Injured Workers

Washington State provides good financial and vocational benefits for injured workers. In fact, the Department of Labor and Industries (LNI) offers some of the best financial and vocational benefits in the nation! But what about health benefits? How can we improve access to quality of treatment following an injury at work?

~ Download white-paper ~

Background and motivation

Workers’ Compensation in Washington State is designed to protect injured workers and their beneficiaries. Its purpose is to provide workers with benefits and reduce to a minimum their suffering and economic harm as a result of industrial injuries and occupational diseases. At its core, the Department of Labor and Industries (LNI) has a fiduciary duty to ensure these benefits are provided to injured workers. But after all, LNI is a government administration body, presenting its own challenges and operational constraints.

We want to improve the quality of healthcare, and access to necessary and proper care for injured workers in Washington State. There are plenty of studies documenting various aspects of the Workers’ Compensation ecosystem. Interestingly, studies usually follow what we refer to as “top-down” analysis and approach. They are driven by government agencies and insurance companies. Please refer to the Resources section at the bottom for some references. Unfortunately, we feel that the system has become significantly bogged down by administrative processes, overhead, and wasteful spending. In top-down studies, it is not surprising that injured workers (who we believe are at the bottom of the workers’ compensation stack) are becoming less of a priority.

In Medabra, we want to challenge the status quo. Our goal is to take a bottom-up approach and rely on detailed low-level data to identify worker-centric issues. We aim to identify and influence opportunities for systematic improvement.

We are well-aware that the workers’ compensation system is very challenging. For one, it is difficult to distinguish the effect of care from non-medical factors such as the workplace, employer accommodation, comfort of financial benefits, worker motivation, and so on. Another reason is insurance and employer incentives (or disincentives), which may or may not work as intended. For example, with LNI payout rates, some providers avoid complex patients or claims. Another challenge is data limitations – medical data can be fragmented making it harder to look at past medical conditions. Finally, there are state rules and regulations (e.g., provider networks, incentives, utilization reviews, etc.), dictating what is “feasible”. These are our main focus areas that we hope to challenge and disrupt. Below are more specific topics that we plan to investigate and influence.

Before we proceed, please note that the entire discussion below focuses on workers’ compensation claims involving medical conditions. We do not touch on mental health claims, which LNI allows for first responders or in sudden and tangible incidents that produce mental health conditions. We also do not look at medical claims where mental health conditions develop (and can be accepted by LNI) over time. Those are far more complex claims that present issues beyond the scope of this write-up. Furthermore, we focus on claim administration with LNI. We do not touch on self-insured and third-party claim administrators (TPAs). In actuality, TPA claims often exhibit many additional challenges impeding recovery and returning individuals to work. We plan to delve into these topics (and others) in future phases of our research.

Access to quality treatment

LNI has the Medical Provider Network (MPN). To treat injured workers, a provider must be in the MPN. On the LNI website, you can use the provider directory tool to find a doctor. The tool shows dozens or even hundreds of available providers. But… If you pick up the phone and call these providers – only a handful might actually see you.

Injured workers spend countless hours trying to find providers. Many end up with unwilling providers. Some are unable to find providers at all. Our first task here is to look at claim data to understand which providers or facilities are actually seeing injured workers, in what capacity, and to what effectiveness. Based on this information, we plan to:

1. Develop an injured-worker-centric framework and metric, based on medical claim data, that will allow us to characterize willing and/or effective providers*;

2. Understand the capacity and utilization of providers that regularly see injured workers;

3. Provide feedback to LNI to minimize the noise/fluff and maintain smaller regional lists of more active providers that are in fact willing to see injured workers;

4. Think about the following – How can we make it easier for more providers to see injured workers? If we focus on willing and effective providers and look at their patients’ claim records – what can we conclude? Remember, these providers accept lower treatment payouts because of LNI payout rates. Yet, we often hear complaints that LNI claims are too administratively burdensome.

5. How can we incentivize and reward willing providers? How can we reduce their administration overhead? What else can we do to help them?

Recently, another issue that relates to quality of treatment has been brought to our attention: The Attending Provider plays a key role in making sure workers receive proper treatment. There are many types of healthcare providers that can serve as the Attending Provider, according to LNI. The same rules are also captured in the Revised Code of Washington (RCW) – see RCW 51.36.010 2(a).

