The use of healthcare data by commercial agencies attracts regular and frequent comment. This is entirely appropriate because it is an activity where risks and benefits need to be carefully considered.
On the one hand, the NHS does rely on Industry to deliver new investigations and treatments, and these can’t be effectively developed without access to clinical information – whether from patients directly or from routine healthcare data. On the other hand, there are clearly concerns about use of information for commercial purposes, as well as the need to maintain the security and confidentiality of information and to ensure the appropriateness of its use.
CRIS-derived data has been used in commercial partnerships at various times during the 15+ years since the development of the database, and these are listed below for transparency, to the best of our knowledge. Where possible, we have sought to ensure that findings are published in open-access sources.
For Industry partnerships to date, the preferred protocol (applying to all studies listed below) has been for data to be accessed by university or Trust staff and the findings generated according to the specifications of the commercial partner, with no direct access to data by employees of the partners themselves. The only exceptions to this have been the preparation of a dataset for LA-SER to carry out a multi-site European study of paliperidone palmitate prescribing, and for Optum to provide a SLaM-commissioned service utilisation review, both projects carried out in 2014. These were carried out by employees of the commercial partner and were covered by the standard confidentiality and security agreements for any agency working with Trust data. CRIS data remained within the Trust’s firewall at all times.
Where funding is received from these partnerships, it is kept distinct in a separate account and its use is governed by the CRIS Oversight Committee. As a principle, the funds are directed towards covering the costs incurred in maximising the utility of the CRIS platform; for example, funds have been used in the past to support the data linkage between CRIS and Hospital Episode Statistics.
Projects completed or in progress
Health and social care costs associated with different levels of cognitive function in dementia: this project was funded by a pre-competitive consortium between King’s College London, Pfizer, Janssen and Lundbeck. Findings have been published.
Developing a mechanism to ascertain negative symptoms of schizophrenia in mental healthcare clinical text: this project was funded by Roche. Findings have been published.
Investigating the performance of an algorithm to detect parkinsonian adverse effects in people receiving antipsychotic polypharmacy: this project was carried out in collaboration with In Silico Biosciences. Findings have been published.
Investigating mental healthcare contacts before and after commencement of paliperidone palmitate maintenance: this project was funded by Janssen. Findings have been published.
Development and evaluation of algorithms to ascertain treatment-resistant depression and depression associated with high suicide risk in routine mental healthcare records: this project was funded by Janssen beginning initially with algorithm development and descriptions of characteristics of these sub-groups of depression, followed by further development of the treatment-resistance algorithm and information provided on levels of service use for economic analyses. This work remains in progress as a core CRIS activity, no longer directly funded.
Identifying markers of treatment-resistance and clozapine non-response using the CRIS dataset. This study was nested within the Schizophrenia Treatment Resistance And Therapeutic Advances (STRATA) research programme and was funded by H Lundbeck A/S. Findings have been published in the following papers:
- A predictor model of treatment resistance in schizophrenia using data from electronic health records
- Using a statistical learning approach to identify sociodemographic and clinical predictors of response to clozapine
Industry-supported CRIS-based PhD programmes
A completed PhD studentship part-supported by GSK, investigated associations between dementia, patterns of hospitalisation and diabetes outcome, resulting in the following publications:
- The association between dementia severity and hospitalisation profile in a newly assessed clinical cohort: the South London and Maudsley case register
- Potentially avoidable causes of hospitalisation in people with dementia: contemporaneous associations by stage of dementia in a South London clinical cohort
- The Impact of Dementia on Diabetes Control: An Evaluation of HbA1c Trajectories and Care Outcomes in Linked Primary and Specialist Care Data
A completed Takeda-funded PhD studentship which enabled the ascertainment of cognitive impairment in schizophrenia, as well as broader applications of natural language processing to multimorbidity and experiences of violence, resulting in the following publications:
- Cognitive Trajectories in Comorbid Dementia With Schizophrenia or Bipolar Disorder: The South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLaM BRC) Case Register
- Cognitive Impairments in Schizophrenia: A Study in a Large Clinical Sample Using Natural Language Processing
- Investigating the association between physical health comorbidities and disability in individuals with severe mental illness
- Mapping multimorbidity in individuals with schizophrenia and bipolar disorders: evidence from the South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLAM BRC) case register
- Can natural language processing models extract and classify instances of interpersonal violence in mental healthcare electronic records: an applied evaluative study
- Multi-domain Clinical Natural Language Processing with MedCAT: the Medical Concept Annotation Toolkit
Other Industry Collaborations
In addition to the specific projects listed above, CRIS data analyses carried out by university/Trust staff have supported several successful research programmes which have incorporated both academic and commercial partners. These include Pheme, KConnect, and an MRC Immunopsychiatry consortium. Industry engagement is also a focus for the DATAMIND HDRUK Mental Health Hub to which CRIS is a key contributor.
