Dataset Information

Use of generalised additive models to categorise continuous variables in clinical prediction.

ABSTRACT:

Background

In medical practice many, essentially continuous, clinical parameters tend to be categorised by physicians for ease of decision-making. Indeed, categorisation is a common practice both in medical research and in the development of clinical prediction rules, particularly where the ensuing models are to be applied in daily clinical practice to support clinicians in the decision-making process. Since the number of categories into which a continuous predictor must be categorised depends partly on the relationship between the predictor and the outcome, the need for more than two categories must be borne in mind.

Methods

We propose a categorisation methodology for clinical-prediction models, using Generalised Additive Models (GAMs) with P-spline smoothers to determine the relationship between the continuous predictor and the outcome. The proposed method consists of creating at least one average-risk category along with high- and low-risk categories based on the GAM smooth function. We applied this methodology to a prospective cohort of patients with exacerbated chronic obstructive pulmonary disease. The predictors selected were respiratory rate and partial pressure of carbon dioxide in the blood (PCO2), and the response variable was poor evolution. An additive logistic regression model was used to show the relationship between the covariates and the dichotomous response variable. The proposed categorisation was compared to the continuous predictor as the best option, using the AIC and AUC evaluation parameters. The sample was divided into a derivation (60%) and validation (40%) samples. The first was used to obtain the cut points while the second was used to validate the proposed methodology.

Results

The three-category proposal for the respiratory rate was ≤ 20;(20,24];> 24, for which the following values were obtained: AIC=314.5 and AUC=0.638. The respective values for the continuous predictor were AIC=317.1 and AUC=0.634, with no statistically significant differences being found between the two AUCs (p =0.079). The four-category proposal for PCO2 was ≤ 43;(43,52];(52,65];> 65, for which the following values were obtained: AIC=258.1 and AUC=0.81. No statistically significant differences were found between the AUC of the four-category option and that of the continuous predictor, which yielded an AIC of 250.3 and an AUC of 0.825 (p =0.115).

Conclusions

Our proposed method provides clinicians with the number and location of cut points for categorising variables, and performs as successfully as the original continuous predictor when it comes to developing clinical prediction rules.

SUBMITTER: Barrio I

PROVIDER: S-EPMC3716996 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectiveSpecific factors affecting generalizability of clinical prediction models are poorly understood. Our main objective was to investigate how measurement indicator variables affect external validity in clinical prediction models for predicting onset of vasopressor therapy.DesignWe fit logistic regressions on retrospective cohorts to predict vasopressor onset using two classes of variables: seemingly objective clinical variables (vital signs and laboratory measurements) and more subjective variables denoting recency of measurements.SettingThree cohorts from two tertiary-care academic hospitals in geographically distinct regions, spanning general inpatient and critical care settings.PatientsEach cohort consisted of adult patients (age greater than or equal to 18 yr at time of hospitalization), with lengths of stay between 6 and 600 hours, and who did not receive vasopressors in the first 6 hours of hospitalization or ICU admission. Models were developed on each of the three derivation cohorts and validated internally on the derivation cohort and externally on the other two cohorts.InterventionsNone.Measurements and main resultsThe prevalence of vasopressors was 0.9% in the general inpatient cohort and 12.4% and 11.5% in the two critical care cohorts. Models utilizing both classes of variables performed the best in-sample, with C-statistics for predicting vasopressor onset in 4 hours of 0.862 (95% CI, 0.844-0.879), 0.822 (95% CI, 0.793-0.852), and 0.889 (95% CI, 0.880-0.898). Models solely using the subjective variables denoting measurement recency had poor external validity. However, these practice-driven variables helped adjust for differences between the two hospitals and led to more generalizable models using clinical variables.ConclusionsWe developed and externally validated models for predicting the onset of vasopressors. We found that practice-specific features denoting measurement recency improved local performance and also led to more generalizable models if they are adjusted for during model development but discarded at validation. The role of practice-specific features such as measurement indicators in clinical prediction modeling should be carefully considered if the goal is to develop generalizable models.

Project description:Recent cohort studies have relied on exposure prediction models to estimate individuallevel air pollution concentrations because individual air pollution measurements are not available for cohort locations. For such prediction models, geographic variables related to pollution sources are important inputs. We demonstrated the computation process of geographic variables mostly recorded in 2010 at regulatory air pollution monitoring sites in South Korea. On the basis of previous studies, we finalized a list of 313 geographic variables related to air pollution sources in eight categories including traffic, demographic characteristics, land use, transportation facilities, physical geography, emissions, vegetation, and altitude. We then obtained data from different sources such as the Statistics Geographic Information Service and Korean Transport Database. After integrating all available data to a single database by matching coordinate systems and converting non-spatial data to spatial data, we computed geographic variables at 294 regulatory monitoring sites in South Korea. The data integration and variable computation were performed by using ArcGIS version 10.2 (ESRI Inc., Redlands, CA, USA). For traffic, we computed the distances to the nearest roads and the sums of road lengths within different sizes of circular buffers. In addition, we calculated the numbers of residents, households, housing buildings, companies, and employees within the buffers. The percentages of areas for different types of land use compared to total areas were calculated within the buffers. For transportation facilities and physical geography, we computed the distances to the closest public transportation depots and the boundary lines. The vegetation index and altitude were estimated at a given location by using satellite data. The summary statistics of geographic variables in Seoul across monitoring sites showed different patterns between urban background and urban roadside sites. This study provided practical knowledge on the computation process of geographic variables in South Korea, which will improve air pollution prediction models and contribute to subsequent health analyses.