Currently, with few exceptions, LNI does not enforce which professionals can serve as the Attending Provider for certain types of injuries or occupational diseases. One exception (for example) is allowing psychologists to serve as Attending Providers only for mental health conditions. While this makes sense, it is unclear what happens if a worker has both physical and mental conditions. However, our broader concern is that nothing prohibits workers from seeing the “wrong” providers considering their underlying conditions.

Consider the following example: Say a worker suffers from back injuries. In general, back injuries often result in complex and lengthy claims. Nothing prevents the worker from choosing a podiatrist to serve as their Attending Provider. While we expect the podiatrist to politely decline, this example demonstrates a fundamental gap. Continuing with the example- the worker might also select a naturopath or a chiropractor as their Attending Provider. Both are good options. However, what happens if proper and necessary treatment involves surgery? Are these the appropriate Attending Providers for the claim? We would like to understand this subject better and investigate the following topic.

6. What type of providers move treatment in the right direction in claims? Are there certain types of providers that cause regression or “mess” in claims? If we look at workers that switch from one Attending Provider to another over the life of the claim – from a treatment standpoint, can we identify “effective” providers or provider types?

* The framework will include the relevant definitions, with focus on the injured-worker benefit and health outcomes. For example, while existing studies focus on early return to work as a function of cost, returning to work when truly medically appropriate is widely ignored.

Benchmarking: Treatment under LNI vs Medicaid or commercial insurance

Based on national comparisons, Washington State LNI has top financial and vocational benefits for injured workers. But we see many cases where treatment is not up to par. Treatment is sometimes unreasonably rushed or delayed, objective medical tests are not being authorized (or incur significant authorization delays), and so on. The list of less-than-optimal treatment issues is long.

At the same time, LNI spends time and significant resources on independent medical examinations (IMEs) that while are relevant at times, are often ineffective and result in administrative delays and/or prolonged litigation, or on fighting/delaying claims due to internal processes and policies. We understand that claim cost is key in the workers’ compensation setting. Our goal here is to develop benchmarks based on medical claim data. More explicitly, we plan to tackle the following topics:

1. Using a bottom-up approach, how can we use claim data to measure access to and quality of care covered by state funded LNI claims or self-insured employer claims? How can we compare LNI treatment to equivalent care covered by other programs (e.g., Medicaid or commercial insurance)? What can we conclude?

2. Consider a person receiving treatment under an LNI claim after a workplace injury. Take that person and consider their conditions and medical history. Next, look at medical claims data to identify individuals with similar characteristics and conditions that arose outside of a workplace setting. What does the treatment journey look like for injured workers compared to non-work-related injuries? What can we learn about the quality of treatment (e.g., availability, authorizations, continuity, etc.)? Are there differences in the path or speed to recovery? What are the explicit and implied cost differences?

3. Where are the areas for improvements? How can we apply our findings to solve “low hanging fruit” to help improve recovery for injured workers?

Continuity of care

We sometimes receive complaints from injured workers in specific industries about changing providers. Take the Concentra Urgent Care clinics for example. In the construction industry, it is common for a construction manager to take a fellow worker immediately after an injury to a Concentra location. We know that Concentra personnel serve as Attending Providers in workers’ compensation claims. However, we received reports that when workers revisit the clinic, they see a different provider every time. While we are certainly thankful to Concentra and its providers for willing to serve injured workers, this seems counter-productive considering the role of the Attending Provider. After all, the Attending Provider is supposed to orchestrate treatment through the life of the LNI claim.

1. If we use our bottom-up approach and look at the system as a whole, starting with medical claim records– what does the continuity of care look like for injured workers? How does it correlate with the length of treatment until workers reach Maximum Medical Improvement (MMI)? Or until actual recovery? In claims where continuity is lacking- what does treatment look like after workers reach MMI?

2. Measure the quality or effectiveness of IME exams – The original motivation behind IME exams is (or was more correctly) good. It’s a way to independently measure the conditions and state of health of an injured worker. Unfortunately, in many cases, IMEs have become a tool used by LNI to close claims in a biased manner (e.g., ignoring medical charts and recommendations from attending providers). This is where long lingering legal fights begin instead of focusing on health and care. Is there a way to measure the effectiveness of IME exams? Can we look at how many IME recommendations are rebutted by attending providers and how such claims progress from that point on? Can we identify IME doctors that perform exams in a truly unbiased manner?

