Management of risks associated with Length of Stay (LoS) is a common business challenge for multi-specialty hospitals, hospices and insurance companies, because there is unpredictable cost and capacity management involved in it. Despite integrating modern technology into operational systems, assuaging key LoS risks is a huge challenge for these organizations. Often, it is the unpredictable cost involved that acts as a bugbear for them.
Apart from capricious cost per treatment factor, there are other risks involved too, such as event risks that lead to early discharge or overstay or re-admittance, or ‘Knock-On’ effect of overstay patients and more.

To elucidate further, picture this: Two patients, namely Patient A and Patient B, have approached a hospital for treatment. Both the patients have been medically diagnosed with Diabetics and have been advised surgery at the hospital due to some medical complications.

(To keep the use case simple, we have considered only two patients and that the hospital has only one bed.)

To provide unbiased care, hospital plans Patient B’s surgery five days ahead of Patient A’s surgery, because patient A’s estimated LoS is five days from the diagnosis perspective. Due to certain medical condition there are high chances of Patient A becoming an overstay-patient. This disrupts admission of Patient B. So the knock-on effect of Patient A becoming an overstay-patient is that the surgery of Patient B will have to be postponed. Not only that the cost incurred by both the hospital and the insurance company would be higher. Above all capacity management goes for a toss.

The best way forward to address such issues and to bring in steadfast operations would be through modeling LoS and predicting risk of over stay. Aptus Data Labs has devised a diligent ecosystem of advanced analytics to model length of stay, using decision tree or SVM (support vector machine) model.

In the above scenario, on the actual admission date of Patient A, an ADT (Admission, Discharge, Transfer) message is sent from EPR to other systems like radiology department. The ADT message is then communicated with the Aptus Data Labs’ Analytics Accelerators solution. Subsequently, data of Patient A is compared with different profiles available in the Analytics Accelerator solution to obtain an APR (Aptus Data Labs’ Patient Risk) score. Refer Figure 1, which depicts Risk Score and LoS for many patients. The Risk Score is used by hospital staff and case workers to follow a patient more closely, to take evasive measures as required. Consequently, Patient A can be treated in a 5-day period, and can be discharged before Patient B comes into the hospital.


With the help of Aptus Data Labs’ LoS solution and Analytics Accelerator solution, hospitals and insurance companies can gain key insights related to overstay data (descriptive) as well as the real-time risk status of an admitted patient (real time matching).

Above all, Aptus Data Labs’ solutions help in predictive bed management and waiting list reduction to enable increased resource utilization and enhanced capacity planning.

To summarize, Aptus Data Labs’ LoS solutions:

  • Ranks inter-disciplinary treatments that lead to uncontrolled costs
  • Identifies patterns that impacts LoS
  • Identifies patterns of re-admittances
  • Classifies Risk Profiles (RAG) and matches patients’ medical history and diagnosis with identified Risk Profiles
  • Avoids knock-on effect of overstay patients

These features incentivize diligent management of Length of Stay (LoS) risks on the work floor, thereby bringing in augmented inter-disciplinary care and advanced point of care for patients, and peace of mind for hospitals and insurance companies.