Calling All Torts Professors: Does Veterans Law Have a Coherent Theory of Proximate Cause?

The essence of veterans law is providing compensation for conditions that are related to military service.  Does the system have a consistent approach to proximate causation?  I am not so sure.  When I teach veterans law, I begin the section on causation by reviewing Palsgraf v. Long Island Railroad Co. and use that to frame the discussion of causation.  Using Palsgraft as a reference point always leads to a lively discussion.

There is a great article waiting to be written examining all of the different entitlement theories and applications in veterans law.  I do not think that there is any question that different groups of veterans feel like they are treated differently from one another.  Whether this can be justified is a very good question.

Let us look at some examples.  Of course, disabilities related to being shot or blown up are compensable.  However, what about the soldier who tears his ACL in a pickup basketball game?  Yes, that is compensable.  There is no concept of “on duty” or “off duty” for veterans benefits.  Whether you are injured in a combat zone or by a burglar that breaks into your off-base house in the middle of the night, you can be granted compensation for your injuries.  It is all the same — unless, of course, your disability is related to smoking or alcohol.  Those are not covered — unless, however, if your abuse of alcohol were caused by a service-related mental health condition.  What if you did not start smoking until you joined the military and found that Uncle Sam was putting cigarettes in your rations?  Sorry, there is no exception for that.

Aside from what counts as having occurred in the line of duty, there are also issues with probability.  The basic probability rule is 50/50.  If it is as least as likely as not that something is related to service, then it is compensable.  A tie goes to the veteran.  This is known as “the benefit of the doubt” (not to be confused with beyond a reasonable doubt, although many veterans do), and the U.S. Court of Appeals for Veterans Claims has traced the origins of this doctrine back to the Civil War.  See Gilbert v. Derwinski, 1 Vet. App. 49 (1990).  However, what about all the special provisions that have been adopted over time?  Are they based upon the 50/50 standard?

If you are a Vietnam veteran, then you are presumed to have been exposed to Agent Orange even if you only had “boots on the ground” for five minutes while you changed planes once in Saigon.  Under 38 U.S.C. § 1116, the Secretary will presume a condition is related to Agent Orange exposure if the scientific evidence shows “a positive association” between the condition and dioxin exposure.  Is that the same as 50/50, or does it just mean that there is a statistical difference between the exposed population and the unexposed population?  Let us look at this interesting student note that considered the extension of the presumption of causation to ischemic heart disease:

The problem is becoming highlighted now that Vietnam veterans are getting older and starting to exhibit illnesses that are more prevalent among an older population.  For example, Dr. Diane Bild, a cardiologist for the National Institute of Health, testified that 80 to 90 percent of men aged sixty to seventy-nine can be expected to have either symptomatic or asymptomatic ischemic heart disease.  Major causes of ischemic heart disease include smoking, high LDL cholesterol, low HDL cholesterol, high blood pressure, and diabetes.  Also thought to contribute are a sedentary lifestyle, poor diet, obesity, and psychosocial factors such as stress and depression.  While the [Institute of Medicine] has recently found that dioxin exposure is associated with ischemic heart disease, this association is modest and most of the studies cited failed to account for other risk factors.  The nine primary studies considered in the review of ischemic heart disease did control for age, but their control for other risk factors varied greatly.  Additionally, only two of these studies actually dealt with Vietnam veterans.  As a result of these shortcomings, a veteran may receive disability benefits under the Agent Orange Act even if he has every other possible risk factor for ischemic heart disease.

Meagan E. Fassinger, Striking A Better Compromise: Suggested Revisions to the Agent Orange Act of 1991, 21 Fed. Circuit B.J. 193, 211-12 (2011).  If 80 to 90% of men of the age of Vietnam veterans have ischemic heart disease, is it as least as likely as not that any given veteran’s ischemic heart disease is related to service?

What if you are a Vietnam veteran, but you suffer from an exceedingly rare condition?  Then your problem may be that there are not enough cases to determine whether there is a positive association.  Judge Newman of the Federal Circuit has argued that it is unjust to apply the same causation standard to both common conditions that are easier to study and to rare conditions that are very difficult to study.  See Bastien v. Shinseki, 599 F.3d 1301, 1308 (Fed. Cir. 2010) (Newman, J., dissenting).

