Douglas Schaubel | University of Michigan (original) (raw)

Papers by Douglas Schaubel

The statistical evaluation of provider performance is an increasingly important component of the ... more The statistical evaluation of provider performance is an increasingly important component of the US healthcare system as policy-makers seek to shift reimbursement from rewarding quantity to rewarding quality and discouraging overuse. Statistically valid and reliable measures of provider performance are needed to guard against provider reactions that result in unintended consequences for patients. Both recent change to the Medicare dialysis reimbursement “bundle” payment for erythropoiesis-stimulating agent (ESA) drugs to manage anemia, and the identification of safety concerns associated with aggressive ESA use may result in increased frequency of red blood cell transfusion in the US chronic dialysis population. Thus, it is important to monitor dialysis facility-specific transfusion rates, relative to a national standard in order to identify specific facility treatment patterns that result in unnecessary blood transfusions. A risk adjusted transfusion measure using Medicare claims d...

Canadian Journal of Statistics, 2014

Liver transplantation : official publication of the American Association for the Study of Liver Diseases and the International Liver Transplantation Society, Jan 9, 2015

Survival benefit of simultaneous liver-kidney transplant (SLKT) over liver transplant alone (LTA)... more Survival benefit of simultaneous liver-kidney transplant (SLKT) over liver transplant alone (LTA) is unclear from the current literature. Additionally, the role of donor kidney quality, measured by kidney donor risk index (KDRI), in survival benefit of SLKT is not studied. We compared survival benefit after SLKT and LTA among recipients with similar pre-transplant renal dysfunction using novel methodology, specifically with respect to survival probability and area under the survival curve by dialysis status and KDRI. Data were obtained from the Scientific Registry of Transplant Recipients. The study cohort included patients with pre-LT renal dysfunction who were waitlisted and received either a SLKT (n=1,326) or a LTA (n=4,283) between 3/1/02-12/31/09. Inverse Probability of Treatment Weighted (IPTW) - SLKT and LTA survival curves, along with the 5-year area under the survival curve were computed by dialysis status at transplant. The difference in the area under the curve represents...

Transplantation, Jan 9, 2015

Simultaneous kidney and pancreas (SPK) transplantation is an attractive option for end-stage rena... more Simultaneous kidney and pancreas (SPK) transplantation is an attractive option for end-stage renal disease patients with type 1 diabetes. Although SPK transplantation is superior to remaining on dialysis, the survival advantage for SPK recipients compared to kidney transplantation alone (KTA) is controversial. Using data obtained from the Scientific Registry of Transplant Recipients, we compared patient and graft survivals for 7308 SPK and 4653 KTA adult patients with type I diabetes transplanted in 1998 to 2009. Because SPK and KTA recipients are differently selected, comparison groups were chosen to maximize overlap in the case mixes. Most previous studies contrasted (unadjusted) Kaplan-Meier survival curves or, if covariate-adjusted, reported hazard ratios (HRs). Using newer statistical methods, we avoid relying on hazard ratios (which are seldom of inherent interest) and directly compare covariate-adjusted survival curves. Specifically, we compare average covariate-adjusted SPK-...

American Journal of Transplantation

Statistics in Biosciences, 2012

In this article, we develop methods for quantifying center effects with respect to recurrent even... more In this article, we develop methods for quantifying center effects with respect to recurrent event data. In the models of interest, center effects are assumed to act multiplicatively on the recurrent event rate function. When the number of centers is large, traditional estimation methods that treat centers as categorical variables have many parameters and are sometimes not feasible to implement, especially with large numbers of distinct recurrent event times. We propose a new estimation method for center effects which avoids including indicator variables for centers. We then show that center effects can be consistently estimated by the center-specific ratio of observed to expected cumulative numbers of events. We also consider the case where the recurrent event sequence can be stopped permanently by a terminating event. Large sample results are developed for the proposed estimators. We assess the finite-sample properties of the proposed estimators through simulation studies. The method is then applied to national hospital admissions data for end stage renal disease patients.

