A Rosemary Tate

The health of UK military personnel who deployed to the 2003 Iraq war: a cohort study.

BACKGROUND Concerns have been raised about the mental and physical health of UK military personnel who deployed to the 2003 war in Iraq and subsequent tours of duty in the country. METHODS We compared health outcomes in a random sample of UK armed forces personnel who were deployed to the 2003 Iraq war with those in personnel who were not deployed. Participants completed a questionnaire covering the nature of the deployment and health outcomes, which included symptoms of post-traumatic stress disorder, common mental disorders, general wellbeing, alcohol consumption, physical symptoms, and fatigue. FINDINGS The participation rate was 62.3% (n=4722) in the deployed sample, and 56.3% (n=5550) in the non-deployed sample. Differences in health outcomes between groups were slight. There was a modest increase in the number of individuals with multiple physical symptoms (odds ratio 1.33; 95% CI 1.15-1.54). No other differences between groups were noted. The effect of deployment was different for reservists compared with regulars. In regulars, only presence of multiple physical symptoms was weakly associated with deployment (1.32; 1.14-1.53), whereas for reservists deployment was associated with common mental disorders (2.47, 1.35-4.52) and fatigue (1.78; 1.09-2.91). There was no evidence that later deployments, which were associated with escalating insurgency and UK casualties, were associated with poorer mental health outcomes. INTERPRETATION For regular personnel in the UK armed forces, deployment to the Iraq war has not, so far, been associated with significantly worse health outcomes, apart from a modest effect on multiple physical symptoms. There is evidence of a clinically and statistically significant effect on health in reservists.

Independent component analysis for automated decomposition of in vivo magnetic resonance spectra.

Fully automated methods for analyzing MR spectra would be of great benefit for clinical diagnosis, in particular for the extraction of relevant information from large databases for subsequent pattern recognition analysis. Independent component analysis (ICA) provides a means of decomposing signals into their constituent components. This work investigates the use of ICA for automatically extracting features from in vivo MR spectra. After its limits are assessed on artificial data, the method is applied to a set of brain tumor spectra. ICA automatically, and in an unsupervised fashion, decomposes the signals into interpretable components. Moreover, the spectral decomposition achieved by the ICA leads to the separation of some tissue types, which confirms the biochemical relevance of the components. Magn Reson Med 50:697–703, 2003.

Exploiting the potential of large databases of electronic health records for research using rapid search algorithms and an intuitive query interface.

Objective UK primary care databases, which contain diagnostic, demographic and prescribing information for millions of patients geographically representative of the UK, represent a significant resource for health services and clinical research. They can be used to identify patients with a specified disease or condition (phenotyping) and to investigate patterns of diagnosis and symptoms. Currently, extracting such information manually is time-consuming and requires considerable expertise. In order to exploit more fully the potential of these large and complex databases, our interdisciplinary team developed generic methods allowing access to different types of user. Materials and methods Using the Clinical Practice Research Datalink database, we have developed an online user-focused system (TrialViz), which enables users interactively to select suitable medical general practices based on two criteria: suitability of the patient base for the intended study (phenotyping) and measures of data quality. Results An end-to-end system, underpinned by an innovative search algorithm, allows the user to extract information in near real-time via an intuitive query interface and to explore this information using interactive visualization tools. A usability evaluation of this system produced positive results. Discussion We present the challenges and results in the development of TrialViz and our plans for its extension for wider applications of clinical research. Conclusions Our fast search algorithms and simple query algorithms represent a significant advance for users of clinical research databases.

