Using secondary knowledge to support decision tree classification of retrospective clinical data

Retrospective clinical data presents many challenges for data mining and machine learning. The transcription of patient records from paper charts and subsequent manipulation of data often results in high volumes of noise as well as a loss of other important information. In addition, such datasets often fail to represent expert medical knowledge and reasoning in any explicit manner. In this research we describe applying data mining methods to retrospective clinical data to build a prediction model for asthma exacerbation severity for pediatric patients in the emergency department. Difficulties in building such a model forced us to investigate alternative strategies for analyzing and processing retrospective data. This paper describes this process together with an approach to mining retrospective clinical data by incorporating formalized external expert knowledge (secondary knowledge sources) into the classification task. This knowledge is used to partition the data into a number of coherent sets, where each set is explicitly described in terms of the secondary knowledge source. Instances from each set are then classified in a manner appropriate for the characteristics of the particular set. We present our methodology and outline a set of experiential results that demonstrate some advantages and some limitations of our approach. © 2008 Springer-Verlag Berlin Heidelberg.

Physical and psychophysical analysis of progressive addition lens embodiments

The extent to which the surface parameters of Progressive Addition Lenses (PALs) affect successful patient tolerance was investigated. Several optico-physical evaluation techniques were employed, including a newly constructed surface reflection device which was shown to be of value for assessing semi-finished PAL blanks. Detailed physical analysis was undertaken using a computer-controlled focimeter and from these data, iso-cylindrical and mean spherical plots were produced for each PAL studied. Base curve power was shown to have little impact upon the distribution of PAL astigmatism. A power increase in reading addition primarily caused a lengthening and narrowing of the lens progression channel. Empirical measurements also indicated a marginal steepening of the progression power gradient with an increase in reading addition power. A sample of the PAL wearing population were studied using patient records and questionnaire analysis (90% were returned). This subjective analysis revealed the reading portion to be the most troublesome lens zone and showed that patients with high astigmatism (> 2.00D) adapt more readily to PALs than those with spherical or low cylindrical (2.00D) corrections. The psychophysical features of PALs were then investigated. Both grafting visual acuity (VA) and contrast sensitivity (CS) were shown to be reduced with an increase in eccentricity from the central umbilical line. Two sample populations (N= 20) of successful and unsuccessful PAL wearers were assessed for differences in their visual performance and their adaptation to optically induced distortion. The possibility of dispensing errors being the cause of poor patient tolerance amongst the unsuccessful wearer group was investigated and discounted. The contrast sensitivity of the successful group was significantly greater than that of the unsuccessful group. No differences in adaptation to or detection of curvature distortion were evinced between these presbyopic groups.

A tree-based decision model to support prediction of the severity of asthma exacerbations in children

This paper describes the development of a tree-based decision model to predict the severity of pediatric asthma exacerbations in the emergency department (ED) at 2 h following triage. The model was constructed from retrospective patient data abstracted from the ED charts. The original data was preprocessed to eliminate questionable patient records and to normalize values of age-dependent clinical attributes. The model uses attributes routinely collected in the ED and provides predictions even for incomplete observations. Its performance was verified on independent validating data (split-sample validation) where it demonstrated AUC (area under ROC curve) of 0.83, sensitivity of 84%, specificity of 71% and the Brier score of 0.18. The model is intended to supplement an asthma clinical practice guideline, however, it can be also used as a stand-alone decision tool.

