Jeffrey Rowe v. Monica Gibson

HAMILTON, Circuit Judge,

concurring in part and dissenting in part.

I agree with the majority’s disposition of most claims and issues: affirming summary judgment for defendants on several claims and reversing on Rowe’s retaliation claim and his claim for complete denial of his Zantac medicine for 33 days in July and August 2011.

*636I must dissent, however, from the reversal of summary judgment on Rowe’s claim regarding the timing for administering his medicine between January and July 2011 and after August 2011. On that claim, the reversal is unprecedented, clearly based on “evidence” this appellate court has found by its own internet research. The majority has pieced together information found on several medical websites that seems to contradict the only expert evidence actually in the summary judgment record. With that information, the majority finds a genuine issue of material fact on whether the timing of Rowe’s Zantac doses amounted to deliberate indifference to a serious health need, and reverses summary judgment. (The majority denies at a couple of points that its internet research actually makes a difference to the outcome of the case, see ante at 629, 630, but when the opinion is read as a whole, the decisive role of the majority’s internet research is plain.)

The majority writes that adherence to rules of evidence and precedent makes a “heartless ... fetish of adversary procedure.” Yet the majority’s decision is an unprecedented departure from the proper role of an appellate court. It runs contrary to long-established law and raises a host of practical problems the majority fails to address.

To explain my disagreement, Part I reviews the facts in the record before us and shows that the majority has actually based its decision on its internet research. Part II explains why the majority’s reliance on its own factual research is contrary to law. Part III addresses the practical problems posed by the majority’s decision to do its own factual research. Finally, Part IV points out problems with the reliability of the majority’s factual research and shows that the enterprise of judicial factual research is unreliable when it loses the moorings to the law of judicial notice.

I. The Facts in the Record

On Rowe’s claim that the timing of his Zantac doses showed deliberate indifference to his health, the evidence in the record consists of two items. First, plaintiff Rowe asserts in his verified complaint and in several affidavits that he believes the prison’s schedule for giving him two 150 mg Zantac pills each day left him in unnecessary and avoidable pain for hours every day after meals. Second, defendants filed an affidavit from defendant Dr. William Wolfe, who was a career physician in the United States Air Force and is now a contract physician for the Indiana Department of Correction. Dr. Wolfe testified: “It does not matter what time of day Mr. Rowe receives his Zantac prescription. Each Zantac pill is fully effective for twelve hour increments. Zantac does not have to be taken before or with a meal to be effective. Providing Mr. Rowe with Zantac twice daily as the nursing staff makes their medication rounds, whatever time that may be, is sufficient and appropriate to treat his heart burn symptoms.”

The record thus shows a prisoner’s diagnosed disease and complaints of pain that prison staff treated with an appropriate medicine. The prisoner is not satisfied with details of the treatment’s timing, but a physician testified that the timing change the prisoner wanted was not called for because the medicine was equally effective as long as he was receiving two doses per day. This evidence does not support a reasonable inference of deliberate indifference.

Proof of deliberate indifference is much more demanding than proof of even medical malpractice. E.g. Petties v. Carter, 795 F.3d 688 (7th Cir.2015); Ray v. Wexford *637Health Sources, Inc., 706 F.3d 864, 866 (7th Cir.2013); Duckworth v. Ahmad, 532 F.3d 675, 679 (7th Cir.2008); see generally Estelle v. Gamble, 429 U.S. 97, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976). This record evidence would not let a reasonable jury find that the prison’s schedule for giving Rowe his medicine departed so far from professional standards to find that any prison staff acted with deliberate indifference to his health. The district court therefore properly granted summary judgment for defendants on this claim. See, e.g., Norfleet v. Webster, 439 F.3d 392, 396 (7th Cir.2006) (reversing denial of summary judgment), citing Estate of Cole v. Fromm, 94 F.3d 254, 262 (7th Cir.1996) (affirming summary judgment); see also, e.g., Pyles v. Fahim, 771 F.3d 403, 409 (7th Cir.2014) (affirming summary judgment; physician’s refusal to order MRI for prisoner’s back pain did not show deliberate indifference).