What if you served in the first Gulf War and you are having health problems?  There is a “positive association” clause in 38 U.S.C. § 1118, but to what were those veterans exposed?  It is still not clear.  Does the “positive association” clause treat these veterans differently because of the difficulty in determining when there is great uncertainty in determining what “biological, chemical, or other toxic agent[s]” to which they were exposed?  There is a special provision in 38 U.S.C. § 1117 that allows those veterans to obtain benefits for “an undiagnosed illness,” but how often do you see a doctor give no diagnosis?  How many conflicting diagnoses do you need to make it as least as likely as not that you do not have a valid diagnosis at all? 

Additionally, what about the special presumptions for veterans who were exposed to radiation or who were prisoners of war?  Are those rules consistent with the 50/50 standard?  What science went into those presumptions, and have those findings aged well?

In a system with so many paths to benefits, it is hard to make each one feel connected to the same fundamental standard.  Of course, maybe the presumptions are not trying to treat those veterans equally with veterans pursing the basic path to compensation.  If so, what is the justification for the difference?  Maybe we just have a patchwork of different provisions adopted at different times in response to different political stimuli that cannot be reconciled into a consistent framework.  Is that a problem?

A specialist in tort theory may enjoy trying to create a unified theory of causation for veterans law, and  ideas for how to accomplish it would be interesting.

Comments

[DrGrishka]

Well, if one takes the view that an active duty soldier is always on duty, subject to be called at any time, and bound by various military rules 24/7 (unlike an employee who is usually not bound by company's policies once he punches out), it would then make sense that all actions that the soldier engages in are part of his employment. So that answers the basketball/ACL example, I think.

As to presumptions, I am not sure that those have to be squared with the Torts theory so much as with administrative ease. Especially since it is not an irrebuttable presumption. It also seems to me to be consistent with the multiple causes rule. And under that rule, it is generally the tortfeasor who has to exculpate himself rather than the tort victim having to prove that the disease is not attributable to an alternative cause.

[James Ridgway]

So is it true that administrative ease explains the spectrum of theories of tort recovery better than cost/benefit analysis or anything else? I'm just curious. I've never had the time to research the literature on Grand Unified Theories of Torts. I always assumed that tort theories were more normatively driven than determined by the practicalities of what can and cannot be judicially managed.

[Brad]

A theory of causality would be interesting, but I think this exchange shows that a more valuable contribution would be a thoery of presumptions. Too often the VA and commentators treat presumptions as codifications of the 50/50 rule. This is the basis for the "administrative ease" argument: in factual circumstances where the causality is likely to be 50% or higher, we'll expedite the analysis by automatically granting service connection. But this is not the "definition" of a presumption. There are other, and potentially more important, reasons for creating a presumption. Some of these were outlined in 69 FR 60,083. For example: unavailability of primary evidence and low quantity of secondary research. These justifications do not address probability, they address evidentiary problems. These examples show that presumptions are not merely short-hand for probability analysis.

As practitioners know, however, the VA tends to treat presumptions as short-hand for probability analysis. This means that the VA only adopts new presumptions when it can show that the 50/50 standard is exceeded. And it means that adjudicators tend to find service connection only when relevant presumptions are satisfied, as if any other circumstances must be less than 50/50. Both of these are errors, I believe, and we might remedy them with a better theory of presumptions.

[James Ridgway]

How disconnected from causation could a theory of presumptions really be? It seems to me that you quickly reach the point where administrative ease and other concerns run up against the political backlash caused when different groups of veterans perceived that they are being treated unequally. In addition, getting lay adjudicators to correctly administer the system gets very hard very quickly when the rules stop making intuitive sense.

I'm just throwing that out there. I would be very interested in seeing a theory of presumptions that looked more broadly than just causation. But I'm concerned about the practical problems that occur when you try to deploy a sophisticated, counter-intuitive theory through instructions to non-attorney agents deciding claims of unrepresented veterans.

[Brad]

I agree that it would have to be grounded in common sense. But there are common sense issues that lie within the scope of presumptions and outside the scope of probability theory. Evidentiary issues is a big one. It defies common sense (mine anyway) that we demand objective documented evidence of events that we know do not get documented: servicemembers simply do not document every blow to the head, for example, so why do we require this to establish mild traumatic brain injury?

As to practical problems, a theory of presumptions would not be applied directly by adjudicators. It would instead guide regulators to encourage them to create more appropriate presumptions (each of which would be relatively bright-line for adjudicators' purposes). This is what has happened with PTSD: several firm rules have been established the reflect the evidentiary realities of different circumstances. This should be the rule, not the exception.