Data science tackles a billion dollar healthcare problem

A leading provider of healthcare analytics and technology-enabled services, Intermedix, has developed a solution to address no-show patients using Dataiku Data Science Studio (DSS), its predictive analytics and data science platform.  The problem of patients missing scheduled appointments costs the healthcare industry billions of dollars of lost revenue each year. 

The predictive analytics software solution developed by Intermedix was built, tested and deployed by a small team of data scientists in just one month using Dataiku DSS and is now being used in more than 50 private clinics across the US. 

The unfortunate reality is that patient no-shows in the healthcare industry are extremely common, and industry-wide, this adds up to billions of dollars of losses each year. The long-term effect of this phenomenon is lowered reimbursement for providers and negative impacts on adherence, quality, and clinical outcome measures for patients.

More and more organisations are turning towards advanced analytics to reduce the probability of no-shows and their associated costs using heterogeneous data to optimise scheduling systems.

The inability of healthcare organisations - big or small, public or private - to deal with the no-show issue has had a profound effect on patients’ medical health and on providers’ financial health.

Studies have shown that 5 – 10% of patients miss scheduled appointments. US primary care physicians lose an average revenue of $US228 for every no-show, and lost revenue for specialists is even higher. In addition, overhead costs including staffing, insurance and utilities remain on the books.

Cancellations with primary care physicians also impact the number of necessary specialist referrals those physicians can make. Combined, these factors contribute to significant revenue loss for physicians associated with patient no-shows.

A Predictive Solution

Intermedix decided to develop and operationalise a no-show predictor that would assist local office managers in reducing the number of patients who miss appointments. The data science team set up Dataiku DSS to ingest and crunch historical appointment and demographic patient data.

From there, they built a predictive model that scores individual patients based on the probability that they will miss a scheduled appointment. Dataiku DSS automatically sends this output to the office managers at regular intervals customised to their practice’s needs.

Thanks to the predictive report, local office managers and schedulers can make informed decisions on scheduling and proactively target reminders to the patients most likely to miss their appointments.

Typically, developing and deploying such an application to cover site-specific patterns would take more than three months. Equipped with Dataiku DSS, Intermedix’s data science team was able to prototype and deliver the solution to more than 50 clinics in just one month.

“DSS slashed the amount of time it took to analyse our data, produce a working model and deploy a solution, all while improving the accuracy of our predictions,” said John Enderele, data scientist at Intermedix. “The platform will enable us to more rapidly identify our clients’ needs and respond with innovative, data-driven solutions to make them successful.”

Intermedix’s solution was made possible by the technology behind Dataiku, maker of the predictive analytics software platform Dataiku Data Science Studio (DSS). Dataiku DSS makes it possible for organisations to reap the benefits of data science thanks to a collaborative interface for both expert and beginner analysts and data scientists.

Dataiku offers a complete and accessible advanced analytics software platform that allows teams made of different skill sets to streamline the process from raw data to predicted output, all in one tool.

Dataiku DSS can be used to quickly build predictive services and data products that transform raw data into business impacting products including:

  • Churn Analytics
  • Fraud Detection
  • Logistic Optimization
  • Data Management
  • Demand Forecasting
  • Spatial Analytics
  • Lifetime Value Optimization
  • Predictive Maintenance
  • and much more