Managing liability ENIDs using Radar

Radar

The Radar service provides an expert view of science-based changes to liability insurance exposure and provides quantitative estimates of that exposure. Some examples from the Radar database are appended here:

• American football and brain disease,
• LED lights and eye injury,
• wood dust and nasal cancer and,
• 3D printing.

History

The Radar service was first developed under contract to ABI[1] as a collective work on emerging liability risks. Radar was then commercialised by Re: Liability (Oxford) Ltd[2]. As ABI evolved, the service was passed to liability insurers to take up individually. Liability emerging risks are now more often referred to as liability ENIDs[3], for example in Solvency II guidance.

Aim

The aim is to inform judgement of factors that could drive changes to liability insurance exposure and to estimate the resulting size of that change.

Method

This follows the familiar method of identify, evaluate and take action. Identification of Radar ‘echoes’ is confined to publicly available science, technology, political and legal papers. There has to be substantial and verifiable evidence of a potential change in liability exposure to trigger further analysis. The Radar service is particularly well developed for the evaluation of science evidence and the potential liability exposure implications.

Subscribers receive narrative reports and interactive Excel based models.

Fit to your business

Subscribers may have an emerging risks specialist but a committee is the more usual forum for deciding the issues and shaping change.

The committee usually agrees and sets action thresholds and conditions e.g.:

• How much can the liability business plan change before we need to do something? Or,
• Would it be cheap and easy to make a useful/profitable change?

Two decision thresholds are normally set for liability ENIDs; the first is whether or not to investigate further, the second is an indicator of the need for action.

The committee Chair ensures that Radar reports and models and reports from other sources are targeted at those best equipped to make a firm-specific evaluation on that topic and to report back. Typical considerations include: do we insure that sector (and how much), do we exclude this risk, is the new liability likely to exceed either decision threshold (within a defined timescale)… The committee sets out what needs to change before a review is called for and tasks people to monitor that change.

Potential business actions include changes to: price, risk appetite, wordings, new products, risk improvement, reserving, PR, relations with brokers, developing a defence…

Continuous improvement for all those involved in managing liability emerging risks is almost a certainty. Knowledge, judgement skills, leadership are all tested when thinking through liability ENIDs.

The Radar Product

Radar reports are circulated to all subscribers and appear in a quarterly journal.

There is a meeting every quarter.  Subscribers are free to propose topics for presentation.

There are nearly 4,000 reports held in a searchable database. Use of the database helps identify insights and confirm basic facts such as number of potential victims, severity of injury, likely date of knowledge etc. Subscribers use the database to develop their own reports.

Consultation on any items mentioned in the Radar reports is free.

The new development from 2017 is the liability ENID modelling software. New models are provided free to those subscribers who were on the books during their development. New subscribers pay an entry fee to benefit from those models. The plan is to develop four new models a year. Thus far, subscribers have agreed that nitrites in food and anti-microbial resistance would be useful in 2018.

Price

Price is not usually the deciding factor. Radar subscribers include Insurers of all sizes, share value, capacity to manage change, expertise and business ambitions.

Who subscribes?

UK registered liability insurers and defendant-only law firms.

Detailed reports

Sometimes an individual insurer has a specific liability exposure or reserving problem. Private reports can be made.

Contact

Dr Andrew Auty andrew@reliabilityoxford.co.uk

 

Sample reports from the Radar journal

These are divided here into ‘probably significant’ and ‘could be significant’.

 

 

Probably Significant

Introduction

The number of concussions experienced in third party insurance scenarios is very large. The rate of degenerative brain disease is also very high. If concussions or mild concussions (mTBI) were a cause of degenerative neurological disease the problem for liability insurance would be significant. Radar has quantified these frequency factors. Until clarity emerges, new science reports in this area are judged to be ‘probably significant’.

One recent Radar report goes as follows.

American football

This study found there was no evidence of cognitive deficit in middle aged former high school American football players.

More broadly, it would appear that associations between TBI and recognised brain disease are limited.

 If TBI has little effect then why would you expect mTBI to have an effect?

Product areas: PL, EL, GL

Evidence from:

SK Deshpande et al. JAMA Neurol. (2017) Vol.74 (8) p 909-918

Association of Playing High School Football with Cognition and Mental Health Later in Life

More than 1 million students played high school American football in 2014

Aim

To assess evidence of cognitive detriment in those who played American Football while at high school.

