Tag Archives: jayant lakshmikanthan

How to Attack the Opioid Crisis

The vastness of the opioid crisis is all around us:

  • 259 million opioid prescriptions are made every year.
  • 91 Americans die every day of opioid overdose.
  • Workplace costs of prescription opioid use are more than $25 billion, driven by lost earnings from premature death, reduced compensation or lost employment and healthcare costs.

It’s time to take action.

See also: Opioids: A Stumbling Block to WC Outcomes  

As with any large-scale, complex phenomenon, there is no silver bullet. But a framework from the Johns Hopkins Bloomberg School of Public Health suggests three areas where we should focus our efforts: preventing new cases of opioid addiction, identifying opioid-addicted individuals early and ensuring access to effective opioid addiction treatment. We believe these areas must be attacked from a variety of clinical and operational angles.

From the clinical side, the emphasis has to be largely around better clinical training and urinary drug testing (UDT). A generation of doctors has been raised based on a curriculum emphasizing the need to manage pain aggressively. Retraining physicians on best practices is needed to reinforce safe opioid prescribing patterns. Research from Utah has shown that physician education on recommended opioid prescribing practices was associated with improved prescription patterns, including 60% to 80% fewer prescriptions for long-acting opioids for acute pain. When an opioid is prescribed, the use of UDT is a cost-effective way to monitor treatment compliance and drug misuse.

To address from the operational side, we need evidence-based opioid prescription guidelines in place and systems to track opioid prescriptions and adherence to guidelines. Further, we must ensure access to effective opioid addiction treatment.

Many health organizations and state health systems are aggressively adopting pain treatment guidelines that clearly lay out when opioids should and should not be used. And the preliminary results of implementing these guidelines are promising. For example, the introduction of opioid prescribing guidelines in the Washington state workers’ compensation system was associated with a decline in opioid prescriptions, the average morphine equivalent doses prescribed and the number of opioid-related deaths.

Prescription drug monitoring programs (PDMP) allow for health systems to analyze opioid prescribing data to find potentially inappropriate prescribing behavior and illegal activity. For example, using its PDMP, New York City found that 1% of prescribers wrote 31% of the opioid prescriptions.

While prevention of initial opioid exposure is important, the treatment of opioid addiction is an important safety net when prevention fails. Pharmacotherapies including methadone, buprenorphine and naltrexone are options for routine care of opioid dependence, but they are still in the early stages of the adoption cycle.

See also: Potential Key to Tackling Opioid Issues  

The foundation to address the clinical and operational approaches to opioid epidemic is two-fold:

  1. A strong system to determine what’s acceptable through well-defined, evidence-based guidelines; and
  2. A system to use these guidelines and trigger the right actions through processes and technology.

The next article will address the nature of these two systems.

A Silicon Valley View on Work Comp

Occupational injuries cost the U.S. more than $250 billion annually. That is nearly three times the financial impact of cancer. Yet to date, the technology and analytics community has largely underserved the challenges of effectively helping injured workers get back to being productive rapidly. Injured workers are being pulled into complex processes unnecessarily. Claims adjusters juggle many balls and are not able to focus their time on what they do best: being a trusted adviser to the injured worker.

The technology and analytics community can make an impact by helping drop combined ratios by more than 20% through better pricing and improved operations. To date, these efforts have been delivered largely through a one-off services model, an approach that works for specific scenarios in which objectives can vary by carrier.

See also: Data and Analytics in P&C Insurance  

For universal challenges across carrier, the one-off services model is suboptimal, and a product-centric model is recommended to maximize impact. Two such carrier challenges that affect the lives of claims adjusters daily and need special attention are:

  1. Connecting the injured worker to the right providers. The choice of the provider for a claim makes a big difference to its outcome. From a total cost perspective, a bottom-tiered provider can cost five times as much as a top-tiered provider. Improving the quality of a medical network and directing claims toward better providers can reduce average claim costs by more than 10%. To suggest the right providers, claims adjusters need a solution that ranks providers in a fair, accurate, comprehensive and defensible way. The system also needs to be very easy to use so that the adjuster can come up with the right answer instantaneously when the injured worker calls.
  2. Reducing claims escalation and focusing the team’s attention. The majority (~75%) of claims are simple and can be fast-tracked. However, the few that are complex (e.g., heading toward litigation or high costs) drive the bulk of the effort. Determining which claims are heading toward a simple outcome and which ones are heading toward complexity is challenging. The ever-changing nature of the claims complicates the situation. The claims team needs a solution that goes through the open claims and helps focus efforts. It needs to be highly accurate, dynamic (i.e., account for the changing nature of the claims) and integrate well into the team’s workflows. In short, the technology solution needs to mirror the dream analyst that every claims team likes to have — the one who is constantly on top of the claims and helps adjusters focus on being a trusted guide to the injured worker.

Why Now?

Analytics, in particular, and technology, in general, have passed through the hype cycle and are now accepted as required parts of the workers’ compensation solution for these reasons:

  • Underlying technology platforms are more mature. Claims management systems are being upgraded or replaced industrywide. They are more flexible, comprehensive and integrated than ever before. With this maturity comes the ability to easily connect one system to another and change workflows, an essential ingredient in accelerating change. Uber wouldn’t have happened if payment systems did not connect easily.
  • Analytics have started proving value. The advances made on analytical models over the past five to 10 years have started showing clear, tangible results. Underwriting and pricing models have brought down combined ratios dramatically. Additionally, provider scoring models have reduced costs by more than 10% year-over-year, and litigation models have brought down attorney involvement rates by several percentage points. The value of analytics is no longer under scrutiny. The question now is: How can we realize impact?
  • Both analytics and technology are essential to attracting new talent. Millennials will not accept archaic, paper-based processes. Most don’t know life without technology, and they treat it as a given. To attract new talent to the workers’ compensation industry, providers need to serve up tools that our future leaders can use and relate to rapidly. There is no alternative.

Why Current Delivery Models Are Obsolete

Most advanced analytics efforts have been one-off projects by internal teams or boutique consulting firms. They are primarily geared toward proving the point but not designed for scale and longevity. They served a purpose while the market was sizing up the value of analytics. However, these services have led to unnecessary redundancy across the industry, and, lacking a long-term strategy, these suboptimal solutions have stalled over time.

Are there exceptions? Sure. There are several models in which the objectives differ from carrier to carrier. For example, pricing models are intricately tied to the strategy of the carrier and will therefore have different goals for each carrier. In carrier-specific models, internal or outsourced analytics projects make sense.

See also: How Technology Breaks Down Silos  

However, for most claims operations, the objective is identical across carriers: reduce the cycle time of processing claims. To solve this challenge comprehensively, carriers need to have a dedicated focus over a long period. It takes hundreds of iterations to get all the pieces in place before one can call the solution complete. What is needed is a product-centric model.

What Is the High-Impact Promise of a Product-Centric Model?

A product-centric model is focused on creating the most robust solution possible across the entire industry. It is about identifying a problem that is common across many customers and then dedicating an R&D effort to it. Differences between customers are handled through configurations, such as switches that can be turned on or off, rather than customization, such as building brand-new models and using different inputs from customers. Product teams focus on select issues and continuously innovate.

A product-centric model delivers:

  • A continuously optimized model. Having a team of smart data scientists, engineers and product managers working toward the same goal for an extended period has an almost magical effect. All situations are thought through, and the solution is deep and complete. Experience builds on experience to create an exponentially rich set of features.
  • A cost-efficient model. R&D costs can now be distributed across the industry, making the cost for each customer much lower than a one-off solution. This is especially true when considering the total cost of the solution, including design, implementation, maintenance and upgrades.
  • A quickly implemented model. The constant refinement of the product makes it as close to plug-and-play as possible. Timelines can be reduced from months to days and minutes.

From our market size estimates, each of these challenges faced today by claims operations represents a $5 billion-and-upward opportunity across the industry. The potential of solving these challenges with advances in technology and analytics is significant from an economics standpoint. More importantly, a product-centric model will empower claims adjusters to do what we set out to do in the first place: “Get injured workers back on track rapidly.”