Statistics in biosciences, 2014

In observational studies of survival time featuring a binary time-dependent treatment, the hazard... more In observational studies of survival time featuring a binary time-dependent treatment, the hazard ratio (an instantaneous measure) is often used to represent the treatment effect. However, investigators are often more interested in the difference in survival functions. We propose semiparametric methods to estimate the causal effect of treatment among the treated with respect to survival probability. The objective is to compare post-treatment survival with the survival function that would have been observed in the absence of treatment. For each patient, we compute a prognostic score (based on the pre-treatment death hazard) and a propensity score (based on the treatment hazard). Each treated patient is then matched with an alive, uncensored and not-yet-treated patient with similar prognostic and/or propensity scores. The experience of each treated and matched patient is weighted using a variant of Inverse Probability of Censoring Weighting to account for the impact of censoring. We p...

Liver Transplantation, 2009

More candidates with creatinine levels Ն 2 mg/dL have undergone liver transplantation (LT) since ... more More candidates with creatinine levels Ն 2 mg/dL have undergone liver transplantation (LT) since the implementation of Model for End-Stage Liver Disease (MELD)-based allocation. These candidates have higher posttransplant mortality. This study examined the effect of serum creatinine on survival benefit among candidates undergoing LT. Scientific Registry of Transplant Recipients data were analyzed for adult LT candidates listed between September 2001 and December 2006 (n ϭ 38,899). The effect of serum creatinine on survival benefit (contrast between waitlist and post-LT mortality rates) was assessed by sequential stratification, an extension of Cox regression. At the same MELD score, serum creatinine was inversely associated with survival benefit within certain defined MELD categories. The survival benefit significantly decreased as creatinine increased for candidates with MELD scores of 15 to 17 or 24 to 40 at LT (MELD scores of 15-17, P Ͻ 0.0001; MELD scores of 24-40, P ϭ 0.04). Renal replacement therapy at LT was also associated with significantly decreased LT benefit for patients with MELD scores of 21 to 23 (P ϭ 0.04) or 24 to 26 (P ϭ 0.01). In conclusion, serum creatinine at LT significantly affects survival benefit for patients with MELD scores of 15 to 17 or 24 to 40. Given the same MELD score, patients with higher creatinine levels receive less benefit on average, and the relative ranking of a large number of wait-listed candidates with MELD scores of 15 to 17 or 24 to 40 would be markedly affected if these findings were incorporated into the allocation policy.

Transplant International, 2011

Statistics in Medicine, 2014

We develop a weighted cumulative sum (WCUSUM) to evaluate and monitor pre-transplant waitlist mor... more We develop a weighted cumulative sum (WCUSUM) to evaluate and monitor pre-transplant waitlist mortality of facilities in the context where transplantation is considered to be dependent censoring. Waitlist patients are evaluated multiple times in order to update their current medical condition as reflected in a time-dependent variable called the Model for End-Stage Liver Disease (MELD) score. Higher MELD scores are indicative of higher pre-transplant death risk. Moreover, under the current liver allocation system, patients with higher MELD scores receive higher priority for liver transplantation. To evaluate the waitlist mortality of transplant centers, it is important to take this dependent censoring into consideration. We assume a 'standard' transplant practice through a transplant model and utilize inverse probability censoring weights to construct a WCUSUM. We evaluate the properties of a weighted zero-mean process as the basis of the proposed WCUSUM. We then discuss a resampling technique to obtain control limits. The proposed WCUSUM is illustrated through the analysis of national transplant registry data.

Statistics in Medicine, 2010

We consider using observational data to estimate the effect of a treatment on disease recurrence,... more We consider using observational data to estimate the effect of a treatment on disease recurrence, when the decision to initiate treatment is based on longitudinal factors associated with the risk of recurrence. The effect of salvage androgen deprivation therapy (SADT) on the risk of recurrence of prostate cancer is inadequately described by the existing literature. Furthermore, standard Cox regression yields biased estimates of the effect of SADT, since it is necessary to adjust for prostate-specific antigen (PSA), which is a time-dependent confounder and an intermediate variable. In this paper, we describe and compare two methods which appropriately adjust for PSA in estimating the effect of SADT. The first method is a two-stage method which jointly estimates the effect of SADT and the hazard of recurrence in the absence of treatment by SADT. In the first stage, PSA is predicted in the absence of SADT, and in the second stage, a time-dependent Cox model is used to estimate the benefit of SADT, adjusting for PSA. The second method, called sequential stratification, reorganizes the data to resemble a sequence of experiments in which treatment is conditionally randomized given the time-dependent covariates. Strata are formed, each consisting of a patient undergoing SADT and a set of appropriately matched controls, and analysis proceeds via stratified Cox regression. Both methods are applied to data from patients initially treated with radiation therapy for prostate cancer and give similar SADT effect estimates.

