The life and health insurance industry are undergoing seismic changes as technology and digitization transforms the global marketplace. The new technologies that have emerged over the past decade have made it possible for established, traditional companies with core legacy systems, to consider new modes of operation. The most common usage of technology in health insurance space is to track customer behaviour is in the form of wearables. Most common wearables are devices that consumers can wear, like Fitbits and smartwatches, wristbands, smart ring, and chest wraps. The definition of wearables stretches to smart scale, CGM devise, metabolic panel,and wearables ECG Monitors. Most common health apps provide tracking activity levels, daily fitness endeavours, personalised diet and exercise plan and on-demand instant access to reputable nutritionists, fitness trainers, and yoga instructors to guide through customer’s fitness ambitions. The latest addition to capabilities in health apps is a Facial Bio Scan which provides key health vitas through a selfie in few seconds. The scope of behavioural data being captured through wearables have also significantly increased over a period. So, besides steps, physical activity etc, wearables are capable of tracking sleep pattern, heart rate, pulse rate, oxygen saturation levels, glucose levels etc. The core of this engagement through wearables is real time customer data.
Use of customer data to assess and price risk is happening more in Health Insurance space than Life insurance. This is because of the fundamental difference in Life & Health pricing approach. In Life Insurance we follow a level annual premium approach (premium doesn’t change throughout term of policy), whereas in Health Insurance, premium is age banded and can increases significantly with age. Insurers use customer data in 3 forms. A reduction in renewal premium (insurers offer up to 10-15% discount worldwide), cover boost (offer more cover at the same price), superior products to engaged and healthier customers are the some of the outcomes of this engagement.
Another major focus area for insurers is use of health & lifestyle data and scoring done from data sets as a proxy for traditional underwriting process. This reduces friction in the underwriting process and makes customer journey seamless. Most customers do appreciate this approach. A study undertaken by a global reinsurer state that, 80% of the customers are willing to share their personal (behavioural) data provided that the insurer gives a better price for their purchase and/or a friction less customer journey.
There has also been significant increase in use of Artificial Intelligence Applications in this space. Infact, Insurers use a combination of Artificial Intelligence (AI), Deep Learning (DL) and Machine Learning (ML) Models. A classical use case of artificial intelligence and machine learning in development and implementation of health insurance policy is Insurers identifying and designing a health insurance plan for an individual based on their historical data and current health conditions of the applicant. This helps the insurer to provide a customised health insurance plan instead of “one size fits all” approach. Also, customers are encouraged to choose a plan suitable to their needs and not pay for healthcare services which they might not utilize. IRDA’s sandbox initiatives have aided this approach in a big way.
AI-led premature mortality and chronic disease risk prediction models is the other key area HealthTech companies are working in collaboration with insurers. Though Cancer, Diabetes, Coronary Artery Disease and Cerebrovascular disease are the most worked upon areas, the models can be adapted rapidly to any disease or population group. Insurers use a multi-endpoint ensemble technique to deliver maximum outcome certainty. An important data point in this regard is, when diagnosed early, chances of survival in a cancer patient is 89% as opposed to merely 21% when diagnosed late. AI led Eye doctor can examine retina scans and identity Diabetes Retinopathy four to five years earlier.AL and ML can predict breast cancer from a mammogram as far as ten years in advance. Another interesting use case of AI (AI, ML & Image recognition) is analysis of 300 pages of attending physician statement and submission of machine-readable summary in less than 3 seconds.
Developing “transparent” AI models are important as insurers need to ensure that their algorithms are fair, unbiased, and explainable. Most common ways insurers do it is by researching the common causes of bias in population data, defining, and applying measurements of fairness, developing data collection and modelling methodologies aimed at creating fair algorithms and providing feedback to teams on how to regulate machine learning models. An important aspect to this exercise is continuous monitoring of parameters used in modelling.
In conclusion that more and more insurers are likely adopt technologies to stay relevant within the space and bring in innovation. The Life insurance industry will also undergo radical changes, not only because of new technologies but because of changing health demographics of the customer. Engagement through wearables will help to develop lifelong relationships with customers, and pricing approach and the nature of cover will change to reflect the health behaviour of the customer.