Using free text information to explore how and when GPs code a diagnosis of ovarian cancer: an observational study using primary care records of patients with ovarian cancer

Background Primary care databases provide a unique resource for healthcare research, but most researchers currently use only the Read codes for their studies, ignoring information in the free text, which is much harder to access. Objectives To investigate how much information on ovarian cancer diagnosis is ‘hidden’ in the free text and the time lag between a diagnosis being described in the text or in a hospital letter and the patient being given a Read code for that diagnosis. Design Anonymised free text records from the General Practice Research Database of 344 women with a Read code indicating ovarian cancer between 1 June 2002 and 31 May 2007 were used to compare the date at which the diagnosis was first coded with the date at which the diagnosis was recorded in the free text. Free text relating to a diagnosis was identified (a) from the date of coded diagnosis and (b) by searching for words relating to the ovary. Results 90% of cases had information relating to their ovary in the free text. 45% had text indicating a definite diagnosis of ovarian cancer. 22% had text confirming a diagnosis before the coded date; 10% over 4 weeks previously. Four patients did not have ovarian cancer and 10% had only ambiguous or suspected diagnoses associated with the ovarian cancer code. Conclusions There was a vast amount of extra information relating to diagnoses in the free text. Although in most cases text confirmed the coded diagnosis, it also showed that in some cases GPs do not code a definite diagnosis on the date that it is confirmed. For diseases which rely on hospital consultants for diagnosis, free text (particularly letters) is invaluable for accurate dating of diagnosis and referrals and also for identifying misclassified cases.

Extracting diagnoses and investigation results from unstructured text in electronic health records by semi-supervised machine learning.

Background Electronic health records are invaluable for medical research, but much of the information is recorded as unstructured free text which is time-consuming to review manually. Aim To develop an algorithm to identify relevant free texts automatically based on labelled examples. Methods We developed a novel machine learning algorithm, the ‘Semi-supervised Set Covering Machine’ (S3CM), and tested its ability to detect the presence of coronary angiogram results and ovarian cancer diagnoses in free text in the General Practice Research Database. For training the algorithm, we used texts classified as positive and negative according to their associated Read diagnostic codes, rather than by manual annotation. We evaluated the precision (positive predictive value) and recall (sensitivity) of S3CM in classifying unlabelled texts against the gold standard of manual review. We compared the performance of S3CM with the Transductive Vector Support Machine (TVSM), the original fully-supervised Set Covering Machine (SCM) and our ‘Freetext Matching Algorithm’ natural language processor. Results Only 60% of texts with Read codes for angiogram actually contained angiogram results. However, the S3CM algorithm achieved 87% recall with 64% precision on detecting coronary angiogram results, outperforming the fully-supervised SCM (recall 78%, precision 60%) and TSVM (recall 2%, precision 3%). For ovarian cancer diagnoses, S3CM had higher recall than the other algorithms tested (86%). The Freetext Matching Algorithm had better precision than S3CM (85% versus 74%) but lower recall (62%). Conclusions Our novel S3CM machine learning algorithm effectively detected free texts in primary care records associated with angiogram results and ovarian cancer diagnoses, after training on pre-classified test sets. It should be easy to adapt to other disease areas as it does not rely on linguistic rules, but needs further testing in other electronic health record datasets.

Optimising use of electronic health records to describe the presentation of rheumatoid arthritis in primary care: a strategy for developing code lists.

Background Research using electronic health records (EHRs) relies heavily on coded clinical data. Due to variation in coding practices, it can be difficult to aggregate the codes for a condition in order to define cases. This paper describes a methodology to develop ‘indicator markers’ found in patients with early rheumatoid arthritis (RA); these are a broader range of codes which may allow a probabilistic case definition to use in cases where no diagnostic code is yet recorded. Methods We examined EHRs of 5,843 patients in the General Practice Research Database, aged ≥30y, with a first coded diagnosis of RA between 2005 and 2008. Lists of indicator markers for RA were developed initially by panels of clinicians drawing up code-lists and then modified based on scrutiny of available data. The prevalence of indicator markers, and their temporal relationship to RA codes, was examined in patients from 3y before to 14d after recorded RA diagnosis. Findings Indicator markers were common throughout EHRs of RA patients, with 83.5% having 2 or more markers. 34% of patients received a disease-specific prescription before RA was coded; 42% had a referral to rheumatology, and 63% had a test for rheumatoid factor. 65% had at least one joint symptom or sign recorded and in 44% this was at least 6-months before recorded RA diagnosis. Conclusion Indicator markers of RA may be valuable for case definition in cases which do not yet have a diagnostic code. The clinical diagnosis of RA is likely to occur some months before it is coded, shown by markers frequently occurring ≥6 months before recorded diagnosis. It is difficult to differentiate delay in diagnosis from delay in recording. Information concealed in free text may be required for the accurate identification of patients and to assess the quality of care in general practice.

