This is Part 3 of a 3-part series. Part 1 can be found here; Part 2, here. In our first blog post on artificial intelligence (AI),  we outlined the challenges of defining AI, and in our second blog post, we described how ubiquitous AI is becoming, defining it as "ubiquitous intelligence." In this post, we define the continuum of AI as "AAAI": assisted, augmented and autonomous intelligence. AI as Assisted Intelligence Over the past couple of decades, AI has replaced many of the repetitive and standardized tasks done by humans. For example, industrial robots are tackling many manufacturing tasks. Similarly, many administrative tasks such as taking meeting minutes, answering phones and searching for information are all done by some form of an automated system. We call this type of automation — where the AI is assisting humans to do the same tasks faster or better — assisted intelligence. The humans are still making some of the key decisions, but the AI is executing the tasks on their behalf. The decision rights are solely with the humans.  AI as Augmented Intelligence We are just now moving to the next stage of augmented intelligence, where humans and machines learn from each other and redefine the breadth and depth of what they do together. For example, in a recent client engagement, we carried out 200,000 go-to-market scenarios generated by an AI system for a service introduction. This provided the human decision-makers with a high degree of granularity and specificity regarding the assumptions, future projections and impact of the new service. While the system learned a lot and modeled the ecosystem, the humans saw the sensitivities and feedback involved in market adoption. Under these circumstances, the human and the machine share the decision rights. In addition, unlike assisted intelligence, in augmented intelligence, the nature of the task fundamentally changes. On a spectrum ranging from no automation to total autonomous operation, each sector, company and individual will set the appropriate level of machine augmentation. Over time, the dial might move more toward totally autonomous, or it might stay somewhere in between. AI as Autonomous Intelligence Lastly, we see autonomous intelligence, in some cases, where adaptive/continuous systems take over. They will do so only after the human decision-maker starts trusting the machine (e.g., fully autonomous self-driving cars), or when the cycle time of decision making is so fast that having the human in the loop is a liability (e.g., automated trading). In autonomous intelligence, the decision rights are with the machine and are fundamentally different from assisted intelligence. See also: Of Robots, Self-Driving Cars and Insurance The decision to move from augmented intelligence to autonomous intelligence will largely be in our hands and will be made based on a number of different factors — including the speed of human decision making, the technical feasibility of making autonomous decisions, the cost of building solutions and the trust we place in these solutions. As enterprises contemplate the introduction of AI across their functional areas, it helps to clearly articulate which stage of AI they are aiming for. Are they merely automating repetitive tasks and providing assisted intelligence? Are they fundamentally changing the nature of work by having humans and machines collaborate with each other to make decisions with augmented intelligence? Or are they delegating all decision making with autonomous intelligence?

Remember the classic '80s film “War Games,” where the computer system (named War Operation Plan Response, or WOPR for short) asks Mathew Broderick in a See’n Say computer voice, “Shall we play a game?” The movie was a tense thriller that was topical for my Cold War childhood, but it showed, among others things, that not all games are fun. Insurance claims should not be a game — where one side is playing games as a tactic to delay or reduce claim payment. Unfortunately, I see this sort of ambush all too often when preparing property and business interruption claims for my clients. My client is usually struggling to recover from a major loss to its business. My client has done its best to protect itself via loss control, insurance procurement and a proper claim filing. So when the client documents losses and presents the claim, the client should not have to battle bullying, stall tactics and misguided theories. As an example, a chemical company client's business depended heavily on the supply of raw materials from specific international locations. The exclusive relationships with these international suppliers and their governments took decades to forge and represented a distinct competitive advantage. The company's business was cyclical, and, during a low point, a hurricane devastated its manufacturing plant. If it was not able to get back up and running quickly, the long-term contracts with suppliers would be canceled, undoing years of supply chain efforts. The CEO recognized the real possibility that his company would not recover if its insurance didn't reimburse the company for its losses in a timely manner. The CEO knew he might have to lay off more than 1,000 employees. The insurer chose to overwhelm the client with requests for information, including sending a large group of insurance investigators that my client had to accommodate at his chemical plant while it was still under water. Contractually, the insurance company has the right to gather information. However, tact and decency should be observed. The insurer leveraged the policyholder’s crisis to establish reasons not to pay the claim based on misguided theories about the business. Because of the chaos, when the insurance consultants arrived to survey the damage, the client was not prepared. The company did not have a proper escort for the consultants, who made incorrect assumptions about the damaged equipment that formed the basis for a theory on the valuation of replacement equipment and lost production. See also: Power of ‘Claims Advocacy’   The client was not informed about these conclusions until some months later. The company was ambushed and had to disprove the incorrect information. All the while, claim payment was being withheld. This tactic is common. While the insurance team may just be doing its job as instructed, a company’s existence is at risk. Adjusters are often cavalier about this process and will hide behind the “duty” or policy wording while, in reality, just playing games with the money owed to the policyholder. I am happy to report that we were able to secure advanced payments to stabilize the client’s operations, maintain supplier relations and create an equitable settlement. The success required an effective strategy and careful execution. Here is the approach we know works best:

