Long term extension of a randomised controlled trial of probiotics using electronic health records.

Most randomised controlled trials (RCTs) are relatively short term and, due to costs and available resources, have limited opportunity to be re-visited or extended. There is no guarantee that effects of treatments remain unchanged beyond the study. Here, we illustrate the feasibility, benefits and cost-effectiveness of enriching standard trial design with electronic follow up. We completed a 5-year electronic follow up of a RCT investigating the impact of probiotics on asthma and eczema in children born 2005-2007, with traditional fieldwork follow up to two years. Participants and trial outcomes were identified and analysed after five years using secure, routine, anonymised, person-based electronic health service databanks. At two years, we identified 93% of participants and compared fieldwork with electronic health records, highlighting areas of agreement and disagreement. Retention of children from lower socio-economic groups was improved, reducing volunteer bias. At 5 years we identified a reduced 82% of participants. These data allowed the trial's first robust analysis of asthma endpoints. We found no indication that probiotic supplementation to pregnant mothers and infants protected against asthma or eczema at 5 years. Continued longer-term follow up is technically straightforward.

The use of secure anonymised data linkage to determine changes in healthcare utilisation following severe open tibial fractures.

Severe open fractures of the lower limbs are complex injuries requiring expert multidisciplinary management in appropriate orthoplastic centres. This study aimed to assess the impact of open fractures on healthcare utilisation and test the null hypotheses that there is no difference in healthcare utilisation between the year before and year after injury, and that there is no difference in healthcare utilisation in the year post-injury between patients admitted directly to an orthoplastic centre in keeping with the joint BOA/BAPRAS standards and those having initial surgery elsewhere. This retrospective cohort study utilising secure anonymised information linkage (SAIL), a novel databank of anonymised nationally pooled health records, recruited patients over 18 years of age sustaining severe open lower limb fractures managed primarily or secondarily at our centre and who had data available in the SAIL databank. 101 patients met inclusion criteria and 90 of these had records in the SAIL databank. The number of days in hospital, number of primary care attendances, number of outpatient attendances and number of emergency department attendances in the years prior and subsequent to injury were recorded. Patients sustaining open fractures had significantly different healthcare utilisation in the year after injury when compared with the year before, in terms of days spent in hospital (23.42 vs. 1.70, p=0.000), outpatient attendances (11.98 vs. 1.05, p=0.000), primary care attendances (29.48 vs. 11.99, p=0.000) and emergency department presentations (0.2 vs. 0.01, p=0.025). Patients admitted directly to orthoplastic centres had significantly fewer operations (1.78 vs. 3.31) and GP attendances (23.6 vs. 33.52) than those transferred in subsequent to initial management in other units. There is a significant increase in healthcare utilisation after open tibial fracture. Adherence to national standards minimises the impact of this on both patients and health services.

Healthcare databases in Europe for studying medicine use and safety during pregnancy.

PURPOSE: The aim of this study was to describe a number of electronic healthcare databases in Europe in terms of the population covered, the source of the data captured and the availability of data on key variables required for evaluating medicine use and medicine safety during pregnancy. METHODS: A sample of electronic healthcare databases that captured pregnancies and prescription data was selected on the basis of contacts within the EUROCAT network. For each participating database, a database inventory was completed. RESULTS: Eight databases were included, and the total population covered was 25 million. All databases recorded live births, seven captured stillbirths and five had full data available on spontaneous pregnancy losses and induced terminations. In six databases, data were usually available to determine the date of the woman's last menstrual period, whereas in the remainder, algorithms were needed to establish a best estimate for at least some pregnancies. In seven databases, it was possible to use data recorded in the databases to identify pregnancies where the offspring had a congenital anomaly. Information on confounding variables was more commonly available in databases capturing data recorded by primary-care practitioners. All databases captured maternal co-prescribing and a measure of socioeconomic status. CONCLUSION: This study suggests that within Europe, electronic healthcare databases may be valuable sources of data for evaluating medicine use and safety during pregnancy. The suitability of a particular database, however, will depend on the research question, the type of medicine to be evaluated, the prevalence of its use and any adverse outcomes of interest. © 2014 The Authors. Pharmacoepidemiology and Drug Safety published by John Wiley & Sons, Ltd.