Returning injured workers back to work

We can all agree that when injured workers are capable of working, then returning them to work is the best outcome of workers’ compensation claims. However, concerns have been raised that within the LNI setting, simply a finding on paper that the worker is capable of working is sufficient to log that as the actual outcome. We are concerned there may be gaps in the data being collected that prevent us from being able to ascertain whether workers are actually able to return to work successfully. The key questions here seem to be: Are injured workers being released to return to work when they are medically ready to work successfully? Is it possible to measure instances of actual/successful return to work?

Plenty of studies have shown that prolonged time-loss compensation (i.e., being off work for a long amount of time) directly correlates with long term disability. Consistent research conclusions show a sharp decline in the likelihood of returning to work once absence extends beyond about 3 months. After 12 months, the odds are much lower. After 2 years, the probably of working again is closer to zero.

Because of this, LNI spends significant time and money to shorten time off work and minimize the time-loss benefits they pay out on claims. However, many times it feels that LNI is trying to “cut corners” and return injured workers to work before they are actually ready to be successful returning to work.

Considering that the largest decline in returning to work is around 12 months, it makes sense that LNI claim managers push to close claims in under a year. But wouldn’t it make more sense to return people to work by removing barriers to treatment faster? For example, consider the following hurdles and administration challenges:

a. When claim managers deny treatment, workers have to appeal those decisions. The appeal is done through the Board of Industrial Insurance Appeals (BIIA). The process takes approximately 18 months, which puts workers well over the 12-month cliff.

b. Claim managers are ultimately responsible for approving or denying treatment. Claim managers review (and react upon) each claim once every 30 days. Considering the number of treatment authorizations needed in certain claims, the potential 30-day delay seems impractical under the 12-month target.

c. The utilization review and authorization contractor for LNI is Comagine (more on this later). In theory, authorizations are supposed to take ~7 days. In practice, it takes 30+ days after Comagine receives any materials they request to authorize treatment.

d. When Attending Providers refer workers to specialists, many specialists refuse the referral due to billing and authorization concerns. Most specialists do not realize that initial consultations do not require LNI authorization. As a result, claim managers often get involved and call specialists to explain the process. Considering a typical ~3-day turnaround for claim managers to answer voicemails (in our experience), there seem to be unnecessary delays that do not contribute to achieving the 12-month goal.

e. Consider the challenges that providers face trying to move treatment forward: If professionals that specialize in serving workers’ compensation claimants encounter so many issues, how can we expect general providers to be willing to serve as Attending Providers? Or do so effectively?

Therefore, we would like to include the following topics in our study: Is there anything we can do to help claim managers move claims faster? Can we propose auto-authorization in certain circumstances? For example, what would it take to auto-approve N hours of physical therapy for lumbar strain? Or auto-extension for physical therapy? Can we trust that providers on the MPN know what they are doing when they recommend treatment and remove (or ease up) lengthy authorization barriers? How can we improve communications between providers and claim managers to shorten claim delays?

In this context, our research is going to focus on the following:

1. Where is the balance between cost savings and proper recovery? Using our bottom-up approach driven by actual medical claim records, if we look at the treatment history from a time of a work injury – what can we learn from the treatment received after LNI care is “complete”? What percentage of people, and under which types of work injuries and diagnoses, actually recover under the treatment covered by LNI vs follow-on treatment?

2. Duration of claim vs length of medical treatment – What can we learn by statistically measuring the length of care for specific conditions for work injuries and related (e.g., symptomatic pre-existing) conditions? How do those compare to the length of the claim and why?

3. Can we develop an unbiased formula or tool that will tell us when injured workers can return to work and in what capacity? Is there a way to come up with such an objective formulation (e.g., AI-driven via private healthcare/vocational LLMs), which examines the health record, employment history, worker profile, etc., to provide objective vocational recommendations? Can we use such a tool for other claim-related decisions? Can such tools assist claim managers, vocational counselors or IMEs? Or perhaps be used to replace IMEs (resulting in high 9-figure annual cost reduction) as a tool to close claims?

4. How will auto-authorization for certain medical conditions and treatment impact claim duration, potential return-to-work outcomes, and cost? Can we look at claim data and compare current LNI claim performance to ones with auto-authorization? What are the cost implications? What are the potential benefits and speed to recovery implications?

Treatment under COHE

Tightly related, we would also like to look closer at the LNI best practices under COHE, which is the Centers of Occupational Health & Education. The COHE initiative is a set of L&I-contracted programs embedded in major health systems that coach clinicians on what L&I perceives as occupational-health best practices. It includes care coordination and focuses on early return-to-work. The program also offers COHE providers certain incentives.