As noted above, the majority claims twice that its decision does not actually depend on its independent factual research, at pages 629 and 630. See also ante at 635 (Rovner, J., concurring). These denials contradict the rest of the majority opinion. If they were accurate, the majority’s long discussion of its research and its justifications for it would amoxmt to a long essay not necessary to the court’s decision. If the denials were accurate, moreover, the majority decision would amount to a significant rewriting of the Eighth Amendment law governing health care for prisoners.

Where prison medical staff just refuse to treat serious pain or disease, a prisoner may well have a viable claim that should go to trial. E.g., Miller v. Campanella, 794 F.3d 878 (7th Cir.2015) (no treatment of prisoner’s GERD); Hayes v. Snyder, 546 F.3d 516, 524-26 (7th Cir.2008). Where the evidence shows, however, that medical staff have provided at least some treatment for pain we almost always hold that the prisoner is not entitled to a jury trial on a claim for deliberate indifference based on a claim that the pain treatment was not adequate. E.g., Pyles v. Fahim, 771 F.3d 403, 409, 411 (7th Cir.2014); Holloway v. Delaware County Sheriff, 700 F.3d 1063, 1073-76 (7th Cir.2012).

If the majority decision did not depend on its own factual research, then the majority would be holding that the prisoner’s dissatisfaction with pain treatment is enough to require a jury trial on whether the prison’s medical staff were deliberately indifferent to his pain. We have not found before this case that such evidence is sufficient to infer deliberate indifference. But we will see a lot more cases like this one. As the average age of the prison population increases, so will the incidence of painful, chronic conditions that cannot be treated to the complete satisfaction of the prisoners. The fact that a treatment for pain is not as effective as the prisoner would like should not be enough to support an inference that the prison staff are deliberately indifferent to his pain.

In fact, the majority’s reversal on this claim is based on a small but important category of cases in which prisoners have shown that medical staff persisted in obviously inadequate courses of treatment. In those cases, we have found triable issues of deliberate indifference. E.g., Arnett v. Webster, 658 F.3d 742, 754 (7th Cir.2011); Berry v. Peterman, 604 F.3d 435, 441-42 (7th Cir.2010); Greeno v. Daley, 414 F.3d 645, 654 (7th Cir.2005) (treatment prisoner received was “blatantly inappropriate”). As we explained in Pyles, these decisions were based on evidence showing that the need for specialized expertise or different treatment was either known by the treating physicians or would have been obvious to a lay person. 771 F.3d at 411.

*638The problem for the majority here is that Rowe himself has made no comparable showing. Only by relying on its independent factual research can the majority establish an arguable basis for applying this theory that the course of treatment was so clearly inadequate as to amount to deliberate indifference. The majority decision to reverse summary judgment on this claim thus depends on that independent factual research.

II. The Law on Judicial Research into the Facts

The ease of research on the internet has given new life to an old debate about the propriety of and limits to independent factual research by appellate courts.1 To be clear, I do not oppose using careful research to provide context and background information to make court decisions more understandable. By any measure, however, using independent factual research to find a genuine issue of material, adjudicative fact, and thus to decide an appeal, falls outside permissible boundaries. Appellate courts simply do not have a warrant to decide cases based on their own research on adjudicative facts. This case will become Exhibit A in the debate. It provides, despite the majority’s disclaimers, a nearly pristine example of an appellate court basing a decision on its own factual research.

The majority’s factual research runs contrary to several lines of well-established case law holding that a decision-maker errs by basing a decision on facts outside the record.

If a district judge bases a decision on such research, we reverse for a violation of Rule 201. E.g., Pickett v. Sheridan Health Care Center, 664 F.3d 632, 648-51 (7th Cir.2011) (district court erred by relying on independent internet research on attorney fees without giving parties opportunity to address information).

If jurors start doing their own research during a trial, a new trial is likely. United States v. Thomas, 463 F.2d 1061, 1062-65 (7th Cir.1972); see also United States v. Blagojevich, 612 F.3d 558, 564 (7th Cir.2010) (noting concern that messages to jurors would tempt them to engage in “forbidden research and discussion”).