1. Take Control — You do not want to put off the insurance company too long, but it is okay to let the company know you are going to control when its people get access and whom they can interview. Claims are often derailed in the first week because of uncontrolled access and miscommunication. 
2. Agreements in Real Time — One of my favorite risk managers uses this mantra during claims: “We make decisions in real time.” What he means is that when confronted with a decision — say the rebuilding or replacement of equipment — you must use all the information you have at that time to make a decision. As long as the adjuster is aware of the decision and your reasoning, he should not second guess what you have done once more information is known. For example, immediately after a loss, you think you need two cranes to move equipment and debris. After the fact, you realize you could have gotten by with just one. You made a decision based on what you knew; if the adjuster does not object at the time of the decision, he has no grounds to object after the fact.
3. Clarify Requests — The insurance company is going to ask for information — a lot of information. In general, these requests are broad and may even be used to fish for something that can be used against the claim. Don’t let this happen. Ask that requests be specific. If they are not specific, send the request back. Ideally, claims are presented with supporting documentation, and that should be the focus of requests. I often filter this information down to what is really specific to what is being claimed. Extraneous information can create confusion and can lead to more requests. Your loss accountant, if she is experienced, will help interpret requests and focus on what is relevant to the claim.
4. Recruit Experts  — Adjusters and their team work on claims every day. It’s their full-time job. For you, it is an infrequent part of your job. If you want a smoother process and a positive outcome, you need experts working on your behalf. In addition to your internal team, your broker's claim experts (as well as independent forensic accountants, engineers and outside counsel) are critically important. Ensure that those on your team are working on your behalf and match up well against the insurance company representatives. In my claim example above, we were not engaged from the onset; it is vital to have your independent team vetted and agreements in place ahead of a loss. Remember, as in my example, many claims are hindered by mistakes made in the initial weeks after a loss. Immediately after a loss is no time for shopping.
5. Don’t Play Games — In other words, focus on the claim, not the games. Prepare an accurate claim from your perspective; be up-front with relevant information; and be reasonable in final negotiations. Just because insurers may play games doesn’t mean you need to do the same. You are much better off being prepared, professional and confidently in control of the process.

See also: The Future of Insurance [Infographic] If you follow this advice, you will stand a much better chance succeeding with claim recovery. As WOPR realized in the movie, with claims war games there are no winners.  Avoid this ambush by being prepared and informed.

Bob Kaplan deserves our respect. Famous for his contribution to management with the balanced scorecard, he is now senior fellow and Marvin Bower professor of leadership development, emeritus at the Harvard Business School. (I have never had the privilege of meeting him.) His colleague, Anette Mikes, was with him at Harvard, and she is now professor of accounting and control at the University of Lausanne (HEC). I am in a network of risk practitioners and thought leaders that includes her. (I have heard her speak but have never met her one-on-one.) She has made important contributions to the academic study of risk management that includes a case study of John Fraser’s Hydro One and a similar case study on Lego. I have shared my thoughts with her on the narrow and highly limiting view that risk management is about mitigating potential harm from adverse events. Unfortunately, I have not been persuasive. Kaplan and Mikes recently published a Harvard Business School working paper, "Risk Management – the Revealing Hand." While there is some value in the paper — such as its insistence that risk management must be continuous as well as its discussion of overreliance on models — it demonstrates very clearly why so many board members and executives do not see how the management of risk enables their organizations to set and deliver on objectives and strategies. For example, the ERM Initiative at North Carolina State University, in its 2016 survey of the state of risk management, found that only 4% of organizations feel their risk management is very mature (up from 3.4% in 2010). In 2013, a Deloitte survey found only 13% of executives believe risk management supports their ability to develop and execute on business strategy very well. See also: How to Remove Fear in Risk Management How can risk management practitioners demonstrate value and significantly contribute to the success of an organization when they:

• Focus on a list of potential harms;
• Don’t focus on enabling intelligent and informed decisions from strategy to tactics; and
• Talk in technobabble instead of the language of the business?

