Large Claims Deconstructed Guide

Large Claims Deconstructed Guide

Posted Date: May 10, 2023  

---

Introduction

NCCI’s Large Claims Deconstructed presentation provides an in-depth review of the characteristics and drivers of large claims. In this presentation, large claims are defined as those that cost $1M or more. This Large Claims Deconstructed Guide provides a slide-by-slide examination of the key insights, data sources, and formulas underlying the Large Claims Deconstructed presentation.

As you review the information contained in this Guide, it may be useful to keep in mind the following main takeaways that NCCI’s Large Claims Deconstructed presentation highlighted:

• Large claims make up less than 0.5% of all lost-time claims, but up to 15% of total workers compensation indemnity and medical losses in any given accident year.
• Since 2002, the relative frequency of large claims has decreased at an annual rate of 3%.
• The factors driving inflation in large claims can be vastly different from the trends affecting more typical workers compensation claims.
• Hospital costs are the biggest driver of large claims. But home health care expenses, medical supplies, and changes in drug prescribing patterns are also significant factors.
• Large claims can be sorted into two groups: fast-emerging claims and slow-emerging claims. Fast-emerging claims reach $1 million within two years and are typically the result of severe, traumatic injuries. Slow-emerging large claims reach $1 million after the second year and are often soft tissue, degenerative injuries that balloon over time due to multiple medical conditions.
• Notable diagnoses for fast-emerging claims include burns, spinal cord injuries, and traumatic brain injuries. The frequency of these diagnoses has grown by nearly 7% per year since 2012.
• Claims from the construction sector make up more than a third of fast-emerging large claims.
• The most prevalent diagnoses for slow-emerging claims include degenerative disc disorders and pain. The frequency of these diagnoses has decreased by 11% per year since 2012.

We hope you find the Large Claims Deconstructed Guide both a beneficial and informative resource.

Defining Large Claims

Defining Large Claims

Key Insights

• Large claims are defined as those that cost $1M or more.
• Large claims make less than 0.5% of all lost-time claims, or approximately 50 out of every 10,000 injured workers with a lost-time claim. However, these claims account for approximately 15% of total workers compensation indemnity and medical losses.
  
• There is limited data to study large claims. In any year there might be 500-1000 of them.

Background

For this analysis, large claims are defined as those that meet the following criteria within the first ten unit report levels:
\[Incurred \text{}\ Medical_{Trended} + Incurred \text{}\ Indemnity_{Trended} ≥ $1M\]

All incurred loss amounts are trended to 12/31/2021. Incurred indemnity losses are trended using Average Weekly Wage values from the Quarterly Census of Employment and Wages (QCEW) as a proxy. Incurred medical losses are trended using the Personal Health Care index published by the Centers for Medicare and Medicaid Services (CMS). Refer to the Unit Statistical Data Valuation and Due Dates table in Appendix B for more information on the unit report levels.

Data

• NCCI’s Statistical Plan data. COVID-19 claims are excluded.

Deconstructing Large Claims

Key Insights

• Large claims can be split into two categories:
  • Fast-emerging large claims (FAST_ELC) exceed 1M in cost within 2 years of initial injury.
  • Slow-emerging large claims (SLOW_ELC) exceed 1M in cost more than 2 years after the initial injury.
• A traumatic brain injury that requires immediate and intensive treatment would be an example of a FAST_ELC claim.
• A strained back injury that degenerates over time would be an example of a SLOW_ELC claim.

Background

FAST_ELC claims exceed the $1M threshold in unit report levels one or two based on incurred loss amounts trended to Accident Year 2021. SLOW_ELC claims exceed the $1M threshold in unit report levels three through ten based on incurred loss amounts trended to Accident Year 2021. Refer to the Unit Statistical Data Valuation and Due Dates table in Appendix B for more information on the unit report levels.

Data

• NCCI’s Statistical Plan data. COVID-19 claims are excluded.

Frequency

Frequency of Large Claims

Key Insights

• This analysis looks at the relative frequency of large claims, meaning the number of large claims per 10,000 lost-time claims. This metric allows us to project large claim frequency to the most recent period.
• Prior to the Great Recession, large claims were on the rise. Large claim frequency decreased sharply during Great Recession, partly due to job cuts in the construction sector.
• In the recovery period, 2012 and onwards, large claim frequency continues to decrease, although not as sharply, by about 2% per year.

Background

\[Large Claim Frequency = FAST_{ELC} \text{ Frequency}\ + SLOW_{ELC} \text{ Frequency}\ \]

See the background section of Deconstructing Frequency into FAST_ELC and Slow_ELC for more information on the frequency calculations.

