was 63% for the 117 study patients. The most usefulness of APACHE III (Acute Physiology, Age, and important determinants of risk for hospital death on Chronic Health Evaluation), a physiologically based ICU day 1 were the acute physiology score of APACHE prognostic system, 27 in assessing prognosis for groups III, ICU admission diagnosis, and operative status. Daily of intensive care unit (ICU) admissions with cirrhosis. changes in the acute physiology score caused a rise or ICU admission for patients with cirrhosis has prefall in the probability of hospital mortality and was useviously been recognized to result in low survival ful in assessing individual response to therapy. APACHE III accurately risk stratifies critically ill patients with rates. [28][29][30][31] Poor outcomes are common for all ICU pacirrhosis because it accounts for many of the factors tients with multiple organ system dysfunction or failknown to influence prognosis. This capability can be ure, 32-34 but survival is particularly poor among individused to assess severity of illness and risk-stratify pa-uals with cirrhosis who develop extrahepatic organ tients with cirrhosis during clinical trials. Daily prog-system failure. [35][36][37][38][39][40] For example, hospital mortality nostic estimates based on physiological changes over ranged from 89% to 95% in three studies involving 220 time reflect patient response and can help physicians to nonoperative ICU admissions with cirrhosis who also poor outcome results, we...

show abstract

“…Pre-admission treatment (lead-time bias), imprecise diagnosis (selection bias), and incomplete data (detection bias) may also confound the predictive ability of objective systems. 5 The paucity of critically ill obstetric patients in any population used to validate scoring systems precludes the routine use of such systems in this subset of patients.…”

mentioning

confidence: 99%