I see risk management as about the following:

• Enabling informed and intelligent decisions that consider what might happen, both good and bad. Those decisions include setting the vision for the organization (including its strategy, plans and objectives) as well as the decisions made every day across the extended enterprise as people at all levels direct and manage the organization toward its objectives.
• Thinking about what lies between where we are and where we go, how it might affect our ability to achieve or exceed our objectives and what (if anything) we need to do about it.
• Taking the right level of the right risks. We cannot survive, let alone thrive, if we do not take risk. The concept that we must mitigate all risks is absurd. Risks need to be assessed in the context of achieving objectives, not in a silo.
• Knowing how to evaluate the potential for any event or situation to have good, bad or a combination of good and bad effects — and providing a structured process for making decisions about the path forward.
• Promoting intelligent and effective management that enables the organization to succeed.

Kaplan and Mikes say there has been no credible academic study that demonstrates that risk management delivers tangible value. (Note: EY and Aon have released studies that say that organizations with better risk management obtain better long-term financial results.) Is the conclusion by Kaplan and MIkes because they don’t understand what risk management should be, that it is not about managing a list of potential harms (what Jim DeLoach calls "enterprise list management")? Focusing on what could go wrong will not help you do what is needed for everything to go right. If you were greeted at your front door by someone with a list of all the bad things that might happen, would you ever go out, or, would you dismiss the pessimist with disdain? Here are just a few quotes to support my view:

• “Enterprise risk management helps an entity get to where it wants to go.” – COSO (the acronym for the Committee of Sponsoring Organizations of the Treadway Commission, which published "Internal Control—Integrated Framework" in 1992).
• "[Risk management enables] a greater likelihood of achieving business objectives [and] more informed risk-taking and decision-making.” – COSO
• “The purpose of managing risk is to increase the likelihood of an organization achieving its objectives by being in a position to manage threats and adverse situations and being ready to take advantage of opportunities that may arise.” – National Guidance on Implementing ISO 31000:2009 from NSAI in Ireland
• “We believe a paradigm shift in risk management is beginning, which is tied to the increasingly complex world in which companies now operate; based on the awareness that uncertainty is embedded in [and affects] everything we do; [and] focused on both capturing upside opportunities as well as protecting the business.” – EY
• “You need [risk management] to become part of the rhythm of the business — meaning within the flow of strategic and business planning, operations, oversight and monitoring that runs from the board to the line.” – EY
• “The job of risk (management) is to make … executives more confident to take strategic risks; to demand objectivity in decision-making; and to focus on value added, not just value preserved.” – Deloitte

I can tell you that the risk management programs at Hydro One and Lego do not limit their work to potential harms. They consider the potential for reward as well as harm and work to help management succeed. See also: Moving to Real-Time Risk Management So how is it that Kaplan and Mikes have such a narrow view? Perhaps it is because the great majority of practitioners limit risk to the negative and their practice to a periodic review of a list of top risks (enterprise list management). That narrow view inevitably creates a disconnect with the desire of management to lead their organization to success. How do you expect a CEO to believe risk management enables success when all the chief risk officer (CRO) gives him is a list of what could go wrong? The CEO needs help to see what might happen, both good and bad, and what to do about it. In other words, the CEO needs to see risk management as helping him or her get where he or she needs to go. Do you share my view? If so, how do we convince both the practitioner and academic community? How can we move the practice forward so that it is recognized by leaders of every organization as contributing to their success? I welcome your views. This article was originally posted here.