Data

• NCCI’s Statistical Plan data. COVID-19 claims are excluded.
• NCCI’s Call 31 data. COVID-19 claims are excluded.

Deconstructing Large Claim Frequency

Key Insights

• SLOW_ELC frequency has decreased significantly since 2008.
• FAST_ELC frequency has increased moderately since 2012.
• The proportion of FAST_ELC and SLOW_ELC claims has changed over time, and they are evenly split as of Accident Year 2021.

Background

Claims are identified as FAST_ELC or SLOW_ELC using the definitions established in the Deconstructing Large Claims slide. For each cohort, the claim counts by Accident Year are developed to a 19th report using the all-year volume-weighted chain ladder method. Lost-time claims are developed to a 10th report, which is assumed to be at ultimate. Similar to large claims, the all-year volume-weighted chain ladder method was employed.

The formulas are as follows:

\[FAST_{ELC} Frequency =\frac{\small\text{At 19th report }\small{FAST_{ELC}}\small\text{ Claim Counts}}{\small\text{(Ultimate Lost-Time Claim Counts / 10,000)}} \] \[SLOW_{ELC} Frequency =\frac{\small\text{At 19th report }\small{SLOW_{ELC}}\small\text{ Claim Counts}}{\small\text{(Ultimate Lost-Time Claim Counts / 10,000)}} \]

The development Factors for the 1st - 10th unit report levels are derived from NCCI’s Statistical Plan data while the development Factors for the 10th - 19th unit report levels are derived from NCCI’s Call 31 data. Refer to the Unit Statistical Data Valuation and Due Dates table in Appendix B for more information on the unit report levels.

Data

• NCCI’s Statistical Plan data. COVID-19 claims are excluded.
• NCCI’s Call 31 data. COVID-19 claims are excluded.

FAST_ELC Frequency

Key Insights

• FAST_ELC burn, spinal cord injury, and TBI frequency has increased by approximately 6% to 7% annually. This could be due, in part, to lifesaving medical advancements for these severe traumas that may otherwise have resulted in the injured worker’s death.
• Other types of fast-emerging claims have remained largely steady.
• Since 2012, the overall frequency for fast-emerging large claims has grown at an annual rate of 2%.

Background

This slide deconstructs the frequency of FAST_ELC claims by diagnosis group for Accident Years 2012 through 2021. The overall FAST_ELC frequency per 10K lost-time claims was calculated using NCCI’s Statistical Plan data, which was then linked to NCCI’s Medical Data Call to identify the diagnosis group. For this analysis, each Accident Year is developed to a 19th report using the all-year volume-weighted chain ladder method.

Data

• NCCI’s Statistical Plan data. COVID-19 claims are excluded.
• NCCI’s Medical Data Call. COVID-19 claims are excluded.

SLOW_ELC Frequency

Key Insights

• SLOW_ELC degenerative disc disorder and pain frequency has decreased by more than 11% annually. This could be due, in part, to increased awareness within the medical community and WC industry regarding pain management and prescribing patterns.
• Other types of slow-emerging large claims have decreased to a lesser extent, and the overall frequency for slow-emerging large claims has decreased at an annual rate of 4.8% since 2012.

Background

This slide deconstructs the frequency of SLOW_ELC claims by diagnosis group for Accident Years 2012 through 2021. The overall SLOW_ELC frequency per 10K lost-time claims was calculated using NCCI’s Statistical Plan data which was then linked to NCCI’s Medical Data Call to identify the diagnosis group. For this analysis, each Accident Year is developed to a 19th report using the all-year volume-weighted chain ladder method.

Data

• NCCI’s Statistical Plan data. COVID-19 claims are excluded.
• NCCI’s Medical Data Call. COVID-19 claims are excluded.

Injured Worker Characteristics

Top Causes and Injuries

Key Insights

• The leading causes of injury for FAST_ELC claims are falls or slips from elevation and motor vehicle accidents. These incidents often result in fractures and injuries to multiple body parts.
• The leading causes of injury for SLOW_ELC claims are strains or injuries caused by lifting. These incidents often result in degenerative disk disorders and lower back injuries.

Background

After claims were identified as FAST_ELC or SLOW_ELC using the definitions established in the Deconstructing Large Claims slide, the most common injury characteristics were identified using the Cause of Injury, Nature of Injury, and Body Part fields as reported in NCCI’s Statistical Plan data.

Data

• NCCI’s Statistical Plan data. Accident Years 2002-2017.