 

Method

This research was developed from a cohort study begun in 1957, Wisconsin. It was based on a random sample of 10,317 “High School” graduates[4], representing around one third of all such graduates.

Participation in American football was established from year books, but did not include data on concussions.  People with participation in other impact sports were excluded. 834 footballers were identified out of 2,692 men eligible for the study.

Covariates included family background, adolescent IQ, educational attainment, participation in leisure time physical activity[5] at age 35 and alcohol consumption prior to cognitive testing. The control group was taken from all those who did not play American football. These were further divided into those who did or did not play impact sports at High school.

Cognitive and psychological wellbeing were assessed at age 54, 65 and 72.This study analysed data from the age 65 assessment. Outcome variables included depression, cognitive speed.

A small number of participants were assessed for variants of the APOE gene; thought to be a risk factor for Alzheimer’s.

 

Results

IQ at baseline was slightly higher in football players. At 35, football players were more likely to be involved in physical activity (OR= 1.23 (1.02 to 1.48)). The APOE genotype had no effect on outcome variables. Alcohol consumption at ages 54, 65 and 72 was the same for players and controls.

At age 65 former football players had lower scores on the depression scale. Smaller scores indicate a lower degree of depression.

There was no evidence of cognitive impairment or benefit.

Age 54, 65 and 72 trajectories for letter fluency, word recall, depression, and intelligence were the same for football players and controls.

 

Authors’ summary

Among men graduating from high school in Wisconsin in 1957, we did not find evidence that playing football had a negative long-term association with cognitive functioning and mental health at 65 and 72 years of age.

 

Comment

There was no data on playing style, skill level, aerobic fitness, BMI, diabetes, teenage substance use, progression to professional football or field position and no data on concussion history. It could be that specific sub populations were at higher risk of decline but if so, when averaged over the whole population, such effects were not big enough to give significant associations. Teenage football playing per se did not have a measurable detrimental effect on cognitive outcomes.

Alzheimer’s and Parkinson’s would therefore not be expected to be revealed earlier in former football players.

If teenage footballers were diagnosable with CTE at age 50+, it made no difference to their cognitive ability in this research. But no-one can say whether CTE was present of not in either footballers or controls.

There is no reason to suggest that teenage footballers in the 1950s were unlikely to be concussed. If concussion causes CTE then there should be some CTE in the footballer cohort. If present, it was either not present at a higher rate than in the controls or, CTE has no effect on cognitive performance.

It could be argued that the protective effect of physical activity offset any cognitive detriments due to head trauma-related pathology. But there is no evidence to inform such a speculation. In another study it might be shown that footballers without concussion, or non-contact sports people had an unusually high cognitive performance, for example.

Teenage IQ and middle-aged consumption of alcohol were the same in footballers and controls. Those who argue that cognitive deficits are caused by increased self-abuse and by lower innate cognitive capacity would not find any support from this study.

 

Some context

Opinion* on this study began with the following statement:

It is increasingly recognized that moderate to severe traumatic brain injury (TBI) (typically defined as loss of consciousness for >30 minutes) increases the risk for dementia in older adulthood, a conclusion based primarily on epidemiologic research.

Moderate brain injury is described as loss of consciousness or amnesia for half an hour or more, severe is defined as 24 hours or more. In support of this opinion it may be useful to read a BMJ article[6] from 2000, reporting on Alzheimer’s disease (AD). This was based on documented closed head injury in 548 World War II veterans.

Both moderate head injury (hazard ratio [HR] = 2.32; CI = 1.04 to 5.17) and severe head injury (HR = 4.51; CI = 1.77 to 11.47) were associated with increased risk of AD. Results were similar for dementia in general. The results for mild head injury were inconclusive.

This was an important study[7] in that it had a low risk of reverse causation. Similar studies[8] in the general population do not find such an association with AD. The above opinion* may be a true reflection of unbiased assessment of all the literature, or it may not. To say that something is ‘increasingly recognised’ is too vague a statement to be interpreted in the context of an emerging risk.

 

It is quite clear that other authors would not make the same statement:

• In a recent systematic review[9], the authors reported a lack of certainty:

Failure to establish such a link may be related to methodological problems in the studies. To shed light on this dilemma, future studies should use a prospective design, define the types and severities of TBI and use standardized AD and TBI diagnostic criteria.