“Who is that guy, and what’s he doing?” Shaun called me, laughing. He explained that he had just heard about a teenager who was at his friend’s house. As they walked through the den – he saw an older man reading a newspaper in a recliner and asked the question above. The man’s son said, “That’s my dad, and he’s reading a newspaper.” The next question was, “What’s a newspaper?” followed by, “Where does he get it?” The son apologized for his dad’s eccentric behavior by explaining that there are stories about news, politics and sports in the paper. It's like what we can read on Google, Facebook, Twitter, Snapchat, Instagram or WhatsApp, or view on YouTube. Every morning a man drives by the house and throws a paper (usually in a plastic bag) out of his window and into our yard. Rain or snow, sleet or shine, dad walks outside to get it. Then he comes in and reads it like he’s doing now. “Why doesn’t he have a smart phone? What’s wrong with him?” the friend asked. These questions may shock those of us who walk to the curb at 5:00 every morning, anxiously awaiting our daily delivery. The newspaper is important to me. It is more than news or journalism. It is a ritual in my life, a daily ritual I’ve enjoyed for more than 50 years. To the teen above, my ritual probably is crazy. To me, he seems stupid. In yesterday’s world, I’m right. In tomorrow’s world, he is. Consider the profits made by the publishers of newspapers in yesterday’s world. Consider the Big 3 broadcast channels, CBS, NBC and ABC, and their dominance in the media world. Now recognize that they are dinosaurs – smoking around the tar pits while awaiting their demise. Is your agency or carrier or brokerage selling yesterday’s products to a population that is the past, or is your agency a living system that is growing with the marketplace as it will be and adapting what you sell and how you sell it to the buyers of the future, or are you focused on your reminiscences? What populations/niches will you serve in tomorrow’s world? What will they be buying? How will you deliver it to them profitably? I can promise they will be different than I am, and you must be different than you are – if you don’t want to go extinct! Will your model be paper or virtual? Think new! Act now!

Healthcare Matters sits down with Dr. Richard Anderson, chairman and CEO of the Doctors Company. In Part 5 of the series, we discuss with Dr. Anderson the most effective provisions of the Medical Injury Compensation Reform Act (MICRA) and how MICRA has shaped the medical malpractice insurance climate in California since 1975.

In the great tradition of discoveries such as germ theory, X-rays, DNA and penicillin, the next stage of our growth in healthcare will come from artificial intelligence (AI). This includes effective machine learning and neural networks and a promise of greater pattern recognition, data analysis, "general thinking," decision-making and efficiency.   Until recently, a large percentage of data was not digitized, nor did we have the storage and processing power. Now that we have this capability, we will soon see significant steps forward in AI innovation, growth of neural networking and deep-learning technology, including:

• Medical imaging and diagnostics;
• Wearables;
• Mental health;
• Virtual assistants;
• Risk management;
• Drug discovery;
• ER and hospital monitoring;
• Health and lifestyle management;
• Biotechnology; and
• Genomics.  

Take a moment to see what's on the horizon: smart visionaries and emerging companies such as  Merge Healthcare, Zephyr Health, Lumiata, Ginger.io, Apixio, Zebra Medical Vision, Babylon Health, Sentrian, AICure and HealthAware. Venture funding will continue to increase in the digital health sector, with respect to dollars invested and deals made. Look to see greater M&A activity in many of healthcare's largest industries. These initiatives appear to be putting our country's health in a great place for the future. Many of these efforts are intertwined with the healthcare's industry's efforts to capture the Triple Aim: improving the patient's experience with care (quality and satisfaction), bettering the health of populations and reducing healthcare cost.  See also: How to Think About the Rise of the Machines Thanks to abundant processing power that was previously available only on the world's most robust supercomputers, AI is developing faster than most predicted. As medicine faces many existing and new challenges, it will seek to address many areas in healthcare rife with inefficiencies and waste. By the end of 2016, the healthcare sector is set to have the largest workforce in all U.S. industries — including government. At nearly 19% of the U.S.' GDP (and growing), healthcare must recognize that technologies such as AI can serve a major purpose in improving human-powered efficiency.   When leaders think about AI in healthcare, they are immediately drawn to innovating, reducing medical error, improving management of patient health and making discoveries in drugs and biotechnology. But the greatest benefit may lie in the reduction of human capital, a large direct and indirect contributor to cost. Many medical professionals fear that, because of AI, they may lose their jobs. Early AI companies, especially those courting large players such as IBM, recognize this. These companies are not keen on holding hands and have been assuring medical clients this will not happen. Clearly, we must walk before we can run, and augmented intelligence, as a form of AI, is this needed step.   However, our healthcare and political leaders must recognize the importance of the last leg of the Triple Aim. Until cost drops significantly, we will continue to see great financial strain on many Americans. Outcome-based payments are set to help with this financial burden, though they might not be enough. Publicly owned, for-profit companies need to satisfy their shareholders. Besides, a recent Johns Hopkins study put medically caused deaths as the No. 3 killer in America today, at nearly 251,000 deaths (nearly 10% of the total), showing that even the best-intentioned and -trained humans have limitations. When AI truly arrives, we must make sure our healthcare and political leaders are held to doing what is best for the health and well-being of the American workers and taxpayers. If greater efficiencies and accuracy through replacing jobs with AI can lead to the Triple Aim and greater healthcare system sustainability, we must not be afraid to move forward. Healthcare is not only about business and profit but about serving what Abraham Lincoln called "the better angels of our nature" by passing on a better system to the next generation.