Industry Mix

Key Insights

• The occupation groups shown above account for a significant portion of all large claims, both fast- and slow-emerging.
• For large claims, the cause of injury is more aligned with claim type (FAST_ELC or SLOW_ELC) than occupation group. In other words, two people working at relatively similar jobs could each have large claims with very different causes.

Background

Large claims were identified as FAST_ELC or SLOW_ELC using the definitions established in the Deconstructing Large Claims slide and evaluated as of the fifth unit report level. For each cohort of large claims, the most common occupation groups were identified using the Class Code field as reported in NCCI’s Statistical Plan data. The ranges represent the minimum and maximum shares that the selected occupation groups account for in any single accident year over the given timeframe. For example, construction-related occupations account for up to 31% of all FAST_ELC claims at the fifth unit report level. Refer to the Unit Statistical Data Valuation and Due Dates table in Appendix B for more information on the unit report levels.

Data

• NCCI’s Statistical Plan data. Accident Years 2002-2017.

Industry Mix — SLOW_ELC

Key Insights

• The occupation groups shown above account for up to 13% of SLOW_ELC claims in any given Accident Year.
• These occupation groups often suffer from strains and cumulative injuries that impact soft tissue, but are generally less prone to injuries that typically result in fast-emerging large claims.

Background

Large claims were identified as SLOW_ELC using the definition established in the Deconstructing Large Claims slide and evaluated as of the fifth unit report level. For slow-emerging claims, the most common occupation groups were identified using the Class Code field as reported in NCCI’s Statistical Plan data. The ranges represent the minimum and maximum shares that the selected occupation groups account for in any single accident year over the given timeframe. For example, clerical occupations account for up to 6% of all SLOW_ELC claims at the fifth unit report level. Refer to the Unit Statistical Data Valuation and Due Dates table in Appendix B for more information on the unit report levels.

Data

• NCCI’s Statistical Plan data. Accident Years 2002-2017.

Age at Injury

Key Insights

• Workers aged 35-55 are more likely to have slow-emerging large claims than fast emerging ones. The most common injuries for SLOW_ELC claims, soft-tissue damage, and cumulative strains are often exacerbated with age and presence of comorbidities.
• Fast-emerging large claims are more common in the under-35 and over-55 cohorts.

Background

FAST_ELC and SLOW_ELC claims were identified using NCCI’s Statistical Plan data, which was then linked to NCCI’s Medical Data Call to identify the ages.

Data

• NCCI’s Statistical Plan data Accident Years 2002-2019.
• NCCI’s Medical Data Call.

Injury Type

Key Insights

• Nearly half of all FAST_ELC claims result in death or permanent total disability.
• Over 75% of all SLOW_ELC claims result in permanent partial disability.

Background

Large claims were identified as FAST_ELC or SLOW_ELC using the definitions established in the Deconstructing Large Claims slide. For each cohort of large claims, the most common outcomes were identified using the Injury Type field as reported in NCCI’s Statistical Plan data. Injury types were evaluated as of the most recent unit report level for each claim. Refer to the Unit Statistical Data Valuation and Due Dates table in Appendix B for more information on the unit report levels.

Data

• NCCI’s Statistical Plan data Accident Years 2013-2017.

Severity

Deconstructing Large Claim Severity

Key Insights

• FAST_ELC claims have higher average severity than their slow counterparts.
• Medical spending accounts for 70% of claim costs for FAST_ELC claims and 50% of claim costs for SLOW_ELC claims.

Background

Large claims were identified as FAST_ELC or SLOW_ELC using the definitions established in the Deconstructing Large Claims slide. Next, incurred loss amounts were trended to Accident Year 2021. Incurred indemnity losses are trended using Average Weekly Wage values from the Quarterly Census of Employment and Wages (QCEW) as a proxy. Incurred medical losses are trended using the Personal Health Care index published by the Bureau of Labor Statistics (BLS).

The losses were then developed to a 19th report using the all-year volume-weighted chain ladder method. The development Factors for the 1st - 10th unit report levels are derived from NCCI’s Statistical Plan data while the development Factors for the 10th - 19th unit report levels are derived from NCCI’s Call 31 data. Refer to the Unit Statistical Data Valuation and Due Dates table in Appendix B for more information on the unit report levels.

Data

• NCCI’s Statistical Plan data. Accident Years 2015-2021. COVID-19 claims are excluded.
• NCCI’s Call 31 data. COVID-19 claims are excluded.

Medical Payment Loss Development

Key Insights

• Though fast-emerging large claims are initially very large, much of the medical payment occurs in a short time from the accident, known as the “acute phase” in medical terms, and the subsequent costs for these claims are not as large comparatively. Put into context with a single claim, this flattening of the curve could indicate the stabilization of the injured worker’s condition.
• Slow-emerging claims exhibit far more relative growth—more than tripling in value from initial medical payments. Put into context with a single claim, this escalation in medical costs could indicate that the patient often faces some sort of deteriorating condition and intensifying treatment.