Kidney International, 2012

The risk of death for hemodialysis patients is thought to be highest on the days following the lo... more The risk of death for hemodialysis patients is thought to be highest on the days following the longest interval without dialysis (usually Mondays and Tuesdays); however, existing results are inconclusive. To clarify this we analyzed Dialysis Outcomes and Practice Patterns Study (DOPPS) data of 22,163 hemodialysis patients from the United States, Europe and Japan. Our study focused on the association between dialysis schedule and day-of-week of all-cause, cardiovascular and non-cardiovascular mortality with day-of-week coding as a time-dependent covariate. The models were adjusted for dialysis schedule, age, country, DOPPS Phase I or II, and other demographic and clinical covariates comparing mortality on each day to the 7-day average. Patients on a Monday-Wednesday-Friday (MFW) schedule had elevated all-cause mortality on Monday, and those on a Tuesday-Thursday-Saturday (TTS) schedule increased risk of mortality on Tuesday in all 3 regions. The association between day-of-week mortality and schedule was generally stronger for cardiovascular than non-cardiovascular mortality, and most pronounced in the United States. Unexpectedly, Japanese patients on a MWF schedule had a higher risk of non-cardiovascular mortality on Fridays, and European patients on a TTS schedule experienced an elevated cardiovascular mortality on Saturdays. Thus, future studies are needed to evaluate the influence of practice patterns on schedule-specific mortality and factors that could modulate this effect.

Kidney International, 2012

Journal of the American Society of Nephrology, 2009

Black renal transplant recipients experience shorter graft survival than white recipients, but no... more Black renal transplant recipients experience shorter graft survival than white recipients, but no published data describe the graft outcomes among black Canadian recipients. Here, we analyzed data from the Canadian national renal replacement therapy registry, which included 20,243 incident dialysis patients (3% black, 97% white), 5036 of whom received a renal transplant during the study period. Black patients were significantly less likely to receive a renal transplant (deceased and living-donor combined) when compared with white patients (hazard ratio 0.59; 95% confidence interval 0.51 to 0.69; P Ͻ 0.0001). Among patients who underwent a renal transplant, there was no significant difference in the likelihood of graft failure between black and white patients, even after adjustment for comorbidities and socioeconomic status; black patients, however, had significantly lower posttransplantation mortality compared with white patients (hazard ratio 0.49; 95% confidence interval 0.28 to 0.88; P ϭ 0.02). In conclusion, graft outcomes between black and white Canadian renal transplant patients are similar. Because this differs from the experience reported from the United States, further direct comparisons between the two populations is warranted.

Hepatology, 2012

Background-Candidates with fulminant hepatic failure (Status-1A) receive the highest priority for... more Background-Candidates with fulminant hepatic failure (Status-1A) receive the highest priority for liver transplantation (LT) in the United States. However, no studies have compared wait-list mortality risk among end-stage liver disease (ESLD) candidates with high Model for End-stage Liver Disease (MELD) scores to those listed as Status-1A. We aimed to determine if there are MELD scores for ESLD candidates at which their wait-list mortality risk is higher than that of Status-1A, and to identify the factors predicting wait-list mortality among Status-1A.

Gastroenterology, 2008

Background & Aims: Liver transplant candidates with mild hepatic synthetic dysfunction and marked... more Background & Aims: Liver transplant candidates with mild hepatic synthetic dysfunction and marked renal insufficiency may have higher Model for End-Stage Liver Disease (MELD) scores than candidates with severe liver disease and normal renal function. We re-estimated MELD coefficients and evaluated the effect of updated MELD on the liver transplant waiting list ranking. Methods: Scientific Registry of Transplant Recipients data was analyzed for 38,899 adults wait-listed between September, 2001 and December, 2006. A time-dependent Cox regression waiting list mortality model estimated updated MELD component coefficients. Rank correlation between existing and updated MELD scores was computed. Results: Existing MELD component coefficient (log e creatinine, 0.957 vs 1.266 [95% confidence interval (CI), 1.21-1.32]; log e bilirubin, 0.378 vs 0.939 [95% CI, 0.91-0.97]

Canadian Journal of Statistics, 2006

Key wonis and phrases: Improper imputation; missing data; multivariate failure time data; proport... more Key wonis and phrases: Improper imputation; missing data; multivariate failure time data; proportional rates model; semiparametric model. MSC 2OOO: Primary 62N99; secondary 62P10.

