Christiaan Heij

Modeling procedure and surgical times for current procedural terminology-anesthesia-surgeon combinations and evaluation in terms of case-duration prediction and operating room efficiency: a multicenter study.

BACKGROUND: Gains in operating room (OR) scheduling may be obtained by using accurate statistical models to predict surgical and procedure times. The 3 main contributions of this article are the following: (i) the validation of Strum’s results on the statistical distribution of case durations, including surgeon effects, using OR databases of 2 European hospitals, (ii) the use of expert prior expectations to predict durations of rarely observed cases, and (iii) the application of the proposed methods to predict case durations, with an analysis of the resulting increase in OR efficiency. METHODS: We retrospectively reviewed all recorded surgical cases of 2 large European teaching hospitals from 2005 to 2008, involving 85,312 cases and 92,099 h in total. Surgical times tended to be skewed and bounded by some minimally required time. We compared the fit of the normal distribution with that of 2- and 3-parameter lognormal distributions for case durations of a range of Current Procedural Terminology (CPT)-anesthesia combinations, including possible surgeon effects. For cases with very few observations, we investigated whether supplementing the data information with surgeons’ prior guesses helps to obtain better duration estimates. Finally, we used best fitting duration distributions to simulate the potential efficiency gains in OR scheduling. RESULTS: The 3-parameter lognormal distribution provides the best results for the case durations of CPT-anesthesia (surgeon) combinations, with an acceptable fit for almost 90% of the CPTs when segmented by the factor surgeon. The fit is best for surgical times and somewhat less for total procedure times. Surgeons’ prior guesses are helpful for OR management to improve duration estimates of CPTs with very few (<10) observations. Compared with the standard way of case scheduling using the mean of the 3-parameter lognormal distribution for case scheduling reduces the mean overreserved OR time per case up to 11.9 (11.8–12.0) min (55.6%) and the mean underreserved OR time per case up to 16.7 (16.5–16.8) min (53.1%). When scheduling cases using the 4-parameter lognormal model the mean overutilized OR time is up to 20.0 (19.7–20.3) min per OR per day lower than for the standard method and 11.6 (11.3–12.0) min per OR per day lower as compared with the biased corrected mean. CONCLUSIONS: OR case scheduling can be improved by using the 3-parameter lognormal model with surgeon effects and by using surgeons’ prior guesses for rarely observed CPTs. Using the 3-parameter lognormal model for case-duration prediction and scheduling significantly reduces both the prediction error and OR inefficiency.

Global total least squares modeling of multivariable time series

Presents a novel approach for the modeling of multivariable time series. The model class consists of linear systems, i.e., the solution sets of linear difference equations. Restricting the model order, the aim is to determine a model with minimal l/sub 2/-distance from the observed time series. Necessary conditions for optimality are described in terms of state-space representations. These conditions motivate a relatively simple iterative algorithm for the nonlinear problem of identifying optimal models. Attractive aspects of the proposed method are that the model error is measured globally, it can be applied for multi-input, multi-output systems, and no prior distinction between inputs and outputs is required. The authors give an illustration by means of some numerical simulations. >

Working With a Fixed Operating Room Team on Consecutive Similar Cases and the Effect on Case Duration and Turnover Time

HYPOTHESIS If variation in procedure times could be controlled or better predicted, the cost of surgeries could be reduced through improved scheduling of surgical resources. This study on the impact of similar consecutive cases on the turnover, surgical, and procedure times tests the perception that repeating the same manual tasks reduces the duration of these tasks. We hypothesize that when a fixed team works on similar consecutive cases the result will be shorter turnover and procedure duration as well as less variation as compared with the situation without a fixed team. DESIGN Case-control study. SETTING St Franciscus Hospital, a large general teaching hospital in Rotterdam, the Netherlands. PATIENTS Two procedures, inguinal hernia repair and laparoscopic cholecystectomy, were selected and divided across a control group and a study group. Patients were randomly assigned to the study or control group. MAIN OUTCOME MEASURES Preparation time, surgical time, procedure time, and turnover time. RESULTS For inguinal hernia repair, we found a significantly lower preparation time and 10 minutes less procedure time in the study group, as compared with the control group. Variation in the study group was lower, as compared with the control group. For laparoscopic cholecystectomy, preparation time was significantly lower in the study group, as compared with the control group. For both procedures, there was a significant decrease in turnover time. CONCLUSIONS Scheduling similar consecutive cases and performing with a fixed team results in lower turnover times and preparation times. The procedure time of the inguinal hernia repair decreased significantly and has practical scheduling implications. For more complex surgery, like laparoscopic cholecystectomy, there is no effect on procedure time.

Consistency of system identification by global total least squares

Global total least squares (GTLS) is a method for the identification of linear systems where no distinction between input and output variables is required. This method has been developed within the deterministic behavioural approach to systems. In this paper we analyse statistical properties of this method when the observations are generated by a multivariable stationary stochastic process. In particular, sufficient conditions for the consistency of GTLS are derived. This means that, when the number of observations tends to infinity, the identified deterministic system converges to the system that provides an optimal appoximation of the data generating process. The two main results are the following. GTLS is consistent if a guaranteed stability bound can be given a priori. If this information is not available, then consistency is obtained if GTLS is applied to the observed data extended with zero values in past and future.

Bariatric surgery with operating room teams that stayed fixed during the day: a multicenter study analyzing the effects on patient outcomes, teamwork and safety climate, and procedure duration.