Background

This slide deconstructs the loss development of large claims by FAST_ELC and SLOW_ELC designation, and provides non-large claims as a benchmark for which to compare. Medical payments are broken out into various categories and trended to December 2021 using inflation factors from the Bureau of Labor Statistics (BLS); refer to the Medical Cost Categories and Inflationary Factors table in Appendix B for more information. The trended medical payments by category are then developed to a ten-year claim maturity (i.e., ten years after the initial accident) using the all-year volume-weighted chain ladder method. Finally, the trended and developed medical payments by category are recombined to produce the overall medical values displayed above.

The “growth factors” displayed in this slide start at 1 and can be interpreted as the increase in medical costs after the first year. For example, the yellow line displays a growth factor of 1.7 ten years after the injury for fast emerging claims, indicating 70% growth. So, if a fast-emerging large claim had $1M of medical treatment costs in the first 365 days after the initial injury, then this claim would incur an additional $700K of medical costs in years 2 through 10, totaling $1.7M over the first ten years.

Data

• NCCI’s Medical Data Call — Accident Years 2012-2021. COVID-19 claims are excluded.
• Bureau of Labor Statistics inflationary factors — refer to Appendix B for more information.

Medical Service Costs — Non-Large Claims

Key Insights

• During the first year after injury, medical service costs for non-large claims are evenly split between physician and hospital services, about 40%-45% each.
• In the tenth year after injury, the composition of medical services differs drastically, with larger shares of prescription drugs and supplies.

Background

This slide deconstructs the medical costs for non-large claims in Year 1 and Year 10 to demonstrate how the medical services for these claims evolve over time. Medical payments are broken out into various categories and trended to December 2021 using inflation factors from the Bureau of Labor Statistics (BLS); refer to the Medical Cost Categories and Inflationary Factors table in Appendix B for more information. The trended medical payments by category are then developed to a ten-year claim maturity (i.e., ten years after the initial accident) using the all-year volume-weighted chain ladder method. Finally, the trended and developed medical payments by category are recombined to produce the respective shares displayed above.

Note that this is an incremental analysis. Meaning, Year 1 represents all medical costs incurred within the first 365 days after the initial accident, and Year 10 represents all medical costs incurred in the tenth year after the accident. The Year 10 metrics displayed above are not a summation of Years 1 through 10.

Data

• NCCI’s Medical Data Call — Accident Years 2012-2021. COVID-19 claims are excluded.
• Bureau of Labor Statistics inflationary factors — refer to Appendix B for more information.

Medical Service Costs — Selected FAST_ELC Claims

Key Insights

• Hospital costs account for over two-thirds of all medical payments in the first year after injury when looking at the selected FAST_ELC injuries (burns, spinal cord injuries, and TBIs).
• In the tenth year after injury, the composition of medical services differs drastically, with the Other category— which is comprised of home health, skilled nursing, and transportation— accounting for more than half the costs.

Background

This slide deconstructs the medical costs for the selected FAST_ELC claims in Year 1 and Year 10 to demonstrate how the medical services for these claims evolve over time. Medical payments are broken out into various categories and trended to December 2021 using inflation factors from the Bureau of Labor Statistics (BLS); refer to the Medical Cost Categories and Inflationary Factors table in Appendix B for more information. The trended medical payments by category are then developed to a ten-year claim maturity (i.e., ten years after the initial accident) using the all-year volume-weighted chain ladder method. Finally, the trended and developed medical payments by category are recombined to produce the respective shares displayed above.

Note that this is an incremental analysis. This means Year 1 represents all medical costs incurred within the first 365 days after the initial accident, and Year 10 represents all medical costs incurred in the tenth year after the accident. The Year 10 metrics displayed above are not a summation of Years 1 through 10.

Data

• NCCI’s Medical Data Call — Accident Years 2012-2021. COVID-19 claims are excluded.
• Bureau of Labor Statistics inflationary factors — refer to Appendix B for more information.

Medical Service Costs — All Other FAST_ELC Claims

Key Insights

• Hospital costs account for more than two-thirds of all medical payments in the first year after injury when looking at all other FAST_ELC injuries.
• In the tenth year after injury, the Other and Medical Supplies categories account for more than half of all medical payments. The Medical Supplies spend is partially attributed to prosthetics, which are typically replaced every 3-5 years.