The statistical evaluation of provider performance is an increasingly important component of the ... more The statistical evaluation of provider performance is an increasingly important component of the US healthcare system as policy-makers seek to shift reimbursement from rewarding quantity to rewarding quality and discouraging overuse. Statistically valid and reliable measures of provider performance are needed to guard against provider reactions that result in unintended consequences for patients. Both recent change to the Medicare dialysis reimbursement “bundle” payment for erythropoiesis-stimulating agent (ESA) drugs to manage anemia, and the identification of safety concerns associated with aggressive ESA use may result in increased frequency of red blood cell transfusion in the US chronic dialysis population. Thus, it is important to monitor dialysis facility-specific transfusion rates, relative to a national standard in order to identify specific facility treatment patterns that result in unnecessary blood transfusions. A risk adjusted transfusion measure using Medicare claims d...

Canadian Journal of Statistics, 2014

Liver transplantation : official publication of the American Association for the Study of Liver Diseases and the International Liver Transplantation Society, Jan 9, 2015

Survival benefit of simultaneous liver-kidney transplant (SLKT) over liver transplant alone (LTA)... more Survival benefit of simultaneous liver-kidney transplant (SLKT) over liver transplant alone (LTA) is unclear from the current literature. Additionally, the role of donor kidney quality, measured by kidney donor risk index (KDRI), in survival benefit of SLKT is not studied. We compared survival benefit after SLKT and LTA among recipients with similar pre-transplant renal dysfunction using novel methodology, specifically with respect to survival probability and area under the survival curve by dialysis status and KDRI. Data were obtained from the Scientific Registry of Transplant Recipients. The study cohort included patients with pre-LT renal dysfunction who were waitlisted and received either a SLKT (n=1,326) or a LTA (n=4,283) between 3/1/02-12/31/09. Inverse Probability of Treatment Weighted (IPTW) - SLKT and LTA survival curves, along with the 5-year area under the survival curve were computed by dialysis status at transplant. The difference in the area under the curve represents...

Transplantation, Jan 9, 2015

Simultaneous kidney and pancreas (SPK) transplantation is an attractive option for end-stage rena... more Simultaneous kidney and pancreas (SPK) transplantation is an attractive option for end-stage renal disease patients with type 1 diabetes. Although SPK transplantation is superior to remaining on dialysis, the survival advantage for SPK recipients compared to kidney transplantation alone (KTA) is controversial. Using data obtained from the Scientific Registry of Transplant Recipients, we compared patient and graft survivals for 7308 SPK and 4653 KTA adult patients with type I diabetes transplanted in 1998 to 2009. Because SPK and KTA recipients are differently selected, comparison groups were chosen to maximize overlap in the case mixes. Most previous studies contrasted (unadjusted) Kaplan-Meier survival curves or, if covariate-adjusted, reported hazard ratios (HRs). Using newer statistical methods, we avoid relying on hazard ratios (which are seldom of inherent interest) and directly compare covariate-adjusted survival curves. Specifically, we compare average covariate-adjusted SPK-...

American Journal of Transplantation

Statistics in Biosciences, 2012

In this article, we develop methods for quantifying center effects with respect to recurrent even... more In this article, we develop methods for quantifying center effects with respect to recurrent event data. In the models of interest, center effects are assumed to act multiplicatively on the recurrent event rate function. When the number of centers is large, traditional estimation methods that treat centers as categorical variables have many parameters and are sometimes not feasible to implement, especially with large numbers of distinct recurrent event times. We propose a new estimation method for center effects which avoids including indicator variables for centers. We then show that center effects can be consistently estimated by the center-specific ratio of observed to expected cumulative numbers of events. We also consider the case where the recurrent event sequence can be stopped permanently by a terminating event. Large sample results are developed for the proposed estimators. We assess the finite-sample properties of the proposed estimators through simulation studies. The method is then applied to national hospital admissions data for end stage renal disease patients.

Statistics in biosciences, 2014

In observational studies of survival time featuring a binary time-dependent treatment, the hazard... more In observational studies of survival time featuring a binary time-dependent treatment, the hazard ratio (an instantaneous measure) is often used to represent the treatment effect. However, investigators are often more interested in the difference in survival functions. We propose semiparametric methods to estimate the causal effect of treatment among the treated with respect to survival probability. The objective is to compare post-treatment survival with the survival function that would have been observed in the absence of treatment. For each patient, we compute a prognostic score (based on the pre-treatment death hazard) and a propensity score (based on the treatment hazard). Each treated patient is then matched with an alive, uncensored and not-yet-treated patient with similar prognostic and/or propensity scores. The experience of each treated and matched patient is weighted using a variant of Inverse Probability of Censoring Weighting to account for the impact of censoring. We p...