It is estimated that commercial property writers lose out on more than $7 billion annually in business interruption insurance, a line that could deliver increased and sustainable earnings upside annually but has often struggled to do so. This loss represents not only undervalued policies, but also income lost because of premium calculations that are not commensurate with risk. As with other property business, the No. 1 culprit is the decades-old difficulty that insurance companies’ face establishing adequate coverage limits for property lines -- and business interruption insurance (BII) often has worse results than insurance on buildings and contents. For the past 10 years especially, the property insurance industry worldwide has been buzzing with concerns about coverage adequacy for BII. The problem affects both business owners policies (BOPs) and the larger package policies (CPP/SMPs). Caroline Woolley, senior vice president at Marsh’s Business Interruption Center, wrote a comprehensive report in 2015 summarizing the challenges the industry faces making BII coverage profitable. Woolley lays out five major obstacles that agents, companies and brokers face when underwriting this coverage line. Woolley says the No. 1 problem is simply “getting the values right” when policies are first written and again at time of renewal. The valuation concern stems from the fact that there has been no standardized, simple-to- learn-and-use insurance-to-value (ITV) system for BII coverages similar to what is done today for buildings and contents. No. 1: Getting the values right According to a survey conducted by the Chartered Institute of Loss Adjusters in 2012 ((and quoting from PMWBG)), 40% of declarations were deemed too low by about 45%. More recently, PMWBG research shows as much as 58% of BII coverages are undervalued by 48%, suggesting the problem is getting worse at a time when demand for property insurance is in decline and competition is fierce. Inadequate coverage disenfranchises consumers, and improper valuation undermines providers. In a very competitive marketplace, where too much supply is chasing dwindling demand, carriers losing on the valuation front lose reputation, financial advantage and long-term revenue. From the inception of BII coverage in the 1930s, calculating risk-specific BII limits has not been easy. The BII coverage addresses shortfalls in the margins corporations face when loss occurs, so underwriters, brokers and agents should understand key variables in the insured's financials. Unfortunately, not enough industry professionals are proficient in this area, leading to costly exposure errors, pricing mistakes and the age-old dilemma of undervaluation. As important is the fact that, unlike with other lines, there has been very little third-party data to aid insurers with BII calculations.  When losses occur, it’s too late in the game to correct undervaluation problems. The impact, especially in today’s economy, where wildfires, storms and other disasters routinely happen, has caused companies like Marsh to look again at the coverage line, suggesting the need for industry-standard ITV calculation tools.   Now, modern web-enabled technology offers both substantive raw data on businesses that actuaries will want to work with to improve pricing models, at the same time carriers will use the program’s web-based ITV system to calculate detailed BII coverage reports for the majority of businesses found anywhere in the U.S. Virtually any enterprise can be valued, with complex insurance specific data sets searched automatically on behalf of the user to both pre-fill input and create BII reports. First Step to Success Vast amounts of insight about corporations and their supply chains can be aggregated on to estimate BII limits in seconds, accessible anywhere from the Internet. In the case of the BOP sector, actuaries and pricing managers have instant access to large amounts of aggregated data for the various sizes and types of business insured, to develop more representative and localized pricing models. Users can also adjust models automatically for the business opportunity rather than offer one-size-fits-all pricing. Additionally, because core data changes annually, savvy users can also upgrade model variables.