For simple injuries, outlets that follow COHE guidelines are fine because they provide a quick path to return to work. But for more complex cases (e.g., injuries that manifest pre-existing conditions, multiple injuries, or injuries that trigger non-medical LNI benefits) – these outfits seem to be rushing to get people out the door. Sometimes, at the expense of injured workers. Moreover, in some cases, COHE providers tend to provide incorrect legal or cost-related advice to workers. For example, we’ve heard reports that certain offices may tell injured workers that because they have reached a certain number of treatment visits, so it’s time to close their claim without regard to whether the treatment has been effective or has resulted in the injured worker having achieved maximum medical improvement. We do not believe that cost-saving factors should be managed by treatment outlets. The potential impact on quality of care is significant and problematic.

Let’s look at more explicit examples. We often get complaints from injured workers that obtain treatment through MultiCare and Kaiser. If a worker walks into MultiCare following a workplace injury, they are immediately directed to see providers in the Occupational Medicine division (or OccMed in short) even if they have had long-term established care with a different primary care provider within the same medical treatment facility. In Keiser, it is the same thing. Both are members of COHE.

Once the injured worker establishes care with a new treatment provider in OccMed, the most common claim-related diagnosis for any physical injury is a sprain or strain (i.e., soft tissue). From there, we tend to see 2 main paths: (1) If the diagnosis is correct, then the injured worker receives some physical therapy or other conservative treatment (e.g., rest, activity modification, and over-the-counter medication). Shortly after, the worker is released to return to work without restriction, the claim closes, and the individual moves on with their life; (2) If there are more objective findings and diagnoses (e.g., after an MRI), then things typically get more complicated and less optimal for the injured worker as we explain below.

Under the law, if an injury causes conditions to be symptomatic or to worsen, then LNI should cover the treatment for those conditions. However, with Keiser, MultiCare, and other COHE practitioners – it is common to see these kinds of aggravation pre-existing condition diagnoses “pushed aside” and not treated under the claim. They will still treat the person, but not under the LNI insurance. Injured workers frequently report being dissuaded from pursuing treatment of these conditions under the claim, being told things like “LNI will never cover that condition, it is a waste of your time, you should just get the recommended treatment using private insurance.”

Consequently, with MultiCare, we see cases where they treat eligible conditions but bill private commercial insurance even though the conditions should be paid under LNI. This works well for providers (who receive higher payouts). Similar issues are reported with Kaiser, although the cooperative nature of their system further complicates the financial incentive picture. It also works well for LNI (through cost savings). Why is this bad? Because it forces injured workers to miss out on other LNI benefits associated with their conditions or treatment. Workers would qualify for those benefits had the conditions been treated under the claim.

Putting all of this together, we aim to look into the following topics:

1. Can we correlate the number of visits that injured workers receive in one clinic with the longevity of treatment/recovery and the quality of treatment outcomes? What can we learn?

2. Are there medical facilities that are “better” than others from treatment standpoint? How do we quantify “better”? Can we find facilities where the percentage of injured workers that leave treatment with them and continue treatment with other facilities/providers is high? And if so – why?

3. For more involved work injuries (i.e., excluding sprain/strain and the like) – if we look at the continuity of treatment that started with conditions covered by LNI, what can we tell from the medical record? How is the evolution in treatment different in COHE members versus other clinics? What are the potential implications for injured workers?

4. How does treating conditions under non-LNI insurance benefit providers? Can we quantify the benefits? What are the corresponding benefits that injured workers are missing out on (in numbers)? Beyond missing out on benefits, are there out-of-pocket implications for injured workers in their path to recovery?

Additional topics of interest

As we dig deeper, we also aim to learn more about several topics in particular. These include measuring the unfortunate transition of some injured workers to homelessness,  learning more about Comagine and LNI processes for authorization, as well as learning more about HCA’s Health Technology Clinical Committee (HTCC).

Too often, work injuries can lead to homelessness. Washington State is especially challenging in this regard because of high cost of living. This transition happens, for example, when financial benefits are terminated following an IME while a worker is medically and practically unable to return to work in spite of the IME report. Other cases are when time-loss payouts are insufficient to cover bills and financial obligations. In our work on homelessness, since we are already looking at records – what can we conclude by correlating our findings with specific record that pertain to work injury instances?

