How data quality affects funding for Ontario hospitals
In 2012, the Ontario government introduced Health System Funding Reform (HSFR) which moved hospitals from global budgets to a form of activity-based funding. Up to 70% of hospital funding in Ontario will eventually be from activity-based allocations.
In order for this funding model to work, the Ministry of Health and Long Term Care (MOHLTC) needs an accurate clinical representation of all patients treated by hospitals in the province. They get this information through the submission of patient encounter data files (patient abstracts) by each hospital. After a patient is discharged, coders in health records departments create an abstract designed to tell the story of a patient’s stay, based on physician documentation.
Unfortunately, ensuring that this data accurately represents what happened to each individual patient is challenging as clinical documentation can often be incomplete or incorrect. Physicians and health records professionals often don’t know which elements are important to capture to ensure that funding factors are comprehensively recorded. Most hospitals also lack sophisticated processes for coders to ask physicians questions about specific patients.
These data quality errors can be costly. Through the work 3terra has done, we’ve seen that data quality errors can reduce a hospital’s Health Based Allocation Model (HBAM) funding by over 1%. That equates to millions of dollars of lost annual revenue. The funding model is structured to reward hospitals that treat increasingly complex patients in cost-effective ways but poor quality data distorts the pie-sharing funding allocation and contributes to an uneven playing field. Hospitals often receive a larger share of funding simply because they are better at finding data quality errors compared to their peers.
The MOHLTC created a relative weighting system to assign a standardized case weight to each inpatient abstract. There are many factors that affect the weight that each patient abstract record receives including:
- If particular procedures (e.g. dialysis, chemotherapy, radiotherapy, mechanical ventilation) were performed during a stay
- Whether a patient was in the Intensive Care Unit (ICU) during their stay
- Whether a patient was transferred in or out to another inpatient facility
- Which patient condition (diagnosis) was responsible for the highest resource utilization during the visit
Any missing or incorrect information can have significant effects on funding for a hospital. Some hospitals have formed HSFR data quality committees to try to address these issues.
Unfortunately, most hospitals see tens of thousands of admitted patients each year and it is a very time consuming and laborious process to ensure that data quality is consistent across the board. 3terra offers a software tool that is already being used by many Ontario hospitals to quickly identify these types of data quality errors.
While we’ve successfully used algorithms to help hospitals correct data issues and improve their revenue performance, it is important that the expense side of the equation remains top-of-mind as well. Hospitals are increasingly focusing on reducing unwarranted clinical variation in the way they treat patients. Apart from quality, safety and patient experience benefits, tackling variation can significantly reduce costs. The bad news is that hospitals are spending a significant amount of resources compiling and analyzing data for compliance, reporting and financial purposes rather than using those resources to actually improve patient care.
Helping hospitals solve these clinical variation problems is 3terra’s focus for 2018. Analytics will continue to be a key factor to enable successful quality improvement (QI) initiatives. In future articles, we’ll show how data can be used to uncover operational problems (and therefore improvement opportunities) using a variety of machine learning, statistical, and visualization techniques. Luckily, hospitals have access to a large amount of raw data that pertain to their clinical practices, their nursing and equipment utilization, and their performance relative to their peers. As we’ll demonstrate, this can be used to help focus hospitals on areas that return the most value for their QI investments.
For more information about our data quality improvement product, please feel free to visit our website or contact us if you have any questions about how we help hospitals improve their provincial funding.