The Internet is a mirror of sorts — a data mirror. Right now, it is a sort of fuzzy data mirror, but the pictures grow clearer as the available data grows. Soon, the image of an insurers’ customer service, pricing and claims experiences will grow crisp. How will it happen? How will insurers respond and remain competitive? In Part 1 and Part 2 of our series, we discussed data symmetry — the leveling of the playing field that is currently happening because insurers are gaining access to many of the same streams of data. The trend runs in contrast to data asymmetry, which allowed insurers to comfortably differentiate themselves by being good at the analysis of their own in-house data. As insurers use more and more of the same data and some of the same analytics tools and methodologies, they will find themselves in a pool of sameness. Differentiation by price and service will be less about introspective analysis and more about finding and delivering on real brand promises. So, in today’s blog we are crossing a bridge of sorts. We are going to look at how the consumer will achieve data symmetry by gaining a clear view of the real insurer. See also: Data Science: Methods Matter Changes in scrutiny are causing data symmetry Insurers are the subjects of constant scrutiny. The NAIC, the Federal Insurance Office, the Department of Labor, every state and every consumer protection organization have an interest in watching insurers. Yet all of that scrutiny may pale in comparison to the impact of the coming wave of individual consumer scrutiny. Consumers are using ratings, stars, comments and shopping patterns to give instant feedback to all service providers. Feedback (real experience) is a sales tool for aggregators and retailers. It is a reason for consumers to choose particular channels or pipelines. Amazon and eBay don’t have to build trust for any one product. They only have to facilitate feedback and let the products, services and suppliers speak for themselves. These outside views are the result of symmetrical data availability. Prospects are now able to compare any product or service, including insurance, with greater real data, including both sources that are verifiable and those that contain unstructured data. Consumers may look at an insurer through the lens of an insurance aggregator, such as Insure.com or The Zebra, or through simple search terms such as “worst auto claims experience in my entire life.” They may also witness an insurance interaction through their relationships with friends on social media. Reputation analysis will hold tremendous power to validate or invalidate brand promises. Does the insurer make it simple to file a claim? Does it have a poor track record in paying claims? Are renewal rates much higher or lower than competitors'? These bits of information weren’t as public in the past. Today, they are common and easy to find. See also: What Comes After Big Data? Data symmetry’s effect on the insurer will operate much like a looking glass. The insurer will begin to see itself, not as it has attempted to portray its brand, but as it is perceived during real interactions. This will lead some insurers to make course corrections. The good news is that data symmetry will supply healthy doses of reality. Insurers will know and understand their competition. They will have an unprecedented, timely idea about what customers really want and how well they are supplying it. If they are prepared for the coming levels of data symmetry, insurers will also be able to make agile shifts and meaningful steps toward selling insurance through many different channels. Many of these details are still food for our insurance visions. One thing is certain, however. Data and analytics will continue to unlock the secrets of market positioning to keep insurers competitive. Data’s relevance to business decisions will always grow.

This is part two of a three-part series. Part 1 can be found here.  As we saw in a previous blog post on AI Everywhere and Nowhere, defining artificial intelligence is like trying to hit a disappearing target. As soon as any aspect of AI gains widespread adoption, people fail to distinguish it as an AI technology, and it dissolves into the sea of general technology. As a result, most detractors of AI, at least until recently, have questioned the real-world applications of AI. In turn, AI never gains the respect and recognition it needs to evolve and reach its full potential. The beauty (and bane) of AI is that it is everywhere and yet nowhere – it is becoming ubiquitous in all of our interactions (at least all of our virtual interactions), yet most people fail to recognize and respect it. Artificial Intelligence Is Ubiquitous Intelligence You wake up in the morning and from your bed ask your digital assistant, "What is the weather like today?" It replies, "We have 80% chance of snow in Lexington later in the evening – with accumulations of one to three inches." The voice recognition, the natural language understanding of our question, the search through the Internet to get the right answer and the translation of that answer into speech is all AI. You get into your office and open your email. Your email gets automatically sorted into "Social," "Forums," "Private" or whatever categories you have created, gets identified as important or not or marked with whatever tags you have provided to make it easier for you to read and clear your email. The classification of your email based on the To, From, Subject and Content fields, the natural language processing to extract the right keywords, the machine learning to determine what is spam or not spam or who is important or not is all AI. You open up your online newspaper to check on the stock market performance from yesterday. You get a description of the overall stock market performance and the movement of your favorite stocks. The news is personalized to the topics, sources and authors that you want to read, and the newspaper has recommendations on what is trending among the sources or people you follow. The natural language generation based on structured stock market performance data, the curation of articles based on personal preferences and the recommendation engine for suggested articles are all AI. You open up your favorite search engine, and, as you type your query in the search box, the system suggests possible completions. Then, the system recommends the right websites from billions of documents on the Internet and the right ad that matches your query, and fulfills the best bid for your search term among competing advertisers who want to personalize their message to you. The statistical inference in suggesting completions, the page rank algorithm that computes the relevant pages to display and the selection of the right ad using a real-time ad exchange are all AI. See also: How to Think About the Rise of the Machines The list goes on and on. In fact, there is very little in our day-to-day life that is not affected by AI in some way. Yet the real power of AI is the insight that it provides us, without our being aware of it. The intelligence hidden behind many of our day-to-day interactions is powered by an AI algorithm related to machine learning, natural language processing or more generally unstructured data processing, intelligent search, intelligent agents and robotics. And, while AI is ubiquitous, we have only scratched the surface regarding what it can mean for us.

