For insurance companies finding and building customer relationships and managing, risks are key to creating a growing, profitable business. Companies that are making extensive use of AI are reaping the benefits of increased customer satisfaction and loyalty while decreasing fraud which adds to their bottom line. These companies are using AI for a number of scenarios including risk management, fraud detection, customer retention, and optimized marketing.AI, is empowering leading insurance companies to deliver AI solutions that are changing the industry.
The insurance industry in Kenya is in a transition phase and will remain so for the next few years as Insurance companies implement IFRS 17 – Insurance Contracts. The standard was released in May 2017 and takes effect 1 January 2021, bringing with it significant changes in the industry.
Insurance companies will now have to re-look at their valuation models, the capability of current systems, and assess the changes to be made to ensure compliance with the new standard. Insurance Companies will require more data at a granular level to meet the new reporting requirements.
AI represents a big opportunity for the industry to grow its data analytics capability. Data analytics refers to the analysis of data with the aim of discovering existing trends in business data and drawing conclusions that inform future decision making. The risk – price relationship in insurance places data analytics at the very core of the insurance set up.
With requirements for increased data capacity and systems to match within the framework of the incoming IFRS 17, insurance companies will have at their disposal a treasure trove of information that remains very relevant in the running and growth of the business.
Auto insurance is an increasingly competitive industry; with customers more able to easily research and switch carriers than ever before, customer churn has become the new norm. A 2014 Global Insurance Consumer Survey by Ernst & Young found that around 40% of customers globally have left their insurer in the last 18 months. When customers are lost at such a high rate, even more money must be spent acquiring new customers to replace them. Acquisition costs are highest in the insurance industry, and in fact, it costs seven to nine times as much for an insurance company to acquire a new customer than it does to retain one. In this environment, it is essential to understand the customers and ensure they experience excellent service at every touchpoint. Major insurance companies provide millions of insurance policies to customers. Some insurance companies undertake customers’ surveys to provide feedback to the company. This Voice of Customer (VoC) feedback, which reflects the experiences, complaints, and suggestions of customers, provides the company with an excellent opportunity to understand the customers and discover specific customer complaints. Therefore, to continually improve products and customer service quality, and in turn, customer satisfaction and retention, the company monitors this feedback at the national and state level.
Insurance companies that will assess risk supported by insights from data analytics will, therefore, be ahead of the curve in complying with future potential guidelines from the regulator to clamp on price undercutting based on the shift in reporting requirements.
Fraud in the insurance sector remains a reality. The numerous players in a typical insurance contract do little to alleviate the risk of collusion. Through data analytics, Insurance companies can establish trends based on historical transactions and in particular loss-making contracts to assess for indicators of fraud. Trends and analytics will allow for comparability among intermediaries enabling insurance companies to interrogate the quality and cost of business from specific intermediaries.
Fraud remains a perennial challenge due to the lack of data sharing across insurance companies. Be it the non-existence of data, lack of uniform data, or the high set up and maintenance cost of a centralized set up to house and stream data to and from the various insurers, none of these challenges measures up to the ever-increasing cost of fraud year on year.
The potential gains of data and analytics in the insurance sector will require commitment and investment by players in the industry. Insurance companies face a significant challenge in the collection of relevant data in a usable format. The use of data and analytics will introduce the need for more advanced business intelligence tools. Insurers will have to invest in the training of resources capable of correctly analyzing, interpreting, and applying the data.
This shifting landscape of data in insurance heralds the dawn of a new age of opportunity, innovation, and differentiation whose impact will undoubtedly translate into gains for the insurer and insured.
Electronic Medical Records ( EMR) represent a wealth of information for hospitals and physicians in improving patient care and public health at large. A large portion of this EMR data is composed of doctor and nurse notes as well as lab test results, making these EMRs a valuable resource for correct diagnoses. Thus, correct and early diagnosis of diseases relies on careful scrutiny of patient EMRs. Typically, medical professionals are tasked with performing these manual reviews of EMRs. Often overworked, these medical professionals generally do not have enough time to scrutinize every EMR carefully. Hence, they may occasionally miss key signs and symptoms present in the EMRs, leading to an overall higher rate of misdiagnosis, delayed diagnosis, and litigation. Due to these factors, the hospital turned to automated text analysis for diagnostic decision auditing and support. Automated text analysis solutions often struggle to accurately analyze textual data due to the complexity of a medical text. EMR’s are written in a unique language style, which is a combination of the vernacular, scientific terms, medical jargon, shorthand abbreviations, and medical coding. For example, a patient complaining of abdominal pain could be recorded as Abdominal pain, bellyache, stomach ache, and pain, RLQ [Right Lower Quadrant] pain, LLQ [Left Lower Quadrant] discomfort
The complex and evolving nature of medicine itself adds to the challenge of automated text analysis with thousands of potential symptoms, diagnoses, medications, and lab tests. Furthermore, an intelligent text analysis solution must be able to perform context-based disambiguation of acronyms, such as for LBP, which can refer to “Lower Back Pain”, “Low Blood Pressure” or several other terms
An automated solution for the extraction of clinical findings from EMRs. Focusing on high-risk patient diagnoses, a system that extracts and structures the presence or absence of symptoms and findings that point to breast, colon, lung, and ovarian cancer in addition to lumbar disc disease, appendicitis, and myocardial infarction. Automated context-based disambiguation of medical acronyms and diagnostic coding can be achieved through machine learning of hand-annotated medical records as well as from medical journals and articles. A predictive model to determine the most likely expansion based on the words in context surrounding the acronyms. For example, LBP is more likely to be interpreted as “Lower Back Pain” if “back”,” pain”,” leg” or “numbness” is mentioned in the surrounding paragraph. Interpretation of vital signs and lab tests. AI Solution can be easily trained to recognize dozens of lab tests utilized by the hospital. Using metrics provided by the hospital and statistical analysis, the system interprets the data and indicates the presence or absence of one of the diseases in focus. true or negative findings. In addition to the extraction of the key clinical findings,
In conclusion, the benefits of AI solutions include; Earlier, more accurate diagnoses. The consistency and transparency of the solution system enhance diagnostic efficiency by quickly and accurately capturing indications within text notes that might otherwise go unnoticed by medical professionals, thereby reducing the number of misdiagnoses and delayed diagnoses. For diseases such as cancer, early detection increases the chances of patient survival significantly. Financial Savings.as a result of reduced medical errors, the solution also decreases the chances of malpractice litigation against the hospital that is engendered by misdiagnoses and delayed diagnoses while improved patient satisfaction and brand reputation. Reduction of diagnosis errors, and thus, bad patient experiences and medical expenses, will have a positive impact on the overall customer satisfaction and brand reputation of the hospital and medical Insurer