Liver Transplantation, 2009

More candidates with creatinine levels Ն 2 mg/dL have undergone liver transplantation (LT) since ... more More candidates with creatinine levels Ն 2 mg/dL have undergone liver transplantation (LT) since the implementation of Model for End-Stage Liver Disease (MELD)-based allocation. These candidates have higher posttransplant mortality. This study examined the effect of serum creatinine on survival benefit among candidates undergoing LT. Scientific Registry of Transplant Recipients data were analyzed for adult LT candidates listed between September 2001 and December 2006 (n ϭ 38,899). The effect of serum creatinine on survival benefit (contrast between waitlist and post-LT mortality rates) was assessed by sequential stratification, an extension of Cox regression. At the same MELD score, serum creatinine was inversely associated with survival benefit within certain defined MELD categories. The survival benefit significantly decreased as creatinine increased for candidates with MELD scores of 15 to 17 or 24 to 40 at LT (MELD scores of 15-17, P Ͻ 0.0001; MELD scores of 24-40, P ϭ 0.04). Renal replacement therapy at LT was also associated with significantly decreased LT benefit for patients with MELD scores of 21 to 23 (P ϭ 0.04) or 24 to 26 (P ϭ 0.01). In conclusion, serum creatinine at LT significantly affects survival benefit for patients with MELD scores of 15 to 17 or 24 to 40. Given the same MELD score, patients with higher creatinine levels receive less benefit on average, and the relative ranking of a large number of wait-listed candidates with MELD scores of 15 to 17 or 24 to 40 would be markedly affected if these findings were incorporated into the allocation policy.

Transplant International, 2011

Statistics in Medicine, 2014

We develop a weighted cumulative sum (WCUSUM) to evaluate and monitor pre-transplant waitlist mor... more We develop a weighted cumulative sum (WCUSUM) to evaluate and monitor pre-transplant waitlist mortality of facilities in the context where transplantation is considered to be dependent censoring. Waitlist patients are evaluated multiple times in order to update their current medical condition as reflected in a time-dependent variable called the Model for End-Stage Liver Disease (MELD) score. Higher MELD scores are indicative of higher pre-transplant death risk. Moreover, under the current liver allocation system, patients with higher MELD scores receive higher priority for liver transplantation. To evaluate the waitlist mortality of transplant centers, it is important to take this dependent censoring into consideration. We assume a 'standard' transplant practice through a transplant model and utilize inverse probability censoring weights to construct a WCUSUM. We evaluate the properties of a weighted zero-mean process as the basis of the proposed WCUSUM. We then discuss a resampling technique to obtain control limits. The proposed WCUSUM is illustrated through the analysis of national transplant registry data.

Statistics in Medicine, 2010

We consider using observational data to estimate the effect of a treatment on disease recurrence,... more We consider using observational data to estimate the effect of a treatment on disease recurrence, when the decision to initiate treatment is based on longitudinal factors associated with the risk of recurrence. The effect of salvage androgen deprivation therapy (SADT) on the risk of recurrence of prostate cancer is inadequately described by the existing literature. Furthermore, standard Cox regression yields biased estimates of the effect of SADT, since it is necessary to adjust for prostate-specific antigen (PSA), which is a time-dependent confounder and an intermediate variable. In this paper, we describe and compare two methods which appropriately adjust for PSA in estimating the effect of SADT. The first method is a two-stage method which jointly estimates the effect of SADT and the hazard of recurrence in the absence of treatment by SADT. In the first stage, PSA is predicted in the absence of SADT, and in the second stage, a time-dependent Cox model is used to estimate the benefit of SADT, adjusting for PSA. The second method, called sequential stratification, reorganizes the data to resemble a sequence of experiments in which treatment is conditionally randomized given the time-dependent covariates. Strata are formed, each consisting of a patient undergoing SADT and a set of appropriately matched controls, and analysis proceeds via stratified Cox regression. Both methods are applied to data from patients initially treated with radiation therapy for prostate cancer and give similar SADT effect estimates.