Pharmacogenetic testing (PGT) has the potential to help clinicians improve outcomes for injured workers and reduce costs for payers. While research showing the clinical value of PGT continues to grow rapidly, evidence of the return on investment in the workers' comp space is just beginning to emerge. Practitioners can benefit from the technology without falling victim to the hype of some proponents by becoming better educated about PGT and those providing it. Because the use of PGT in the workers' comp population is relatively uncommon, practitioners may find it challenging to realize the true value of the tests. "A few of our customers are trying PGT on select claimants," said Dianne Tharp, pharmacist and executive clinical liaison for pharmacy benefit manager Healthcare Solutions, an Optum company. "This is a complex area; everything is evolving. It's relatively new for the industry, and we are all still learning." One growing area of interest is in genetic tests that can identify injured workers most at risk for addiction and abuse. However, there are many challenges with such tests, including uncertainty about their predictive performance in clinical settings, which must be overcome before clinicians can use them to help identify whether an injured worker may misuse or abuse a prescribed opioid. While PGT could be a welcome tool, the science is not yet at a level where clinical application is appropriate. "On the other hand, pharmacogenetic testing for drug response is often more — and in some cases highly -- predictive,” said Naissan Hussainzada, senior director of genetics strategy and commercialization at Millennium Health. “For example, certain genetic variations can change how an individual metabolizes some opioid medications. Using this information, clinicians can identify patients at higher risk for medication failure and/or side effects, which may help them make more informed and tailored treatment decisions.” Injured workers with preexisting conditions or those who develop comorbid conditions post-injury may especially benefit from PGT — as they may be receiving multiple medications that could potentially elevate their risk for drug-drug and gene-drug interactions. PGT information could also help the clinician better understand whether drugs prescribed for comorbid conditions will be effective. "In the workers' comp space, PGT could be used to help the clinician optimize medication prescribing and avoid trial and error,” Hussainzada said. “This has the potential to translate to faster recovery, less time away from work and shorter claim duration for the injured worker.” See also: Genetic Testing: The New Wellness Frontier Polypharmacy challenges Multiple medication regimens and comorbid conditions are frequently present in workers who are injured on the job. The inability to work and the presence of pain can result in additional comorbidities, especially depression. Metabolism can play an important role in how patients respond to medications, particularly antidepressants, opioids, certain anticoagulants and cardiovascular medications. Mental health providers, in fact, were among the first to recognize the value of PGT in guiding medication therapy and dosing. "Mental health disorders are often assessed subjectively, and drug therapy can be lengthy, unpredictable and suboptimal,” Hussainzada said. “It may take several months to stabilize a patient on an effective antidepressant using trial and error.” PGT can be especially useful for antidepressants. “There are actionable PGT results with good evidence for the antidepressants,” Tharp said. “That would be an instance where PGT may be useful [among injured workers].” In addition to antidepressants, Tharp said PGT is also being used to help determine a patient's ability to properly metabolize warfarin, which is used to prevent blood clots. Drug-drug interactions Individuals metabolize medications differently, partly depending on a person's genetic makeup and partly on clinical factors, such as hepatic (liver) disease, lifestyle factors and administration of other medications. For example, introducing a new medication may change how existing drugs are metabolized, which can change their effectiveness or tolerability. Conversely, an existing medication may have an impact on the metabolism of a new medication. “There are well-documented drug-drug interactions between opioid analgesics and certain antidepressants,” Hussainzada said. “This is because some antidepressants can inhibit or ‘turn off’ the enzymes responsible for metabolizing opioids. This can lead to the opioid becoming less effective, or in some cases, intolerable or potentially toxic. Making matters more challenging, there are some individuals that carry certain genetic variations that can make them more susceptible to a phenomenon called ‘phenoconversion,’ which can elevate their risk for certain types of drug-drug interactions. For injured workers receiving polypharmacy, PGT may help clinicians identify these higher-risk individuals and help mitigate some of the risks of phenoconversion.” There are four categories of metabolizer type that correspond to how individuals may metabolize certain medications via hepatic enzymes. Individuals classified as “extensive” metabolizers possess fully functional enzymes and are able to metabolize medications normally. However, some individuals carry genetic variations that lead to reduced or significantly reduced enzyme function, and are classified as “intermediate” or “poor” metabolizers. Finally, some people may have genetic variations that lead to significantly increased enzyme function and are classified as “ultra-rapid” metabolizers. What