A case study of the Secure Anonymous Information Linkage (SAIL) Gateway: a privacy-protecting remote access system for health-related research and evaluation.

With the current expansion of data linkage research, the challenge is to find the balance between preserving the privacy of person-level data whilst making these data accessible for use to their full potential. We describe a privacy-protecting safe haven and secure remote access system, referred to as the Secure Anonymised Information Linkage (SAIL) Gateway. The Gateway provides data users with a familiar Windows interface and their usual toolsets to access approved anonymously-linked datasets for research and evaluation. We outline the principles and operating model of the Gateway, the features provided to users within the secure environment, and how we are approaching the challenges of making data safely accessible to increasing numbers of research users. The Gateway represents a powerful analytical environment and has been designed to be scalable and adaptable to meet the needs of the rapidly growing data linkage community.

Volunteer bias in recruitment, retention, and blood sample donation in a randomised controlled trial involving mothers and their children at six months and two years: a longitudinal analysis.

BACKGROUND: The vulnerability of clinical trials to volunteer bias is under-reported. Volunteer bias is systematic error due to differences between those who choose to participate in studies and those who do not. METHODS AND RESULTS: This paper extends the applications of the concept of volunteer bias by using data from a trial of probiotic supplementation for childhood atopy in healthy dyads to explore 1) differences between a) trial participants and aggregated data from publicly available databases b) participants and non-participants as the trial progressed 2) impact on trial findings of weighting data according to deprivation (Townsend) fifths in the sample and target populations. 1) a) Recruits (n = 454) were less deprived than the target population, matched for area of residence and delivery dates (n = 6,893) (mean [SD] deprivation scores 0.09[4.21] and 0.79[4.08], t = 3.44, df = 511, p<0.001). b) i) As the trial progressed, representation of the most deprived decreased. These participants and smokers were less likely to be retained at 6 months (n = 430[95%]) (OR 0.29,0.13-0.67 and 0.20,0.09-0.46), and 2 years (n = 380[84%]) (aOR 0.68,0.50-0.93 and 0.55,0.28-1.09), and consent to infant blood sample donation (n = 220[48%]) (aOR 0.72,0.57-0.92 and 0.43,0.22-0.83). ii) Mothers interested in probiotics or research or reporting infants' adverse events or rashes were more likely to attend research clinics and consent to skin-prick testing. Mothers participating to help children were more likely to consent to infant blood sample donation. 2) In one trial outcome, atopic eczema, the intervention had a positive effect only in the over-represented, least deprived group. Here, data weighting attenuated risk reduction from 6.9%(0.9-13.1%) to 4.6%(-1.4-+10.5%), and OR from 0.40(0.18-0.91) to 0.56(0.26-1.21). Other findings were unchanged. CONCLUSIONS: Potential for volunteer bias intensified during the trial, due to non-participation of the most deprived and smokers. However, these were not the only predictors of non-participation. Data weighting quantified volunteer bias and modified one important trial outcome. TRIAL REGISTRATION: This randomised, double blind, parallel group, placebo controlled trial is registered with the International Standard Randomised Controlled Trials Register, Number (ISRCTN) 26287422. Registered title: Probiotics in the prevention of atopy in infants and children.

Toxic marine microalgae and shellfish poisoning in the British isles: history, review of epidemiology, and future implications.

The relationship between toxic marine microalgae species and climate change has become a high profile and well discussed topic in recent years, with research focusing on the possible future impacts of changing hydrological conditions on Harmful Algal Bloom (HAB) species around the world. However, there is very little literature concerning the epidemiology of these species on marine organisms and human health. Here, we examine the current state of toxic microalgae species around the UK, in two ways: first we describe the key toxic syndromes and gather together the disparate reported data on their epidemiology from UK records and monitoring procedures. Secondly, using NHS hospital admissions and GP records from Wales, we attempt to quantify the incidence of shellfish poisoning from an independent source. We show that within the UK, outbreaks of shellfish poisoning are rare but occurring on a yearly basis in different regions and affecting a diverse range of molluscan shellfish and other marine organisms. We also show that the abundance of a species does not necessarily correlate to the rate of toxic events. Based on routine hospital records, the numbers of shellfish poisonings in the UK are very low, but the identification of the toxin involved, or even a confirmation of a poisoning event is extremely difficult to diagnose. An effective shellfish monitoring system, which shuts down aquaculture sites when toxins exceed regularity limits, has clearly prevented serious impact to human health, and remains the only viable means of monitoring the potential threat to human health. However, the closure of these sites has an adverse economic impact, and the monitoring system does not include all toxic plankton. The possible geographic spreading of toxic microalgae species is therefore a concern, as warmer waters in the Atlantic could suit several species with southern biogeographical affinities enabling them to occupy the coastal regions of the UK, but which are not yet monitored or considered to be detrimental.