Comagine Health is the utilization review (UR) contractor for LNI. Providers submit certain treatment requests to LNI. From there, LNI forwards requests to Comagine. In turn, Comagine reviews the requests under various considerations (e.g., medical necessity, appropriateness, cost-effectiveness, etc) and against LNI Treatment Guidelines. Then, Comagine provides authorization recommendations to LNI claim managers.

Based on our observations, it seems that Comagine uses the worker’s medical history to run automated computer algorithms for utilization review decisions. We also see cases with human intervention. For example, after denying MRI, a treating doctor can do peer-to-peer review with a clinician at Comagine to get it authorized.

We would like to learn more. It is not clear to us what data is used by Comagine and how. In addition, we sometimes see systematic-delays in receiving authorization decisions. At times, it seems that Comagine and/or LNI get “stuck” and delay authorization decisions. Delays obviously hinder recovery. We would like to understand more about how utilization reviews work under the hood.

Finally, with respect to the Health Technology Clinical Committee – we understand that this committee has a lot of influence over LNI’s treatment guidelines. Some information about the committee is available on the website of the Washington State Health Care Authority (HCA). However, we hope to learn more about this committee. More specifically, we would like to understand how the committee is involved with LNI, and if/how the committee considers special treatment circumstances that arise in the workers’ compensation settings.

References and selected resources (among many others, always changing):

1. American College of Occupational and Environmental Medicine, “Preventing Needless Work Disability by Helping People Stay Employed,” Jun. 2006, Available: https://acoem.org/acoem/media/News-Library/Preventing_Needless_Work_Disability_by_Helping.pdf

2. G. W. Wynne-Jones,  J. Cowen, J. L. Jordan, O. Uthman, C. J. Main, N. Glozier, and D. van der Windt, “Absence from work and return to work in people with back pain: a systematic review and meta-analysis,” Nov. 2013, Available: https://pmc.ncbi.nlm.nih.gov/articles/PMC4033140/

3. J. Crook, H. Moldofsky, “The probability of recovery and return to work from work disability as a function of time,” Dec. 1994, Available: https://pubmed.ncbi.nlm.nih.gov/7866379/

4. J. M. Sears, A. T. Edmonds, E. MacEachen, D. Fulton-Kehoe, “Appraisal of Washington State workers’ compensation-based return-to-work programs and suggested system improvements: A survey of workers with permanent impairments,” Aug. 2021, Available: https://pmc.ncbi.nlm.nih.gov/articles/PMC8500921/

5. L&I’s Advisory Committee on Health Care Innovation and Evaluation (ACHIEV) – see http://lni.wa.gov/patient-care/advisory-committees/advisory-committee-on-health-care-innovation-and-evaluation

6. L&I’s Centers of Occupational Health & Education (COHE) – See https://www.lni.wa.gov/patient-care/provider-partnership-best-practices/centers-of-occupational-health-education-cohe

7. O. Anderson, V. McLennan, and C. Randall, “Treatment and provider choice in worker injury rehabilitation: A systematic literature review,” Nov. 2021, Available: https://journals.sagepub.com/doi/full/10.3233/JVR-211171

8. T. H. Mekonnen, M. Di Donato, A. Collie, and G. rant Russell, “Time to Service and Its Relationship with Outcomes in Workers with Compensated Musculoskeletal Conditions: A Scoping Review,” Jan. 2024, Available: https://link.springer.com/article/10.1007/s10926-023-10160-0

9. T. M. Wickizer, G. M. Franklin, and D. Fulton-Kehoe, “Innovations in Occupational Health Care Delivery Can Prevent Entry into Permanent Disability: 8-Year Follow-up of the Washington State Centers for Occupational Health and Education,” Dec. 2018, Available: https://pubmed.ncbi.nlm.nih.gov/30234763/

10. T. M. Wickizer, G. Franklin, R. Plaeger-Brockway, and R. D. Mootz, “Improving the quality of workers’ compensation health care delivery: the Washington State Occupational Health Services Project,” Jun. 2003, Available: https://pubmed.ncbi.nlm.nih.gov/11286095/

11. W. Santos, C. Rojas, R. Isidoro, A. Lorente, A. Dias, G. Mariscal, A. H. Zabady, R. Lorente, “Return to Work After Work-Related Injuries: A Systematic Review and Meta-Analysis of Incidence and Determinants,” Jun. 2025, Available: https://pubmed.ncbi.nlm.nih.gov/40566086/