that means is: Two people taking the same drug at the same dose can have very different responses because of their metabolizer status. Individuals susceptible to phenoconversion can “switch” metabolism type, for example, from an intermediate or extensive metabolizer to a poor metabolizer. The trigger for these conversions is non-genetic extrinsic factors, such as administering a drug that inhibits the enzyme pathway. Certain metabolizer types are associated with higher risk of phenoconversion and risk of drug-drug interactions. "Intermediate metabolizers may be at higher risk for phenoconversion compared to normal metabolizers," Hussainzada said. "However, it can be difficult to identify these patients because they may display normal or typical response to a medication, even if they are metabolizing that drug at a reduced rate. However, if an inhibitor of the drug is added to their regimen, this can shift the individual from intermediate to poor metabolism and lead to medication failure and/or potentially serious side effects.” For some claimants who take medications for pre-existing conditions, adding a pain medication can increase the risk for drug-drug interactions and phenotypic conversion. "So a claimant who has been taking antidepressants for years is now also prescribed an opioid because of his injury," Hussainzada said. "If he is an intermediate metabolizer for the opioid, the antidepressant may convert him to a poor metabolizer. This could lead to inadequate pain relief, which may delay recovery and increase risk of poor outcomes.” In another scenario, an injured worker who is taking opioids for his injury and who later develops depressive symptoms may be treated with concomitant antidepressant therapy. “In this case, the opioid may have been initially effective, but certain opioids would lose analgesic potency once the inhibitor, or antidepressant, is added," Hussainzada said. PGT can also help a clinician identify patients who may need to be started with atypical or non-standard doses of certain analgesics. One particular enzyme responsible for the metabolism of a large number of medications is cytochrome P450 2D6, or CYP2D6. Claimants who are reduced metabolizers for the pathway may not respond adequately to a standard dose of oxycodone. “If you are a CYP2D6 poor metabolizer, standard doses of oxycodone or hydrocodone may not effectively control your pain,” Hussainzada said. “However, without knowing this type of genetic information beforehand, it may appear to the clinician that these individuals are drug-seeking if they continue to ask for higher doses.” Some poor metabolizers may not get any pain relief, even with very high doses of a medication. Identifying these patients through PGT can lead the clinician to prescribe a different pain medication from the start, something that can be critical to getting an injured worker back to function. According to a recent position paper from Healthcare Solutions, the rates of comorbidity and polypharmacy are on the rise in workers’ comp and can lead to increased medical costs, delayed returns to work and longer claim durations. Clinical depression is a common comorbidity, and the use of antidepressants is prevalent; however, both are associated with poor recoveries and outcomes. "For patients taking multiple medications, there may be multiple enzymes that are recruited to metabolize and eliminate these drug combinations from the body,” Hussainzada said. "Some recent data indicates that when you look across multiple enzymes, genetic variation is much more common than when you look at a single enzyme. So for the claimant receiving polypharmacy, it may be even more important to understand how their genetics will contribute to their medication response since it is likely that at least one enzyme system may be variant.” Clinicians can use PGT information at the beginning of a claim to optimize initial prescribing and dosing of opioids and other medications, which may hasten the recovery time. "In workers' comp, the data are pretty clear: The faster we can facilitate post-injury recovery and get the claimant back to work, the better their overall prognosis,” Hussainzada said. "Particularly with opioid therapy, we want to use these drugs judicially and effectively. See also: Urine Drug Testing Must Get Smarter The future Researchers and workers’ comp practitioners continue to monitor the clinical evidence for testing in an effort to help clearly identify those injured workers who would benefit most from PGT — in terms of better outcomes and lower costs. For now, there are several types of injured workers who may be good candidates for testing. "A claimant taking multiple medications from several therapeutic classes, one who has failed several therapies and changing dosages or a patient on ultra-high daily morphine equivalent doses may be a good candidate for PGT,” Healthcare Solutions' Tharp said. Ultimately, proponents hope PGT can be a useful tool in getting the right medication at the right dose to each patient. If test interpretations are based on firm clinical evidence, PGT can provide clinicians with a road map for navigating prescribing decisions that can often be complex and subjective. However, providers are advised to become familiar with PGT and, especially, the companies marketing these services. "Payers, clinicians and patients need to be aware that not all pharmacogenetic testing is equal. Ask questions about the evidence for specific genes and drugs and make sure there are clinical standards in place for how results are interpreted,” Hussainzada advised. "Some tests may not be ready for clinical use, so it’s important to be informed."