Improving data quality in English healthcare: from case studies to an applied framework
Leicester, H. J. (2004). Improving data quality in English healthcare: from case studies to an applied framework. (Unpublished Doctoral thesis, City, University of London)
Abstract
This Thesis argues that the quality of data for professionals and information for patients may be limiting healthcare in England. Data are required for individual care, for monitoring services for populations and as part of broader research and development. Similarly, informed patient consent and real involvement in their own care depend on information. Failings in these areas can lead to a fragmented National Health Service which is slow to change.
A Data Quality Framework (DQF) is therefore the main product of the Thesis. Prominent Themes which should be addressed have been identified through four representative Case Studies. From national data collection in Intensive Care to Visual Impairment Notification, the Studies span the range from quantity to quality of life and related healthcare services.
Datasets in the Studies are found to be poorly researched; lacking support for collection and management; and unlinked to information for patients. Assistance from technology is under developed, while overall costs and benefits of both data collection and provision of patient information are poorly documented.
These Prominent Themes are augmented by National Constraints derived from a review of healthcare policies and strategies for Information Management & Technology (IM&T). Demands for data and information are increasing and the delivery structures of care are changing through policy pledges. At the same time, healthcare is officially described as "largely based on manual systems". Problems with the introduction of technology are demonstrated by three IM&T strategies from 1998 to 2002. Culture and change management in the NHS have only recently been identified as major research issues. The necessary "information infrastructure" for data collection and information provision is still not in place at 2004.
The full Data Quality Framework has separate components for assessing an existing or proposed system (Step 1. Appraisal Tool) and introducing managed change (Step 2. Implementation Programme). It draws in particular on recent central initiatives paralleling the Prominent Themes with adjustments for the National Constraints.
The central initiatives cover: evolving mechanisms for appraising and approving all national healthcare datasets; care process modelling to highlight sources of data and points for information provision to patients; principles for accrediting information providers, paralleling those for organisations involved in NHS research; and standards for labelling information resources for indexing and easier retrieval (meta-data) as part of the eGovemment Interoperability Framework for the whole public sector.
Benefits are assessed by applying the DQF to the Visual Impairment Notification process and comparing a review of the same process by the Department of Health (still ongoing at late 2003).
Application of the DQF produces formal evidence to justify and direct change. Recommendations include a new visual impairment identification form linked to information sources and monitoring mechanisms, with pilots in both electronic and paper formats. In contrast, interim proposals from the official review has provided only anecdotal evidence and three new forms without a clear logic to support content or completion practices. The official approach contradicts the Department of Health's own policies on broad consultation and standards for developing national datasets. Moreover, alternative approaches derived from the DQF cannot be developed without prior Department of Health approval.
The DQF has not been fully applied and needs for refinement are acknowledged. Nevertheless, it is possibly the only example of its kind in English healthcare with general principles supported by evidence from the real world Case Studies.
Download (16MB) | Preview
Export
Downloads
Downloads per month over past year