Data driving improvement in healthcare

Improving the documentation, accuracy and use of patient data is imperative within the evolving healthcare landscape of today. Actively improving how patient related data is captured and then utilised has become high on every hospital’s priority list as data is now an important source of proving activity and increasingly used in areas such as securing adequate funding and gaining hospital accreditation. Thus, understanding data and improving the integrity of patient data have now become essential elements of many healthcare roles and professions.

The question now is how can we now use the masses of patient data available to affect real change?

Ensuring the validity of patient data is a real challenge for many healthcare professionals, as the required skillet for many traditional hospital roles is speedily changing to match the demands of this new, increasingly competitive era in healthcare. It is an advantage that there are already some exciting advancements happening to assist with the capture and breakdown of the large amounts of data that a hospital captures and stores, with the introduction of electronic health information systems.

Thankfully new technology has emerged to assist with data extraction, data validity, and the subsequent analyses of the masses of patient data that a hospital collates over time, however many are asking where to start, and what can you do on a smaller budget?

The upcoming Patient Data Improvement Forum is about the cross pollination of ideas, to find ways to better and more accurately document, classify, internally audit and analyse patient data, in addition to sharing ways to train staff around the change of work process that the electronic era brings. Bringing together health information managers, business and data analysts, costing and coding professionals.

The key areas of discussion at the Forum include:

GENERAL STREAM: Patient Data Driving Change

The use of patient data to explore healthcare coordination and collaboration:

• Use of patient data to extract different coordination and collaboration networks among healthcare professionals
• Explore and model the impact of different coordination and collaboration networks on patient outcomes (e.g. cost)
• Employ sophisticated methods (i.e. multi-level binomial logistic regression) to investigate differences in healthcare outcomes across different hospitals by utilising patient data

The role of data analytics and data governance in the delivery of Health Information to the acute health sector:

• The acute health sector in Australia is a “data rich and information poor” and does not meet all the clinical and business needs of their organisations.
• Information is crucial, yet an underutilised asset for managing patients in health organisations. To ensure that information being mined and analysed is of quality, leveraging the power of data analytics tools, a data governance framework has to be in place.
• Analytics and good data governance are used to manage current and future requirements both from a management and “changing models of care” perspective. Analytics and data governance can work hand in hand with an organisation’s strategic plan that can provide evidence based data and information to support their plan.

Data improvement and governance to ensure fitness for data and care improvement:

• Making primary & integrated care data accessible/available
• Mainstream data quality management across health enterprise
• Sociotechnical approaches to data quality management
• Integrated data and information governance across enterprise
• More intelligent use of health informatics and data analytics

COSTING STREAM: ABF Classification Updates

Improving transparency and efficiency in the delivery of public hospital services:

• How ABF provides incentives for greater efficiency
• Increasing transparency in delivery and funding
• What ABF means in terms of hospital management
• Working with hospitals for successful implementation

Using metrics to score the quality of costing data:

• NSW Health ABF data processes include a newly developed step which focuses on data quality and issue a scoring system
• The Reasonableness and Quality Application allows all LHDs to review how their scored against the metric for patient level costing data before formally submitting the results to the Ministry.
• The process has been designed as part of the continuous improvement cycle around patient level costing.

Budgeting using Activity Based Funding data – using the outputs from clinical costing to improve hospital operations:

• Commonwealth and state governments have sponsored national costing studies that collect cost data by patient across all states and care types.
• The result is a very large database of costs by care type allocated down to doctors, nurses, allied health, etc.
• Outlining a method for taking this costing data for use in a multiple regression model to budget for healthcare costs.

CODING STREAM: Classification & eHealth Updates

The impact of coding chronic disease patients:

• Reviewing how coding impacts on the accurate identification of Chronic disease patients when analysing costing data for opportunities for improvement.
• Outlining the significantly different cost and utilisation profile of chronic disease patients, however these patients can become quickly hidden as non- resource homogenous patients within a DRG in the absence of accurate coding.
• Using de-identified real examples over a four year period the impact of chronic disease on health service delivery will be discussed.

Coding of non-admitted procedures and diagnosis project:

• History/ background (non-admitted setting, classification, the subacute and non-admitted costing study- 341,000 patient level episodes with diagnosis and procedures)
• The project (outline of the project, development of coding rules, specific coding related issues identified during the project such as free text, blank fields etc.)
• Results of the project (clinics, procedures, diagnoses, basic analysis of data)
• Future use of the information/ conclusions.

Big data needs little data – the core of eHRs and coding:

• Traditional view of health data as purpose specific, form based resource (Message based model)
• eHealth needs singular shared view of health record and the data in it (the information based model)
• Discussion of the opportunity and requirements to support this change
• Impact on health information systems including coding processes and discharge summaries

For the latest information and program update, please visit the conference website.