The prescribed duration algorithm: utilising 'free text' from multiple primary care electronic systems.

PURPOSE: To develop and test an algorithm that translates total dose and daily regimen, inputted as 'free text' on a prescription, into numerical values to calculate the prescribed treatment duration. METHOD: The algorithm was developed using antibiotic prescriptions (n = 711,714) from multiple primary care computer systems. For validation, the prescribed treatment duration of an independent sample of antibiotic scripts was calculated in two ways: (a) computer algorithm, (b) manually reviewed by a researcher blinded to the results of (a). The outputs of the two methods were compared and the level of agreement assessed, using confidence intervals for differences in proportions. This was repeated on sample of antidepressant scripts to test generalisability of the algorithm. RESULTS: For the antibiotic prescriptions, the algorithm processed 98.5% with an accuracy of 99.8% and the manual review processed 98.5% with 98.9% accuracy. The differences between these proportions are 0.0% (95%CI of -0.9, 0.9%) and 1.0% (95%CI of -0.1, 2.3%), respectively. For the antidepressant prescriptions, the algorithm processed 91.5% with an accuracy of 96.6% compared to the manual review with 96.4% processed and 99.8% accuracy; difference between these proportions is 4.9% (95%CI of 2.0, 8.0%) and 3.2% (95%CI of 1.6, 5.3%), respectively. CONCLUSION: The algorithm proved to be applicable and efficient for assessing prescribed duration, with sensitivity and specificity values close to the manual review, but with the added advantage that the computer can process large volume of scripts rapidly and automatically.

Frequency and characteristics of disease flares in ankylosing spondylitis.

OBJECTIVE: To examine the characteristics and frequency of disease flares in a cohort of people with AS. METHODS: A prospective data set from a randomized controlled trial (RCT) of a probiotic treatment/placebo was utilized to examine disease flares in 134 people with AS. Disease flares were defined as either minor/localized flares (pain/swelling localized to one area with fatigue and stiffness) or major/generalized flares (generalized pain, hot burning joints, muscle spasm, fever, sweating, extreme fatigue and stiffness). Results. One hundred and thirty-four people were followed up for 1216 person follow-up weeks and there were 71.4 flares per 100 person-weeks. Of these, 12 were major/generalized flares and 59.4 were minor/localized flares. People who experienced at least one major/generalized flare during the study period had worse disease during flare-free periods in terms of disease activity, impaired function, self-reported night pain and iritis compared with those who did not experience any major/generalized flares during the study. Major/generalized flares lasted for an average of 2.4 weeks (s.d. 2.7), and were preceded by and followed by a minor/localized flare in 92% (55/60) of cases. CONCLUSION: Seventy per cent of people with AS felt they had a flare in any given week. Those who experienced major/generalized flares appear to have more severe and active disease even during periods when not in flare, compared with those who do not experience major/generalized flares. These results have implications for the timing of assessments prior to starting anti-TNF therapy and suggest that the presence of major flares may be helpful in identifying patients with severe disease.

The SAIL Databank: building a national architecture for e-health research and evaluation.