Examining the relationship between risk assessment and risk management in mental health

Risk assessment is crucial for developing risk management plans to prevent or minimize mental health patients' risks that will impede their recovery. Risk assessments and risk management plans should be closely linked. Their content and the extent to which they are linked within one Trust is explored. There is a great deal of variability in the amount and detail of risk information collected by nurses and how this is used to develop risk management plans. Keeping risk assessment information in one place rather than scattered throughout patient records is important for ensuring it can be accessed easily and linked properly to risk management plans. Strengthening the link between risk assessment and management will help ensure interventions and care is tailored to the specific needs of individual patients, thus promoting their safety and well-being. Thorough risk assessment helps in developing risk management plans that minimize risks that can impede mental health patients' recovery. Department of Health policy states that risk assessments and risk management plans should be inextricably linked. This paper examines their content and linkage within one Trust. Four inpatient wards for working age adults (18-65 years) in a large mental health Trust in England were included in the study. Completed risk assessment forms, for all patients in each inpatient ward were examined (n= 43), followed by an examination of notes for the same patients. Semi-structured interviews took place with ward nurses (n= 17). Findings show much variability in the amount and detail of risk information collected by nurses, which may be distributed in several places. Gaps in the risk assessment and risk management process are evident, and a disassociation between risk information and risk management plans is often present. Risk information should have a single location so that it can be easily found and updated. Overall, a more integrated approach to risk assessment and management is required, to help patients receive timely and appropriate interventions that can reduce risks such as suicide or harm to others. © 2011 Blackwell Publishing.

Relationship between sunlight and the age of onset of bipolar disorder:an international multisite study

Background - The onset of bipolar disorder is influenced by the interaction of genetic and environmental factors. We previously found that a large increase in sunlight in springtime was associated with a lower age of onset. This study extends this analysis with more collection sites at diverse locations, and includes family history and polarity of first episode. Methods - Data from 4037 patients with bipolar I disorder were collected at 36 collection sites in 23 countries at latitudes spanning 3.2 north (N) to 63.4 N and 38.2 south (S) of the equator. The age of onset of the first episode, onset location, family history of mood disorders, and polarity of first episode were obtained retrospectively, from patient records and/or direct interview. Solar insolation data were obtained for the onset locations. Results - There was a large, significant inverse relationship between maximum monthly increase in solar insolation and age of onset, controlling for the country median age and the birth cohort. The effect was reduced by half if there was no family history. The maximum monthly increase in solar insolation occurred in springtime. The effect was one-third smaller for initial episodes of mania than depression. The largest maximum monthly increase in solar insolation occurred in northern latitudes such as Oslo, Norway, and warm and dry areas such as Los Angeles, California. Limitations - Recall bias for onset and family history data. Conclusions - A large springtime increase in sunlight may have an important influence on the onset of bipolar disorder, especially in those with a family history of mood disorders.

Ethical issues about shared medical records

Focal points: Patient views on pharmacists' access to medical records was studied using a self-completion questionnaire in medical practice and two community pharmacies There was some support for pharmacist access to records with a third of the sample being unsure There was a majority support when the purpose was clearly pharmacy related A clear majority was confident of confidentiality in the pharmacy

Improving the description of the retinal vasculature and patient history taking for monitoring systemic hypertension

Systemic hypertension is an important public health concern. If optometrists are to perform a more active role in the detection and monitoring of high blood pressure (BP), there is a need to improve the consistency of describing the retinal vasculature and to assess patient's ability to correctly report the diagnosis of hypertension, its control and medication. One hundred and one patients aged >40 years were dilated and had fundus photography performed. BP was measured and a self-reported history of general health and current medication was compared with the records of their general practitioner (GP). The status of the retinal vasculature was quantified using a numeric scale by five clinicians and this was compared to the same evaluation performed with the aid of a basic pictorial grading scale. Image analysis was used to objectively measure the artery-to-vein (A/V) ratio and arterial reflex. Arteriolar tortuosity and calibre changes were found to be the most sensitive retinal signs of high BP. Using the grading scale to describe the retinal vasculature significantly improved inter- and intra-observer repeatability. Almost half the patients examined were on medication for high BP or cardiovascular disease. Patients' ability to give their complete medical history was poor, as was their ability to recall what medication they had been prescribed. GPs indicated it was useful to receive details of their patient's BP when it was >140/90 mmHg. The use of improved description of the retinal vasculature and stronger links between optometrists and GPs may enhance future patient care. © 2001 The College of Optometrists. Published by Elsevier Science Ltd. All rights reserved.