A close reading of the medical record

Both Andrew Sullivan and Spencer Ackerman raise alarms about a new report in the online journal PLoS Medicine.  The report concerns the medical providers at the detention facility at Guantanamo Bay and their complicity in the torture of detainees there, and was written by two physicians who were given access to the detainees’ records by their legal team.

Before I go any further, it’s important that I make a few things clear.  I have absolutely no doubt that detainees at Gitmo were routinely subjected to inhumane treatment that meets the definition of torture.  As far as I am concerned, there is no serious argument on this point any longer.  I am inclined to believe that the subjects of the report (the medical providers) probably did collude with interrogators in eliding details from the medical record that would have provided evidence of brutality and torture.  I strongly suspect the report is true.

It is, unfortunately, also a deeply flawed document.  Whatever its strengths from a political perspective, as a medical report it is easy to pick apart and thus weaken its impact.  If I were not already inclined to believe its conclusions, there would be much I could use to persuade myself that it is not credible.

The first and most glaring problem is that of potential observer bias.  This report is rife with reasons to suspect it.  The authors are both members of Physicians for Human Rights, an organization whose work I applaud but one that is obviously ideological.  A reader cannot plausibly expect them to be neutral on the subject matter at issue.  Furthermore, there is this:

A total of five non-governmental medical experts, including the authors, consulted on behalf of the nine detainees, only one non-governmental medical expert was partially financially compensated for his services.

It is not made clear which investigator this is, but one of the authors served as a medical expert on behalf of the detainees. A reader might reasonably conclude that one of the authors was remunerated for the work that led to this report, even if the payment did not fund it directly.  Just as one might question Andrew Wakefield’s objectivity when he wrote his MMR study, so one can similarly question the authors’ objectivity in evaluating the evidence before them now.  At least one of them cannot claim to be disinterested.

From the Editors’ Summary that opens the report:

The findings are also limited by their reliance on medical records and case files that were sometimes heavily edited and on psychological assessments based on questionnaires rather than on direct examination.

To my reading, that is a significant limitation.  A reader might reasonably wonder if the heavily edited portions of the record contain exculpatory information with regard to the medical providers’ behavior and documentation.  If they did document the cause of various injuries, it is plausible to suspect that that information was edited out before the record was made available for review.  That may raise suspicions about torture having occurred, but the report concerns itself with the medical providers specifically, and the data contain gaps that may weaken the case against them.

As to those injuries:

In three of the nine cases, the GTMO medical records documented injuries that were consistent or highly consistent with detainee allegations of abuse: contusions (2), bone fractures (3), lacerations (2), peripheral nerve damage (1), and sciatica (2). There was no mention of any cause for these injuries.

Documentation of injuries without documenting a history to account for them is genuinely problematic. It is difficult to explain away this finding in the report, and it is because of this that I suspect there really was actual collusion on the part of the medical providers. Some of the other claims are a bit shakier.

Another detainee indicated that he observed interrogators with his medical records and that his chronic back pain was exploited by interrogators with the use of prolonged, painful stress positions.

I will leave it to those with legal training to say whether this meets the definition of hearsay, though that’s what it looks like to me. It is certainly not an objective medical finding.

I have yet more trouble with the authors’ findings with regard to psychological problems.

According to the GTMO medical records, DoD mental health providers with the Behavioral Health Service (BHS) evaluated six of the nine detainees and diagnosed the following: depression (4), passive aggressive personality (4), borderline personality (2), adjustment disorder (3), routine stressors of confinement (2), narcissistic traits (1), psychosis or depression with psychotic features (2), and anxiety NOS (not otherwise specified) (2).


Nearly all BHS visits were initiated after suicide attempts and/or detainee hunger strikes. Under these circumstances, BHS notes indicate that the visits were unwelcome and the detainees often refused to cooperate.

That’s a pretty impressive list of diagnoses. Furthermore, making a reliable diagnosis for an uncooperative patient in an area with so much room for subjectivity as behavioral health is very difficult. I do not believe the authors acknowledge this difficulty sufficiently.

From the authors’ medicolegal assessment:

The assessments conducted by non-governmental medical experts in each of the nine cases indicate that the specific allegations of torture and ill treatment were highly consistent with and supported by physical and psychological evidence observed in all cases. The psychological component of the evaluation revealed diagnostic criteria for current major depression and/or PTSD in all nine cases.

There are a couple of problems with this.  First of all, as I’ve just quoted, the BHS providers did diagnose depression, at least in some cases.  And that “revealed diagnostic criteria” language is slippery.  There is a big difference between displaying some diagnostic criteria for a disorder and meeting enough of them to make the diagnosis.  The hallucinations reported by some patients are indeed troubling, but there is enough overlap between the diagnostic criteria for PTSD, generalized anxiety disorder and major depression with psychotic features that differentiating between them (particularly with an uncooperative patient) can be difficult, and a failure to diagnose PTSD does not read to me as an overt attempt to hide evidence of torture.

The report spends a lot of time on the question of whether the detainees were tortured, and offers evidence in support of the now widely-accepted view that they were.  But as to the central question of the study, which is to what degree medical and mental health providers colluded with interrogators to torture detainees or cover up evidence that they had been tortured, there are sufficient gaps to create some doubt.  Documentation of the source of injury could be included in the redacted portion of the record.  Differentiating between the effects of detention and PTSD in an uncooperative patient could simply have been difficult, and not due to willful neglect.

Second and third readings of the report leave me with the same impression.  It is suspicious that the source of injuries such as lacerations are not documented.  PTSD does make sense as a diagnosis, and it is curious that it was not listed among those given.  But the authors do not sufficiently acknowledge that the former may be due to the redaction of the record, and the latter problem could be due (at least in part) to the difficulties of working with a uniformly uncooperative patient population.  I am inclined to believe the report, because I believe Guantanamo has proven to be one of the most unsightly blots on my country’s history in my lifetime, but if I were not so inclined there is much in this report that would allow me to discount it.

Russell Saunders

Russell Saunders is the ridiculously flimsy pseudonym of a pediatrician in New England. He has a husband, three sons, daughter, cat and dog, though not in that order. He enjoys reading, running and cooking. He can be contacted at blindeddoc using his Gmail account. Twitter types can follow him @russellsaunder1.


  1. Calling a victim of torture’s own report of his experience hearsay is shameful.

    • Really? How so?

      If you’ll read the post, you’ll note that I pretty much accept that the detainees in question were tortured. (If I didn’t, I might note that you are begging the question.) That is not the point of my post. My point is that, whatever the merits of the case against interrogators at Guantanamo, the evidence presented in this report is shaky in a variety of ways. The detainee’s description of what happened likely bears further investigation, but it does not stand on its own as objective medical fact or proof of wrongdoing on the part of medical staff.

Comments are closed.