You’re spending HOW MUCH on overhead costs?!?

What is your hospital currently spending on overhead costs? 10% of your budget? 20%? Chances are it’s closer to 25%. According to a study by The Commonwealth Fund, administrative costs account for 25.3% of hospitals budgets.  This is money that does not typically improve patient care, but rather is associated with IT, scheduling or billing. There are multiple ways to reduce these costs by using different AI or IoT methodologies.

IT Storage Costs

In 2010, the average hospital was managing one billion terabytes of data. In 2020, it’s expected that it will have increased by 50 fold.  However, many times this data is stored in multiple locations and in multiple formats in numerous data warehouses. In the majority of cases, patient data that is over 7 years old is rarely referenced, however, hospitals are extremely reluctant to archive patient data offline, because having this data provides a healthcare professional with the complete view of the patient, which is can be helpful during treatments of seriously ill people or people with chronic diseases.

In order to reduce data storage costs, hospitals could implement data lakes, which are a more flexible and less costly solution for data storage.  When a traditional data warehouse is created, schemas need to be created first and then again when the data is being exported. When new data sources are added to a data warehouse, the schema then have to be modified. Whereas, data lakes only require schemas to be created when you’re ready to build a solution – thus, cutting the development by at least half. Also, data, regardless of format, can be easily added to a data lake. This can be especially useful since many times a patient record consists of free form text in the provider notes, structured data like height and weight and many times medical images.

Possibly the biggest area where hospitals could see the cost benefit of using a data lake is with medical images. For example, 1 MR scan of 150 images averages 30MBs of storage. If that machine runs 22 scans a day six days a week, that’s 205 GBs of data every year – just for that one machine. Now let’s take into account all of the MRIs, CTs, X-Rays, and ultrasounds that are in a hospital and you’re easily looking at hundreds of terabytes of data that are being produced each year. Storing this amount of data indefinitely can be extremely costly. With a data lake, you would still maintain real-time access to all of these images, but t a Hadoop based data lake has the potential to reduce the amount of storage space necessary by 50-75%, which could translate to tens of thousands to hundreds of thousands of dollars saved each year.

Optimizing scheduling

Did you know that it takes the average new patient in a major metropolitan area 24 days* to see a provider? And if they’re in a mid-sized city, this time increases to 32 days.  This can lead to patient frustration and poor outcomes.  

Often these long wait times are the result of scheduling inefficiencies in a practice.  During the time when doctors and staff aren’t being utilized providers still have to pay for the fixed costs of running an office. To make sure these overhead costs are justified, scheduling needs to be optimized so less time is spent on finding an appropriate appointment, matching the patient to the right provider and incorrectly utilizing a provider’s time.

However, scheduling can be extremely difficult to optimize manually. Many times, the scheduling coordinator just looks at the next available appointment and puts the patient in with the next available doctor.  This can lead to a variety of problems. Many times, high value time slots go to one time patients, patients who could have seen a physician’s assistant or are better off seeing another doctor, or patients who were insured by companies that have notoriously low reimbursement rates. All of these factors contribute to lower revenues than could be generated if these slots had been filled by patients with recurring appointments, patients who actually needed to see a doctor rather than an assistant or patients insured by companies with higher reimbursement rates.

Optimization of the scheduling process can also show office managers the best days the most popular locations, times and days for patients. This can be combined with the providers’ preferences to make the most revenue generating schedule possibl. For example, if there are a large amount of requests for Tuesday appointments, then providers should be on-site then. Walk-in hours or the number of reserved same-day appointments could also be reduced on Tuesdays to account for the popularity of Tuesday appointments.  On the otherhand, maybe a very small number of patients request appointments on Wednesday mornings. This could lead to possibly walk-in hours with a skeleton staff or the office being closed on Wednesday mornings.

Overall, by optimizing scheduling of patients and providers, medical practices can make sure that the right services are provided to patients while ensuring that overhead costs are minimized.

 

Hospitals today are under significant financial pressure. They want to maintain or increase the level of patient care while reducing costs, but this can be quite challenging. However, by reducing overhead costs, hospitals can lower costs while maintaining – or sometimes increasing – the level of care they provide. So, hospitals, look to technology to help reduce waste in your back end system, so you can go ahead and buy that MR machine that you’ve been eyeing!

For more information on Entrigna’s medical image data lake, The Image Vault, or how AI can help you reduce overhead costs, please visit our website or contact us.