• In another review[10]

…when considering the status of unconsciousness, head injury with loss of consciousness did not show significant association with dementia (RR = 0.92, 95% CI 0.67–1.27) and AD (RR = 1.49, 95% CI 0.91–2.43).

A review of reviews would seem to be in order.

• Focussing on just the higher quality studies[11]

Pooled analysis showed that a history of TBI with loss of consciousness was not associated with AD dementia or the neuropathologic features of AD;

• And[12]

 The results showed no effects of TBI history on cognition or AD biomarkers.

‘Increasingly recognised’ is not a useful term when the evidence is considered in detail.

 

Why does it matter?

The argument would be that if moderate and severe TBI increase the risk of AD then might it be possible that multiple mTBI could have the same effect? This would add biological coherence. However, if moderate and severe TBI DO NOT increase the risk of AD then why would you expect mTBI to do so? It would appear that the weight of evidence (when assessed for quality) is against the proposed association between TBI and AD. This doesn’t rule out mTBI causing AD but does reduce the potency of the argument.

The opinion* goes on to say that associations between CTE and sport have not been quantified. The rates of CTE in sports and non-sport people are not known, risk factors for CTE are not known. The opinion finishes with:

Future studies of this type should include comprehensive measurements of outcomes, including thorough neuropsychological, neuropsychiatric, and neurologic assessments to investigate the potential clinical phenotype of concussive-related neurodegeneration and its symptom course over time.

It would seem from this that not much if anything is known with sufficient certainty.

Similar studies of amyotrophic lateral sclerosis and head injury have been undertaken. A recent review [13]noted that an association between ALS and head injury was found (OR = 1.45 (1.21–1.74)) but weakened to insignificance when a three year lag time was allowed for; OR 1.16 (0.84–1.59). The authors propose reverse causation is a likely explanation.

A recent and powerful study[14] of concussion and multiple sclerosis (MS) found a positive association and a doubled risk for those with more than one concussion. This is reviewed in the next Radar article.

 

What would Radar subscribers make of this?

Alzheimer’s and Parkinson’s disease seem unlikely to be caused or accelerated by concussions or American Football.

If CTE was present, it made no difference to cognitive performance and so would be a biological change without significant impairment. Other Radar reports deal with the validity of CTE. The common law on impairment/biological change has been changed very recently.

Summary statements made by researchers may not adequately represent the field of research even if those researchers are involved in that research. It should be borne in mind that research is an industry just like any other. The problem for insurers would be that a court might be persuaded that there potentially was or actually was expert consensus. This can give rise to costs and may encourage baseless claims.

Given that the delay between football and the manifestation of degenerative brain disease can be decades the problem is for policies with causation wordings; accident and health type policies would usually be safe from exposure. Increased reserving would be a normal response if new claim types were emerging. The justification would be on the basis of “more likely than not”. In my view for Alzheimer’s, Parkinson’s and CTE the justification is not in place for payment for personal injury but could be in place for the cost of defending on causation.

---

Less obviously significant

The Radar journal includes short reports such as the following:

SCHEER (2017)

Potential risks to human health of Light Emitting Diodes 10 (LEDs)

Blue LEDs may cause car crashes, but there are signs of direct eye damage too.

Key findings:

The Committee concluded that there is no evidence of direct adverse health effects from LEDs emission in normal use (lamps and displays) by the general healthy population. There is a low level of evidence that exposure to light in the late evening, including that from LED lighting and/or screens may have an impact on the circadian rhythm. At the moment, it is not yet clear if this disturbance of the circadian system leads to adverse health effects.

Published studies show that the blue light-weighted (for eyes) radiance from screens is less  than 10% of the blue light photochemical retinal hazard limit, assuming viewing greater than about 3 hours (acute exposure),.

Although there are cellular and animal studies showing adverse effects raising concerns particularly in susceptible population, their conclusions derive from results obtained using exposure conditions that are difficult to relate to human exposures or using exposure levels greater than those likely to be achieved with LED lighting systems in practice.

The SCHEER is concerned about the high-luminance exterior sources used on some vehicles.

[Editor’s note: the main SCHEER concern with headlamps is the increased experience of glare (which may cause car crashes).

A paper[15] which may have been missed by SCHEER concerns retinal damage caused by the same sort of display as used in smartphones.  Emissions at 460 nm were 150 lux and were applied for 0, 0.5h 1.0h and 3 h every day for 28 days. Normal lab light was 250 lux white light. The 3h group suffered retinal damage. The authors refer to it as akin the age-related macular degeneration.]

