Prior Auth - A Burden to lessen?

What is a Prior Auth?

It's a request from a provider to a health plan to obtain authorization for healthcare services or a response from a health plan for an authorization.

A report from CAQH have said that in 2021, Prior Auth Process is often cited by providers as a major source of administrative burden, since it has been one of the most costly and time-consuming transactions to conduct among those studied.

• To help reduce administrative burden on strained staff during COVID times and to provide timelier patient care during the early months of the pandemic, requirements around prior authorization was suspended or waived off.
• These suspensions and waivers, along with reductions in medical services and freezes on elective procedures, resulted in a 23 percent decrease in prior authorization volume.

However, the same report in 2022 has a different picture to share.

• Volume increased considerably (~61 percent) as restrictions placed on prior authorizations to help ease administrative burden during the COVID-19 outbreak were lifted.
• Elective procedures, which were postponed due to the pandemic, were rescheduled as doctors and nurses had now time to conduct these procedures and beds occupied by COVID patients became available.
• While volume rose for all three modes, portal use experienced the largest increase in volume among providers in comparison to Fully Electronic & Fully Manual and accounted for almost 40% of the overall volume—a proportion that has increased over the last three years.

What is the Time Saving Opportunity here?

• Move towards Fully Electronic Mode is the next step for the holistic industry.
• The increase in automation will result in effective time savings for medical providers. Providers saves on average, 11 minutes by conducting a prior authorization using the HIPAA mandated standard as opposed to doing the same process manually.
• Deemed as one of the most time-consuming administrative transactions by many providers, time associated with conducting a prior authorization manually (20 minutes), via a portal (12 minutes) and electronically (nine minutes). We have to ensure that we narrow the delta between Manual & Electronic in the coming period.

Leveraging AI in Prior Authorization can improve the outcome holistically. Let's understand that with a case study:

Suppose, A healthcare organization is facing challenges in efficiently managing the Prior Auth Process. The inadvertent delays and uncertainties are impacting patient care and increasing administrative burden. The organization decides to implement Predictive Analytics to enhance the Prior Auth workflow.

So, as a Product Manager / Leader, what should be the implementation steps:

1. Gather historical Prior Auth data which includes patient data, medical history, kind of treatments provided and probably some medical trends across different healthcare customers.
2. Clean and organize the information & lookout for inconsistencies or missing information. If there are any standardization required in data formats (this part is very crucial) and ensure compatibility for analysis.
3. Identify relevant features influencing Prior Auth outcomes which can be like demographics, certain medical conditions, treatment specifics.
4. Use some statistical analysis to prioritize important factors leading to aforementioned features.
5. Then comes the important part i.e. Model Development. Since the task is around Predictive Analytics, one must utilize machine learning algos viz. Decision Tree Analysis, Random Forests or even something tougher like Neural Networks. I have no reservations around a model, but this is important for the AI engineer to train a certain model on historical data especially emphasizing approved and denied prior auth cases & reasoning behind them.
6. Once the model is trained, it is time for Validation. Validation is an important step because it ensures that it generalizes well to the new cases as well.
7. Based on the outcome, it is important to fine tune the model parameters.
8. Last step would be to integrate the Predictive Analytics model into Prior Auth workflow & see if it works for the new Prior Auth Cases & compute it exactly how you need.

If this kind of process be in place, this is going to give spectacular results in long term, like:

1. Early Identification of High-Risk Cases and removing the complications around Prior Auth Process to ensure improved patient care and satisfaction.
2. Healthcare Providers can allocate resources more efficiently, focusing additional attention on cases predicted to face challenges.
3. Improve & Streamline Prior Auth Workflows.
4. Overall reduction in Processing Time. May be much lesser than 9 minutes (how it was in 2022).

This is not a straightforward implementation for healthcare payers & providers. But many organizations have started taking this pretty seriously & the coming years will be a testimony for this implementation. A report estimate says that it can save $449 million a year for medical industry.

What are your thoughts around Prior Authorization Process? How do you think you can simplify the existing manual time-taking process? Please put your thoughts in the comments section.

Thanks,

Gourab