pMD prides itself on helping providers and practices accomplish their goals and measure their quality improvement initiatives. In the inpatient environment, we often work with groups that are interested in tracking patients’ length of stay (LOS) and assessing how their data compares to estimated values for particular medical conditions or patient populations.
While there are many views on which calculation is the most accurate for a patient’s estimated length of stay, for the sake of this discussion, I will use the geometric mean length of stay (GMLOS) as the value associated with the estimated LOS.
The GMLOS is based off of the patient’s diagnosis-related group (DRG), which is a system of grouping together clinically similar patients. Several different classification systems exist with varying levels of grouping precision and levels of specificity. For example, CMS utilizes MS-DRGs (Medicare Severity Diagnosis Related Groups) to assign a specific GMLOS to each DRG in their system.
When clinically viable, reducing hospital length of stay has been proven to provide both positive results for patients and financial benefits for the institution. In many cases, hospitals do not receive additional reimbursement once a patient’s stay has passed the GMLOS for their assigned DRG.
The DRG, calculated from the patient’s diagnosis, relies on accurate and thorough coding. One study estimated that 40.6% of patients in a specific facility could have benefitted from having a more accurate DRG assigned. Allina Health, a not-for-profit health care system based in Minnesota, has realized a financial gain of 13 million dollars by implementing a length of stay optimization effort across their system.
Several strategies, such as provider education on statistics and benchmarks, earlier discharge order entry, and increased case management and care coordination efforts, have proven successful for increasing the percentage of patients meeting - or beating - the GMLOS for their condition. However, providing real-time data to care providers stands out as one of the strongest ways to optimize for length of stay. If the care team and administration members do not have a way to access the comparative statistics while the patient is still in the facility, they cannot address this data during their clinical decision-making process, making it impossible to effectively target interventions.
Benchmarking LOS is most useful when the information is supplied in an on-demand manner, allowing educated care teams to use the relevant information when needed. Implementing a system that provides this feedback in real-time can be difficult and involves concurrent coding teams working together with clinical care teams.
Once the data is readily available, health systems can then act to improve it, taking on projects such as selecting specific DRGs to target for improvement and starting conversations between providers and coders when benchmarks don’t seem clinically accurate. In the end, the patient benefits from a more collaborative care planning initiative that takes full advantage of all available information and includes options for care outside of the inpatient environment.
pMD’s Charge Capture, Clinical Communication, and Care Navigation tools all offer the capability to give relevant information to caregivers at the right time, while avoiding cognitive overload. We’ve worked with enterprises to introduce functionality to support their quality initiatives with real-time data. If you’d like to discuss your organization’s current initiatives and goals around length of stay, reach out to us! We’re all ears.