BACKGROUND: Vast quantities of electronic data are collected about patients and service users as they pass through health service and other public sector organisations, and these data present enormous potential for research and policy evaluation. The Health Information Research Unit (HIRU) aims to realise the potential of electronically-held, person-based, routinely-collected data to conduct and support health-related studies. However, there are considerable challenges that must be addressed before such data can be used for these purposes, to ensure compliance with the legislation and guidelines generally known as Information Governance. METHODS: A set of objectives was identified to address the challenges and establish the Secure Anonymised Information Linkage (SAIL) system in accordance with Information Governance. These were to: 1) ensure data transportation is secure; 2) operate a reliable record matching technique to enable accurate record linkage across datasets; 3) anonymise and encrypt the data to prevent re-identification of individuals; 4) apply measures to address disclosure risk in data views created for researchers; 5) ensure data access is controlled and authorised; 6) establish methods for scrutinising proposals for data utilisation and approving output; and 7) gain external verification of compliance with Information Governance. RESULTS: The SAIL databank has been established and it operates on a DB2 platform (Data Warehouse Edition on AIX) running on an IBM 'P' series Supercomputer: Blue-C. The findings of an independent internal audit were favourable and concluded that the systems in place provide adequate assurance of compliance with Information Governance. This expanding databank already holds over 500 million anonymised and encrypted individual-level records from a range of sources relevant to health and well-being. This includes national datasets covering the whole of Wales (approximately 3 million population) and local provider-level datasets, with further growth in progress. The utility of the databank is demonstrated by increasing engagement in high quality research studies. CONCLUSION: Through the pragmatic approach that has been adopted, we have been able to address the key challenges in establishing a national databank of anonymised person-based records, so that the data are available for research and evaluation whilst meeting the requirements of Information Governance.

Are blood pressure levels taken during a secondary care diabetic clinic likely to be higher than when measured in primary care?

Blood pressure management guidelines are widely available. Our aim was to examine differences in blood pressure on the same subjects in primary and secondary care. We identified 661 diabetic subjects with measurements with the aid of the SAIL databank. Blood pressure in primary care was significantly lower than secondary care.

Residential Anonymous Linking Fields (RALFs): a novel information infrastructure to study the interaction between the environment and individuals' health.

BACKGROUND: The necessity of aggregating health data over areas can impede our understanding of health determinants. METHODS: We demonstrate the possibility of creating anonymous links between individual residences and the local environment using digital map data and a data linkage system. RESULTS: Digital map data were used successfully to anonymously link 1.3 million addresses to the local environment. The data linkage system allows detailed environment data surrounding each residence to be linked both to each resident therein and to their medical records. CONCLUSIONS: Local environment data specific to each house can be effectively and anonymously linked to the population registered with the National Health Service. Our integrated approach potentially enables flexible fine-scale, large-area observational studies of communities and health.

The SAIL databank: linking multiple health and social care datasets.

BACKGROUND: Vast amounts of data are collected about patients and service users in the course of health and social care service delivery. Electronic data systems for patient records have the potential to revolutionise service delivery and research. But in order to achieve this, it is essential that the ability to link the data at the individual record level be retained whilst adhering to the principles of information governance. The SAIL (Secure Anonymised Information Linkage) databank has been established using disparate datasets, and over 500 million records from multiple health and social care service providers have been loaded to date, with further growth in progress. METHODS: Having established the infrastructure of the databank, the aim of this work was to develop and implement an accurate matching process to enable the assignment of a unique Anonymous Linking Field (ALF) to person-based records to make the databank ready for record-linkage research studies. An SQL-based matching algorithm (MACRAL, Matching Algorithm for Consistent Results in Anonymised Linkage) was developed for this purpose. Firstly the suitability of using a valid NHS number as the basis of a unique identifier was assessed using MACRAL. Secondly, MACRAL was applied in turn to match primary care, secondary care and social services datasets to the NHS Administrative Register (NHSAR), to assess the efficacy of this process, and the optimum matching technique. RESULTS: The validation of using the NHS number yielded specificity values > 99.8% and sensitivity values > 94.6% using probabilistic record linkage (PRL) at the 50% threshold, and error rates were < 0.2%. A range of techniques for matching datasets to the NHSAR were applied and the optimum technique resulted in sensitivity values of: 99.9% for a GP dataset from primary care, 99.3% for a PEDW dataset from secondary care and 95.2% for the PARIS database from social care. CONCLUSION: With the infrastructure that has been put in place, the reliable matching process that has been developed enables an ALF to be consistently allocated to records in the databank. The SAIL databank represents a research-ready platform for record-linkage studies.