Explore chapters and articles related to this topic
Healthcare Data Organization
Published in Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam, Introduction to Computational Health Informatics, 2019
Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam
There may be more than one disease code associated with mortality and morbidity. There are a general principle and three selection rules to derive the cause of mortality and morbidity. The general principle is: If there are more than one condition and the lowest condition can derive all other conditions, then the lowest condition is treated as the cause of mortality or morbidity. In the absence of general principle, the conditions are reported in the order of severity likely to cause mortality. There are three selection rules: 1) If there are more than one cause, then the originating cause is selected; 2) If no reported condition derives the first condition, the first condition is treated as the cause of mortality; 3) in the absence of Rules 1 and 2, if a condition Q is the consequence of another condition P, then the root-condition P is chosen. If a condition reported is a trivial condition unlikely to cause mortality, then more serious condition is chosen.
Methodological Issues In The Analysis Of Vital Statistics
Published in Michele Kiely, Reproductive and Perinatal Epidemiology, 2019
Cause of death is reported for all infant deaths. The procedure used to assign a cause of death can be summarized as follows: For a given death the underlying cause is selected from an array of conditions given in the cause-of-death section on the death certificate. These conditions are translated into medical codes through use of the classification structure and selection and modification rules contained in the applicable revision of the International Classification of Diseases (ICD) published by the World Health Organization (WHO). Selection rules provide guidance for systematically identifying the underlying cause of death in terms of the format of reported conditions and their causal relationship. Modification rules are intended to improve the usefulness of mortality statistics by giving preference to certain classification categories over others and/or to consolidate two or more conditions on the certificate into a single classification category.2
Birnbaum–Saunders sample selection model
Published in Journal of Applied Statistics, 2021
Fernando de Souza Bastos, Wagner Barreto-Souza
In this direction, Marchenko and Genton [23] introduced a sample selection model based on the bivariate Student-t distribution to associate the outcome with the selection rule. The advantage of this model is the robustness concerning the normal assumption. Disadvantages of this model are the difficulty to estimate the degrees of freedom (additional parameter to the Heckman model) and that the correlation equal to zero does not imply independence between the outcome and the selection rule (unless when the degrees of freedom goes to ∞, which corresponds to Heckman model). A Bayesian approach of this Student-t sample selection model was proposed by [10]. An approach based on copulas was proposed by [19] and semiparametric alternatives were suggested by [1], [31] and [28]. A non-parametric approach was studied by [9].
Risk assessment of post-discharge mortality among recently hospitalized Medicare heart failure patients with reduced or preserved ejection fraction
Published in Current Medical Research and Opinion, 2020
Mark Stampehl, Howard S. Friedman, Prakash Navaratnam, Patricia Russo, Siyeon Park, Engels N. Obi
The following default rules were implemented during the development of the CART model: selection criteria by chi-square; maximum of two branches per node; maximum depth of 5, minimum of 100 observations per leaf; and minimum of 250 observations required for search at each node. These pre-pruning specifications, along with the general process of data partitioning, were implemented to reduce the risk of overfitting. The CART output consisted of each patient being assigned to a final leaf and construction of a patient tree diagram. A sorted list of the variable importance (a measure of the influence that inclusion of that variable has in the output model) as defined in the SAS CART modeling program (SAS Enterprise Miner, Version 13.2) was also compiled. The output identified the key characteristics of the highest and lowest probability leaves including the selection rules of the leaf, the percentage of all patients assigned to the leaf and the observed probability in each leaf.
Measuring emotion perception following traumatic brain injury: The Complex Audio Visual Emotion Assessment Task (CAVEAT)
Published in Neuropsychological Rehabilitation, 2019
Hannah Rosenberg, Skye McDonald, Jacob Rosenberg, Reginald Frederick Westbrook
CAVEAT was developed in a number of stages. First, potential emotional terms were identified using Google, Google Scholar and Ovid search engines, with the keywords of “emotion”, “emotions”, and “feelings”. A list of 60 emotions was generated based on emotions that commonly repeated in the search. The identified emotions consisted to one of the three categories: basic emotions (Ekman & Friesen, 1976), revised basic emotions (Ekman, 1999; Ortony & Turner, 1990), and other, more complex emotions identified in the literature. The selection rules which guided emotion selection were, combining semantically or conceptually similar emotions into a single emotion category (e.g., revolted, nauseated, disgusted, were given a single label of “disgust”), and aiming at having a similar representation of positively and negatively valanced emotions. The selection rules were applied by the first author. This resulted in 34 emotions.