Kidney International, 2012

The risk of death for hemodialysis patients is thought to be highest on the days following the lo... more The risk of death for hemodialysis patients is thought to be highest on the days following the longest interval without dialysis (usually Mondays and Tuesdays); however, existing results are inconclusive. To clarify this we analyzed Dialysis Outcomes and Practice Patterns Study (DOPPS) data of 22,163 hemodialysis patients from the United States, Europe and Japan. Our study focused on the association between dialysis schedule and day-of-week of all-cause, cardiovascular and non-cardiovascular mortality with day-of-week coding as a time-dependent covariate. The models were adjusted for dialysis schedule, age, country, DOPPS Phase I or II, and other demographic and clinical covariates comparing mortality on each day to the 7-day average. Patients on a Monday-Wednesday-Friday (MFW) schedule had elevated all-cause mortality on Monday, and those on a Tuesday-Thursday-Saturday (TTS) schedule increased risk of mortality on Tuesday in all 3 regions. The association between day-of-week mortality and schedule was generally stronger for cardiovascular than non-cardiovascular mortality, and most pronounced in the United States. Unexpectedly, Japanese patients on a MWF schedule had a higher risk of non-cardiovascular mortality on Fridays, and European patients on a TTS schedule experienced an elevated cardiovascular mortality on Saturdays. Thus, future studies are needed to evaluate the influence of practice patterns on schedule-specific mortality and factors that could modulate this effect.

Kidney International, 2012

Journal of the American Society of Nephrology, 2009

Black renal transplant recipients experience shorter graft survival than white recipients, but no... more Black renal transplant recipients experience shorter graft survival than white recipients, but no published data describe the graft outcomes among black Canadian recipients. Here, we analyzed data from the Canadian national renal replacement therapy registry, which included 20,243 incident dialysis patients (3% black, 97% white), 5036 of whom received a renal transplant during the study period. Black patients were significantly less likely to receive a renal transplant (deceased and living-donor combined) when compared with white patients (hazard ratio 0.59; 95% confidence interval 0.51 to 0.69; P Ͻ 0.0001). Among patients who underwent a renal transplant, there was no significant difference in the likelihood of graft failure between black and white patients, even after adjustment for comorbidities and socioeconomic status; black patients, however, had significantly lower posttransplantation mortality compared with white patients (hazard ratio 0.49; 95% confidence interval 0.28 to 0.88; P ϭ 0.02). In conclusion, graft outcomes between black and white Canadian renal transplant patients are similar. Because this differs from the experience reported from the United States, further direct comparisons between the two populations is warranted.

Hepatology, 2012

Background-Candidates with fulminant hepatic failure (Status-1A) receive the highest priority for... more Background-Candidates with fulminant hepatic failure (Status-1A) receive the highest priority for liver transplantation (LT) in the United States. However, no studies have compared wait-list mortality risk among end-stage liver disease (ESLD) candidates with high Model for End-stage Liver Disease (MELD) scores to those listed as Status-1A. We aimed to determine if there are MELD scores for ESLD candidates at which their wait-list mortality risk is higher than that of Status-1A, and to identify the factors predicting wait-list mortality among Status-1A.

Gastroenterology, 2008

Background & Aims: Liver transplant candidates with mild hepatic synthetic dysfunction and marked... more Background & Aims: Liver transplant candidates with mild hepatic synthetic dysfunction and marked renal insufficiency may have higher Model for End-Stage Liver Disease (MELD) scores than candidates with severe liver disease and normal renal function. We re-estimated MELD coefficients and evaluated the effect of updated MELD on the liver transplant waiting list ranking. Methods: Scientific Registry of Transplant Recipients data was analyzed for 38,899 adults wait-listed between September, 2001 and December, 2006. A time-dependent Cox regression waiting list mortality model estimated updated MELD component coefficients. Rank correlation between existing and updated MELD scores was computed. Results: Existing MELD component coefficient (log e creatinine, 0.957 vs 1.266 [95% confidence interval (CI), 1.21-1.32]; log e bilirubin, 0.378 vs 0.939 [95% CI, 0.91-0.97]

Canadian Journal of Statistics, 2006

Key wonis and phrases: Improper imputation; missing data; multivariate failure time data; proport... more Key wonis and phrases: Improper imputation; missing data; multivariate failure time data; proportional rates model; semiparametric model. MSC 2OOO: Primary 62N99; secondary 62P10.