What would Radar subscribers make of this?

If this expert report is any guide, the date of knowledge for blue light LED related injury has not arrived. However, there is evidence to suggest that this is an area to keep as a watching brief.

Motor insurers might want to consider changes in headlamp technology as a potential cause of accidents for which the insured driver would rarely be at fault.

_______

 IIAC Cmnd 9499 (2017)

Nasal carcinoma and occupational exposure to wood dust

10% of sinonasal cancer may be caused by wood dust exposure. 10 cases a year in the UK.

IIAC propose to extend the terms of prescription to include more people with occupational exposure to wood dust in the course of the machine processing of wood. This could include outdoor operations. There is a suggestion that daily exposure for at least 5 years would be needed to pass the doubled risk test. But this is not expressed in such clear terms.

[Editor’s note: there are around 60 new sinonasal cancer cases each year UK men. According to HSE the wood dust attributable fraction is[16] 10% (3.9% to 19.6%). Mineral oils had AF = 13.8% (1.4% to 38.9%) making wood dust the second most likely cause.]

What would Radar subscribers make of this?

6 cases of wood dust-related sinonasal cancers each year in the UK would usually be below the threshold for further investigation of liability exposure.

_______

 European Agency for Safety and Health at Work Discussion Paper

3D Printing and Additive Manufacturing – The Implications for OSH

3D printers produce emissions that should be controlled.

Key statements:

The aim of this discussion paper is to introduce 3D printing and explore its possible impact on both the existing and the new working environment.

 A desktop 3D printer will cost around €1,000. A professional 3D printer for prototyping and limited-edition production will cost anywhere between €2,000 and €20,000.

Despite the suggestion in the title, 3D printing and additive manufacture are synonymous in this paper.

Printers work either as a controlled extrusion of self-adhesive layers or by extrusion of materials which are then further treated (heat, UV light, laser) to bind them together. But both seem to present the same types of H&S hazards (heat or light energy, fumes from the assembly point and dusts from handling powdered substrates).

Printing is associated with the emission of nanoparticles and vapours. Both may contain known toxins e.g. isocyanates. Named substrates include: polylactic acid (CAS number 26100-51-6, a biodegradable extrusion plastic sourced from plants), acrylonitrile butadiene styrene (CAS number 9003-56-9, a synthetic extrusion plastic (used to make logo bricks)) polyamide (nylon), epoxy resins, aluminides (a mix of nylon and aluminium particles, laser sintered for mechanical strength), polysulfon (applied as an ink which dries hard, or used as an extrusion plastic), polyphenylsulfon (an extrusion plastic), metals[17], and nano-carbons.

Mentioned in the report are 3D pens[18]:

…the 3D pen is a printer that looks like a pen and enables you to draw in 3D. For health and safety reasons these new handy 3D printers must be thoroughly tested and strictly regulated because the heating can pose safety threats. More and more of these pens use UV light.

The report delves extensively into sociology, e.g. the loss of skilled trades, the establishment of small local businesses. Of greater interest for liability is:

…the legal division between the designer, the manufacturer and the entrepreneur is vague, liability in case of malfunction or inferior quality is unclear. This creates uncertainty in liability. With all sorts of products having become freely available by (illegal) downloading from the internet, piracy and copyright infringement are imminent.

 

For the future:

Food

Currently, mostly liquid foods, such as chocolate and pancake batter, are used with 3D printing. In the near future, 3D printing will be used for raw food that will later undergo treatment such as heating, or will be processed by natural processes such as fermentation or germination. This will pose new challenges to hygiene, safety and general working conditions (clean air, ergonomics, and so on).

4D printing

Smart materials have one or more properties that can be changed significantly in a controlled way by external stimuli, such as temperature, force, light, moisture, pH, and electric or magnetic fields. When manufactured with a 3D printer, these smart materials can form objects that respond to their environment by changing shape, tactility or hardness.

Bioprinting

The 3D printing of organic and/or living tissue is referred to as bioprinting. Bioprinters output cells from a bioprint head.

As with smart materials, this technique poses risks to health and hygiene. Furthermore, it raises ethical issues.

Nano-printing

By combining 3D printing with nanotechnology, it will be possible to shape objects at a nano- or molecular level. In theory this means that through additive manufacturing it will be possible to manufacture any form of object of any kind of material, in any shape or volume.

