As part of employee benefits, organisations right now intuitively zero in on a single health check package which they think fits all their employees and offer this for free to all their employees. This approach of One package fits all’ not only fails to be relevant to most employees but also increases the corporate’s wellness budget.
The two challenges involved in bringing a scientific method to solve this problem are: 1) Identifying employees who are at high risk to the chronic health disorders 2) Recommending the most relevant packages for a given employee’s health profile.
At eKincare, we solve this by implementing machine learning techniques namely decision tree classification models. An employee once registered, goes through a mandatory 2-min health risk questionnaire addressing specific questions having the highest impact on health disorders. Our model leverages over 6 Million data points to assign a risk factor to the user for different disorders. This approach identifies high risk employees who would require a further health screening and hence, recommends the right health package.
eKincare’s wellness solution is being deployed at one of the leading MNCs with over 30,000 employee base. Our models, help them identify employees with risk for metabolic disorders viz., hypertension, diabetes, cardiovascular and spend their budget effectively by screening employees for further health checks.
At eKincare, we continuously improve our models by adding more voluntary secondary questions and with rapidly growing user base our prediction accuracy also increases with time. We are also constantly expanding these models to identify risks for various other chronic disorders.