Development and evaluation of a de-identification procedure for a case register sourced from mental health electronic records.

BACKGROUND: Electronic health records (EHRs) provide enormous potential for health research but also present data governance challenges. Ensuring de-identification is a pre-requisite for use of EHR data without prior consent. The South London and Maudsley NHS Trust (SLaM), one of the largest secondary mental healthcare providers in Europe, has developed, from its EHRs, a de-identified psychiatric case register, the Clinical Record Interactive Search (CRIS), for secondary research. METHODS: We describe development, implementation and evaluation of a bespoke de-identification algorithm used to create the register. It is designed to create dictionaries using patient identifiers (PIs) entered into dedicated source fields and then identify, match and mask them (with ZZZZZ) when they appear in medical texts. We deemed this approach would be effective, given high coverage of PI in the dedicated fields and the effectiveness of the masking combined with elements of a security model. We conducted two separate performance tests i) to test performance of the algorithm in masking individual true PIs entered in dedicated fields and then found in text (using 500 patient notes) and ii) to compare the performance of the CRIS pattern matching algorithm with a machine learning algorithm, called the MITRE Identification Scrubber Toolkit - MIST (using 70 patient notes - 50 notes to train, 20 notes to test on). We also report any incidences of potential breaches, defined by occurrences of 3 or more true or apparent PIs in the same patient's notes (and in an additional set of longitudinal notes for 50 patients); and we consider the possibility of inferring information despite de-identification. RESULTS: True PIs were masked with 98.8% precision and 97.6% recall. As anticipated, potential PIs did appear, owing to misspellings entered within the EHRs. We found one potential breach. In a separate performance test, with a different set of notes, CRIS yielded 100% precision and 88.5% recall, while MIST yielded a 95.1% and 78.1%, respectively. We discuss how we overcome the realistic possibility - albeit of low probability - of potential breaches through implementation of the security model. CONCLUSION: CRIS is a de-identified psychiatric database sourced from EHRs, which protects patient anonymity and maximises data available for research. CRIS demonstrates the advantage of combining an effective de-identification algorithm with a carefully designed security model. The paper advances much needed discussion of EHR de-identification - particularly in relation to criteria to assess de-identification, and considering the contexts of de-identified research databases when assessing the risk of breaches of confidential patient information.

Cardiovascular disease treatment among patients with severe mental illness: a data linkage study between primary and secondary care.

BACKGROUND: Suboptimal treatment of cardiovascular diseases (CVD) among patients with severe mental illness (SMI) may contribute to physical health disparities. AIM: To identify SMI characteristics associated with meeting CVD treatment and prevention guidelines. DESIGN AND SETTING: Population-based electronic health record database linkage between primary care and the sole provider of secondary mental health care services in south east London, UK. METHOD: Cardiovascular disease prevalence, risk factor recording, and Quality and Outcomes Framework (QOF) clinical target achievement were compared among 4056 primary care patients with SMI whose records were linked to secondary healthcare records and 270 669 patients without SMI who were not known to secondary care psychiatric services, using multivariate logistic regression modelling. Data available from secondary care records were then used to identify SMI characteristics associated with QOF clinical target achievement. RESULTS: Patients with SMI and with coronary heart disease and heart failure experienced reduced prescribing of beta blockers and angiotensin-converting enzyme inhibitor/angiotensin receptor blockers (ACEI/ARB). A diagnosis of schizophrenia, being identified with any indicator of risk or illness severity, and being prescribed with depot injectable antipsychotic medication was associated with the lowest likelihood of prescribing. CONCLUSION: Linking primary and secondary care data allows the identification of patients with SMI most at risk of undertreatment for physical health problems.

Developing a new model for patient recruitment in mental health services: a cohort study using Electronic Health Records.

OBJECTIVES: To develop a new model for patient recruitment that harnessed the full potential of Electronic Health Records (EHRs). Gaining access to potential participants' health records to assess their eligibility for studies and allow an approach about participation ('consent for contact') is ethically, legally and technically challenging, given that medical data are usually restricted to the patient's clinical team. The research objective was to design a model for identification and recruitment to overcome some of these challenges as well as reduce the burdensome (and/or time consuming) gatekeeper role of clinicians in determining who is appropriate or not to participate in clinical research. SETTING: Large secondary mental health services context, UK. PARTICIPANTS: 2106 patients approached for 'consent for contact'. All patients in different services within the mental health trust are gradually and systematically being approached by a member of the clinical care team using the 'consent for contact' model. There are no exclusion criteria. PRIMARY AND SECONDARY OUTCOME MEASURES: Provision of 'consent for contact'. RESULTS: A new model (the South London and Maudsley NHS Trust Consent for Contact model (SLaM C4C)) for gaining patients' consent to contact them about research possibilities, which is built around a de-identified EHR database. The model allows researchers to contact potential participants directly. Of 2106 patients approached by 25 October 2013, nearly 3 of every 4 gave consent for contact (1560 patients; 74.1%). CONCLUSIONS: The SLaM C4C model offers an effective way of expediting recruitment into health research through using EHRs. It reduces the gatekeeper function of clinicians; gives patients greater autonomy in decisions to participate in research; and accelerates the development of a culture of active research participation. More research is needed to assess how many of those giving consent for contact subsequently consent to participate in particular research studies.

Evaluation of facial soft tissue changes with optical surface scan after surgical correction of Class III deformities.

PURPOSE: To evaluate the soft tissue changes after correction of Class III dentoskeletal deformity with orthognathic surgery using the optical surface scanner as a 3-dimensional imaging tool and thin-plate splines as a morphometric analysis. PATIENTS AND METHODS: Forty-six patients with Class III dentoskeletal relationship who underwent a bimaxillary orthognathic surgical procedure involving advancement of the maxilla and mandibular setback to correct the deformity and create a Class I dentoskeletal relationship were included. Optical surface scans (lasergraphs) were obtained from all patients preoperatively and 6 months postoperatively. They were evaluated for the soft tissue changes 3-dimensionally and on the profile line. The cephalographs taken pre- and postoperatively were digitized and evaluated using the CogSoft digitizing software (Consultant Orthodontists Group Software, Bristol, United Kingdom) to confirm whether the preoperative surgical plan was applied. RESULTS: The soft tissue change in different areas of the face ranged from 30% on the nasal tip to 80% on subnasale and from 50% on subcomissural region corresponding to lateral parts of the mandibular body to 100% on the pogonion. The direction of soft tissue movement was not solely forwards and backwards on a horizontal plane but had upward and downward vectors. CONCLUSIONS: The soft tissue changes following orthognathic surgery depend on various factors; hence it can be difficult to predict. However, there is a general trend as discussed in this study for the direction and amounts of facial soft tissue changes in the middle and lower facial thirds.