[Editor’s note: Plastic nanoparticles will often be biopersistent even if derived from plant materials. 3D printers employing such materials should be used with covers in place and LEV. This would be more difficult for 3D pens which may have to rely on directed airflow to prevent inhalation.

The unique properties of nanomaterials, such as very low melting point and high aggregation number, make them potentially ideal for additive manufacturing. The scope of materials which could be used for 3D printing is greatly expanded by presenting them in nanoform. Nano layers could be readily annealed during/post deposition.

Food printers should be especially careful to avoid bacterial contamination and the emission of food allergens.]

 

What would Radar subscribers make of this?

Radar readers will know that there is a liability risk assessment tool for nanomaterials. http://www.reliabilityoxford.co.uk/nanomaterials/ . The tool could be used to differentiate routine risks and nano-specific risks. There is no charge for the use of this tool.

3D printing technology will continue to grow in scope, volume and complexity. Copyright theft claims will be increasingly common, attribution of liability to designer, manufacturer or pirate will be a complex area but could be clarified in wordings.

[1] Association of British Insurers. https://www.abi.org.uk/ . Dr Andrew Auty served on the ABI Liability Committee.

[2] http://www.reliabilityoxford.co.uk/

[3] Events not in data (ENIDs).

[4] High school refers to ages 14 to 18 inclusive.

[5] It was thought that physical activity at age 35 would have a protective effect against cognitive decline.

[6] BMJ. (2000) Vol.321 (7269): 1100. Head injury doubles the risk of Alzheimer’s disease. This is based on Neurology (2000) Vol.55 p 1158-66.

[7] But the result fails the Radar precision test.

[8] Neurology. (2014) Vol.82 (1) p 70-6.

[9] Ann Phys Rehabil Med. (2017) Vol.60(5) p 347-356

[10] https://doi.org/10.1371/journal.pone.0169650 Yanjun Li et al. Head Injury as a Risk Factor for Dementia and Alzheimer’s disease: A Systematic Review and Meta-Analysis of 32 Observational Studies.

[11] JAMA Neurol (2016) Vol.73 p 1062–1069. Crane PK, Gibbons LE, Dams-O’Connor K, et al. Association of traumatic brain injury with late-life neurodegenerative conditions and neuropathologic findings.

[12] Alzheimer Dement (2017) Vol.3 p 177–188. Weiner MW, H D, Hayes J, et al; Department of Defense Alzheimer’s Disease Neuroimaging Initiative. Effects of traumatic brain injury and posttraumatic stress disorder on development of Alzheimer’s disease in Vietnam veterans using the Alzheimer’s disease Neuroimaging Initiative: preliminary report.

[13] Eur J Epidemiol DOI 10.1007/s10654-017-0327-y. Y Watanabe et al. Meta-analytic evaluation of the association between head injury and risk of amyotrophic lateral sclerosis.

[14] Scott Montgomery, Ayako Hiyoshi, Sarah Burkill, Lars Alfredsson, Shahram Bahmanyar, Tomas Olsson. Concussion in adolescence and risk of multiple sclerosis. Ann Neurol (2017) Vol.82 p 554–561.

[15] Toxicological Sciences. (2017) Vol.157(1) p 196–210. Cheng-Hui Lin, Man-Ru Wu, Ching-Hao Li, Hui-Wen Cheng, Shih-Hsuan Huang, Chi-Hao Tsai, Fan-Li Lin, Jau-Der Ho, Jaw-Jou Kang, George siao, and Yu-Wen Cheng. Periodic Exposure to Smartphone-Mimic Low-Luminance Blue Light Induces Retina Damage Through Bcl-2/BAX-Dependent Apoptosis

[16] http://www.hse.gov.uk/research/rrpdf/rr933.pdf

[17] Metals are the fastest-growing segment of 3D printing, with printer sales growing at 48% and material sales growing at 32%. These so far include gold powder, palladium, silver, copper, bronze and tungsten, titanium alloys, cobalt alloys, nickel alloys and aluminium alloys. https://www.3dprintingmaterialsconference.com/3d-printing-materials/metals-are-the-fastest-growing-segment-of-3d-printing-metal-sales-growing-by-32/

[18] These are very cheap indeed. http://www.currys.co.uk/gbuk/computing-accessories/printers-scanners-and-ink/ink-and-cartridges/3d-printer-cartridges/polaroid-play-3d-pen-10160143-pdt.html