BACKGROUND: Bariatric surgery durations vary considerably because of differences in surgical procedures and patient factors. We studied the effects on patient outcomes, teamwork and safety climate, and procedure durations resulting from working with operating room (OR) teams that remain fixed for the day instead of OR teams that vary during the day. METHODS: Data were collected in 2 general teaching hospitals, consisting of patientrelated demographic and intraoperative data and of staffrelated survey data on team work and safety climate. The procedure durations of fixed and conventional OR teams were analyzed by comparison of means tests and by regression methods to control for the effects of surgeon, surgical experience, and procedure type. RESULTS: For both hospitals, we obtained the following 4 results for working on bariatric procedures with OR teams that remained fixed for the day. First, patient outcomes did not worsen. Second, teamwork and safety climate (both measured on a 5-point scale) improved significantly, for teamwork + 0.86 (95% confidence interval [CI], 0.54 to 1.18) and for safety climate + 0.75 (95% CI, 0.40 to 1.11). Third, the procedures were performed significantly faster, as both the mean and the SD of procedure durations decreased. After correcting for learning effects, the average reduction of durations was 10.8% (99% CI, 5.0% to 15.3%, or 4 to 13 minutes). This gain was mainly realized for surgical time (12%; 99% CI, 5% to 18%, or 3 to 11 minutes). The effect on peripheral time, defined as procedure time minus surgical time, is not significant (3%; 99% CI, −6% to 12%, or −1 to 3 minutes). Fourth, additional gains were obtained by performing the same type of procedure multiple times within the same day (5% per every next procedure of the same type; 99% CI, 3% to 7%, or 3 to 6 minutes). CONCLUSIONS: Working with fixed teams in bariatric surgery reduced procedure durations and improved teamwork and safety climate, without adverse effects on patient outcomes.

Forecasting the Yield Curve in a Data‐Rich Environment Using the Factor‐Augmented Nelson–Siegel Model

Various ways of extracting macroeconomic information from a data-rich environment are compared with the objective of forecasting yield curves using the Nelson-Siegel model. Five issues in factor extraction are addressed, namely, selection of a subset of the available information, incorporation of the forecast objective in constructing factors, specification of a multivariate forecast objective, data grouping before constructing factors, and selection of the number of factors in a data-driven way. Our empirical results show that each of these features helps to improve forecast accuracy, especially for the shortest and longest maturities. The data-driven methods perform well in relatively volatile periods, when simpler models do not suffice.

Forecast comparison of principal component regression and principal covariate regression

Forecasting with many predictors is of interest, for instance, in macroeconomics and finance. The forecast accuracy of two methods for dealing with many predictors is compared, that is, principal component regression (PCR) and principal covariate regression (PCovR). Simulation experiments with data generated by factor models and regression models indicate that, in general, PCR performs better for the first type of data and PCovR performs better for the second type of data. An empirical application to four key US macroeconomic variables shows that PCovR achieves improved forecast accuracy in some situations.

Estimated incident cost savings in shipping due to inspections

The effectiveness of safety inspections of ships has been analysed from various angles, but until now, relatively little attention has been given to translate risk reduction into incident cost savings. This paper provides a monetary quantification of the cost savings that can be attributed to port state control inspections and industry vetting inspections. The dataset consists of more than half a million ship arrivals between 2002 and 2007 and contains inspections of port state authorities in the USA and Australia and of three industry vetting regimes. The effect of inspections in reducing the risk of total loss accidents is estimated by means of duration models, in terms of the gained probability of survival. The monetary benefit of port state control inspections is estimated to range, on average, from about 70 to 190 thousand dollars, with median values ranging from about 20 to 45 thousand dollars. Industry inspections have even higher benefits, especially for tankers. The savings are in general higher for older and larger vessels, and also for vessels with undefined flag and unknown classification society. As inspection costs are relatively low in comparison to potential cost savings, the results underline the importance of determining ships with relatively high risk of total loss.

Exact modelling and identifiability of linear systems

In this paper we consider the question of identifiability of a system on the basis of observed data. In particular we present a framework for exact modelling and identifiability of finite dimensional, linear, time invariant systems, on the basis of time series of infinite or finite length. Systems are characterized in terms of their behaviour, i.e. the set of time series which are allowed by the system. In this behavioural approach, no a priori distinction is made between input and output variables, and also non-controllable systems can be considered. We describe identification procedures for exact modelling of multivariable time series of infinite or finite length. These procedures are based on the concept of corroboration of models by data, reflecting that the data should incorporate evidence for the acceptability of the identified models. The assumed a priori information on the data generating system consists of the qualitative behavioural properties of linearity, time invariance, and completeness. No knowledge is required concerning quantitative properties like the number of input and output variables, the number of state variables, or the observability indices. A complete characterization is obtained of the minimal number of time series required for system identifiability. The class of systems identifiable by one time series contains the controllable systems as a strict subclass.

Nonlinear Forecasting with Many Predictors using Kernel Ridge Regression

This paper puts forward kernel ridge regression as an approach for forecasting with many predictors that are related nonlinearly to the target variable. In kernel ridge regression, the observed predictor variables are mapped nonlinearly into a high-dimensional space, where estimation of the predictive regression model is based on a shrinkage estimator to avoid overfitting. We extend the kernel ridge regression methodology to enable its use for economic time-series forecasting, by including lags of the dependent variable or other individual variables as predictors, as is typically desired in macroeconomic and financial applications. Monte Carlo simulations as well as an empirical application to various key measures of real economic activity confirm that kernel ridge regression can produce more accurate forecasts than traditional linear methods for dealing with many predictors based on principal component regression.