Delivering on the promise of making things better

Enabling insight-led enterprise change in healthcare


Against the backdrop of the worst global pandemic in a century, unlocking efficiencies in the healthcare sector is becoming increasingly key. Patient care is paramount and the availability of healthcare workers to provide that much-needed care is crucial. But how can healthcare workers focus on what they need to do if they’re manually capturing the same information up to 32 times? How can nursing staff provide care to critically ill patients – while juggling Covid protocols – when they spend 92% of their time on administrative tasks?

Simply put, they can’t. But that doesn’t mean efficient healthcare is impossible. By tapping into the wealth of data available to healthcare providers, hospital groups and medical aid schemes – and where possible combining that data to create a holistic picture – the opportunities for operational efficiencies are everywhere. 

Making an impact quickly. And cost-effectively.

A value case is a rapid proof-of-value exercise, designed to address a specific business need using a small multi-disciplinary team, and delivers tangible business benefit. What this translates to is a cost effective way to demonstrate return on investment to cost-conscious executives. 

BSGs layered iterative approach

The claims’ management process is a known area of inefficiency and a major headache, both for healthcare providers and patients – a perfect place to prove the value of insight-led operations. 

Upon discharge, patients are at their most vulnerable, and receiving a large unexpected bill leaves a very sour taste – especially as hospital stays require pre-authorisation, so you assume you’ll be covered. But, we rarely read the fine print. For example, certain plans only cover one blanket per patient, per night, so if your claim is submitted with two blankets, the medical aid might reject the claim in its entirety. These rejected and short-paid claims account for up to 30% of all billing and take up to 10 times as long to resolve, resulting in hundreds of wasted hours and millions of Rands in lost revenue.

Rejected and short-paid claims account for up to 30% of billing and take up to 10 times as long to resolve

By flagging claims likely to be rejected before submission, healthcare providers could significantly improve outcomes of the claims process. Working with a leading South African healthcare provider, BSG built a predictive model using machine learning to identify claims with a high likelihood of rejection. Over one and half million cases – roughly 80 million line items – were fed into the model. This helped it to understand reasons for rejection, enabling proactive management of cases. 

Significant Shift

The high volume of claims rejected prior to the implementation of the model, meant that any rejection “trends” identified were based on manual interpretation and intuition. Being able to accurately monitor trends meant the healthcare provider could also identify shifts in trends, and proactively adjust their processes to manage those shifts. This approach resulted in a saving of R18m over 7 months for the provider, and translated into a smoother process for both the medical aid schemes and the patients.

For business data to be valuable it must be analysed and insights drawn from it. And even then, if those insights are not used, it’s wasted effort. By consolidating individual data points into a more complete and trusted data asset, rich and powerful features are made available. 

Regardless of your industry or type of business, your existing data can unlock operational efficiencies, opportunities for proactive servicing and automation, and even predictive operations. 

Get in Touch

If you’d like to find out more about how BSG can help you enable insight-led business operations – get in touch.

Recommended Posts
Contact Us

We love talking about change. If you've been inspired by anything you've read here, please get in touch.

Not readable? Change text. captcha txt
BSG Opinion: Using machine learning to triage patientsalgorithm-driven financial services