Classification of brain tumours from MR spectra: the INTERPRET collaboration and its outcomes.

The INTERPRET project was a multicentre European collaboration, carried out from 2000 to 2002, which developed a decision‐support system (DSS) for helping neuroradiologists with no experience of MRS to utilize spectroscopic data for the diagnosis and grading of human brain tumours. INTERPRET gathered a large collection of MR spectra of brain tumours and pseudo‐tumoural lesions from seven centres. Consensus acquisition protocols, a standard processing pipeline and strict methods for quality control of the aquired data were put in place. Particular emphasis was placed on ensuring the diagnostic certainty of each case, for which all cases were evaluated by a clinical data validation committee. One outcome of the project is a database of 304 fully validated spectra from brain tumours, pseudotumoural lesions and normal brains, along with their associated images and clinical data, which remains available to the scientific and medical community. The second is the INTERPRET DSS, which has continued to be developed and clinically evaluated since the project ended.

Quality of recording of diabetes in the UK: how does the GP's method of coding clinical data affect incidence estimates? Cross-sectional study using the CPRD database

Objective To assess the effect of coding quality on estimates of the incidence of diabetes in the UK between 1995 and 2014. Design A cross-sectional analysis examining diabetes coding from 1995 to 2014 and how the choice of codes (diagnosis codes vs codes which suggest diagnosis) and quality of coding affect estimated incidence. Setting Routine primary care data from 684 practices contributing to the UK Clinical Practice Research Datalink (data contributed from Vision (INPS) practices). Main outcome measure Incidence rates of diabetes and how they are affected by (1) GP coding and (2) excluding ‘poor’ quality practices with at least 10% incident patients inaccurately coded between 2004 and 2014. Results Incidence rates and accuracy of coding varied widely between practices and the trends differed according to selected category of code. If diagnosis codes were used, the incidence of type 2 increased sharply until 2004 (when the UK Quality Outcomes Framework was introduced), and then flattened off, until 2009, after which they decreased. If non-diagnosis codes were included, the numbers continued to increase until 2012. Although coding quality improved over time, 15% of the 666 practices that contributed data between 2004 and 2014 were labelled ‘poor’ quality. When these practices were dropped from the analyses, the downward trend in the incidence of type 2 after 2009 became less marked and incidence rates were higher. Conclusions In contrast to some previous reports, diabetes incidence (based on diagnostic codes) appears not to have increased since 2004 in the UK. Choice of codes can make a significant difference to incidence estimates, as can quality of recording. Codes and data quality should be checked when assessing incidence rates using GP data.

Characterisation of Data Quality in Electronic Healthcare Records

The quality of information depends on the quality of data from which it is derived, but data are of high quality only if they are fit for their intended use. Although electronic healthcare records are collected primarily for patient care and audits, their use in research can also greatly benefit the quality of life of patients. We aim to discuss the challenges and issues involved with measuring data quality in electronic health records for epidemiological and clinical research using the Clinical Practice Research Datalink as a model. We also will share our experiences of assessing data quality in medical primary care database CPRD GOLD and discuss a suggested framework that can help ensure compatibility of data quality measures for different European Electronic Healthcare Records.

A pragmatic approach for measuring data quality in primary care databases

There is currently no widely recognised methodology for undertaking data quality assessment in electronic health records used for research. In an attempt to address this, we have developed a protocol for measuring and monitoring data quality in primary care research databases, whereby practice-based data quality measures are tailored to the intended use of the data. Our approach was informed by an in-depth investigation of aspects of data quality in the Clinical Practice Research Datalink Gold database and presentations of the results to data users. Although based on a primary care database, much of our proposed approach would be equally applicable to other health care databases.