Background

This slide deconstructs the medical costs for all other FAST_ELC claims in Year 1 and Year 10 to demonstrate how the medical services for these claims evolve over time. Medical payments are broken out into various categories and trended to December 2021 using inflation factors from the Bureau of Labor Statistics (BLS); refer to the Medical Cost Categories and Inflationary Factors table in Appendix B for more information. The trended medical payments by category are then developed to a ten-year claim maturity (i.e., ten years after the initial accident) using the all-year volume-weighted chain ladder method. Finally, the trended and developed medical payments by category are recombined to produce the respective shares displayed above.

Note that this is an incremental analysis. Meaning, Year 1 represents all medical costs incurred within the first 365 days after the initial accident, and Year 10 represents all medical costs incurred in the tenth year after the accident. The Year 10 metrics displayed above are not a summation of Years 1 through 10.

Data

• NCCI’s Medical Data Call — Accident Years 2012-2021. COVID-19 claims are excluded.
• Bureau of Labor Statistics inflationary factors — refer to Appendix B for more information.

Medical Service Costs — SLOW_ELC Claims

Key Insights

• Slow-emerging claims are similar to our other claim cohorts in that hospital costs account for the majority of all medical spending in the first year after injury.
• However, in the tenth year after injury, slow-emerging claims have significant medical expenditure on prescription drugs, and this sets them apart from their fast counterparts.
• NCCI’s recent publication on Medical Inflation and Workers Compensation Costs noted a 6% annual decline in prescription drug utilization since 2012, which equates to a 40% decline between 2012 and 2021.

Background

This slide deconstructs the medical costs for slow-emerging claims in Year 1 and Year 10 to demonstrate how the medical services for these claims evolve over time. Medical payments are broken out into various categories and trended to December 2021 using inflation factors from the Bureau of Labor Statistics (BLS); refer to the Medical Cost Categories and Inflationary Factors table in Appendix B for more information. The trended medical payments by category are then developed to a ten-year claim maturity (i.e., ten years after the initial accident) using the all-year volume-weighted chain ladder method. Finally, the trended and developed medical payments by category are recombined to produce the respective shares displayed above.

Note that this is an incremental analysis. Meaning, Year 1 represents all medical costs incurred within the first 365 days after the initial accident, and Year 10 represents all medical costs incurred in the tenth year after the accident. The Year 10 metrics displayed above are not a summation of Years 1 through 10.

Data

• NCCI’s Medical Data Call — Accident Years 2012-2021. COVID-19 claims are excluded.
• Bureau of Labor Statistics inflationary factors — refer to Appendix B for more information.

Distribution of Other Medical Cost Category — FAST_ELC

Key Insights

• Home health services account for the majority of “Other” expenses for FAST_ELC claims.

Background

This slide displays the composition of the “Other” medical cost category for FAST_ELC claims. Medical payments for FAST_ELC claims are broken out into the categories displayed above and trended to December 2021 using inflation factors from the Bureau of Labor Statistics (BLS); refer to the Medical Cost Categories and Inflationary Factors table in Appendix B for more information. The trended medical payments by category are then developed to a five-year claim maturity (i.e., five years after the initial accident) using the all-year volume-weighted chain ladder method.

Data

• NCCI’s Medical Data Call — Accident Years 2012-2021. COVID-19 claims are excluded.
• Bureau of Labor Statistics inflationary factors — refer to Appendix B for more information.

Injured Worker with a Low Back Strain

Key Insights

• The example scenario depicted in the timeline above demonstrates how medical costs can balloon over time for a slow-emerging claim.

Background

This slide highlights key points in the medical journey of a hypothetical injured worker with a low back strain to demonstrate how a SLOW_ELC claim might develop over time. The services shown are an illustration of such claims.

Data

• NCCI’s Medical Data Call

Prescription Drugs — SLOW_ELC

Key Insights

• Drug spend for both opioids and non-opioids has consistently decreased for slow emerging large claims.

Background

This slide displays the share of overall medical costs for SLOW_ELC claims that are attributable to prescription drugs, broken out by opioids and non-opioids, for Accident Years 2012 through 2019. Medical payments for SLOW_ELC claims are broken out into various categories and trended to December 2021 using inflation factors from the Bureau of Labor Statistics (BLS); refer to the Medical Cost Categories and Inflationary Factors table in Appendix B for more information. The trended medical payments by category are then developed to a five-year claim maturity (i.e., five years after the initial accident) using the all-year volume-weighted chain ladder method.

Data

• NCCI’s Medical Data Call — Accident Years 2012-2019.
• Bureau of Labor Statistics inflationary factors — refer to Appendix B for more information.

Medical Inflation