If an immigration judge or administrative law judge bases a decision on facts without record support, we reverse it. See, e.g., Huang v. Gonzales, 403 F.3d 945, 948-50 (7th Cir.2005) (reversing immigration decision based on alien’s answers to questions based on judge’s personal beliefs about alien’s religion); Nelson v. Apfel, 131 F.3d 1228, 1236-37 (7th Cir.1997) (ALJ’s reliance on evidence outside record was erroneous but harmless).

We are in no better a position to go outside the record for decisive facts. Our job is to reverse in cases where the decision-maker has gone outside the record. The majority in this case, however, not only does what we treat as reversible error when others do it; it holds in essence that *639the district judge erred by not doing such independent factual research. What was forbidden is now required.

In addition to the case law holding that a decision-maker is not permitted to base a decision on evidence outside the record, another body of law is relevant to this issue: Federal Rule of Evidence 201 and the law of judicial notice. The majority opinion runs contrary to that law and misunderstands how Rule 201 and judicial notice fit together with the ordinary, adversarial presentation of facts.

The vast majority of facts that courts consider when deciding cases comes from the familiar, adversarial presentations of evidence by opposing parties. The foundation of our legal system is a confidence that the adversarial procedures will test shaky or questionable evidence: “Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence.” Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 596, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Those protective procedures are not available when a court decides to do its own factual research and bases its decision on what it finds.

The law of evidence allows a narrow exception permitting some judicial research into relevant facts, under Federal Rule of Evidence 201 and the concept of judicial notice. Judicial notice “substitutes the acceptance of a universal truth for the conventional method of introducing evidence,” and as a result, courts must use caution and “strictly adhere” to the rule before taking judicial notice of pertinent facts. General Elec. Capital Corp. v. Lease Resolution Corp., 128 F.3d 1074, 1081 (7th Cir.1997); see also Hennessy v. Penril Datacomm Networks, Inc., 69 F.3d 1344, 1354 (7th Cir.1995) (“In order for a fact to be judicially noticed, indisputability is a prerequisite.”).

The majority says twice it is not taking judicial notice of all the cited medical information from the internet. Ante at 628-29, 631-32. I agree it could not properly take judicial notice of this information under Evidence Rule 201(b) and (e). The proper timing of a patient’s doses of Zantac is not “generally known within the trial court’s territorial jurisdiction” and is not beyond “reasonable dispute,” nor can it be “accurately and readily determined from sources whose accuracy cannot reasonably be questioned,” as Rule 201(b) requires.And the majority has made no effort to comply with the procedural requirements of Rule 201(e), essential to basic fairness, of giving the parties an opportunity to be heard on the evidence.

If the majority is not taking judicial notice, what exactly is it doing? It seems to have created an entirely new, third category of evidence, neither presented by the parties nor properly subject to judicial notice. The majority writes:

When medical information can be gleaned from the websites of highly reputable medical centers, it is not imperative that it instead be presented by a testifying witness. Such information tends to fall somewhere between facts that require adversary procedure to determine and facts of which a court can take judicial notice, but it is closer to the second in a case like this in which the evidence presented by the defendants in the district court was sparse and the appellate court need only determine whether there is a factual dispute sufficient to preclude summary judgment.

Ante at 628-29 (emphasis added). In other words, the majority acknowledges that its “evidence” neither comes from adversarial presentation by the parties nor *640meets the strict substantive and procedural standards for judicial notice under Rule 201.

Before this decision, American law has not recognized this category of evidence, which might be described as “non-adversarial evidence that the court believes is probably correct.” Compare the comments of the authors of Rule 201, the Advisory Committee Notes from 1972:

The usual method of establishing adjudicative facts is through the introduction of evidence, ordinarily consisting of the testimony of witnesses. If particular facts are outside the area of reasonable controversy, this process is dispensed with as unnecessary. A high degree of indisputability is the essential prerequisite.

In other words, the Federal Rules of Evidence allow no room for the majority’s innovation. Adversarial evidence and judicial notice are not opposite poles on a wide spectrum, with a middle ground for the majority’s evidence that has neither been subjected to adversarial testing nor a proper subject of judicial notice. These are two distinct categories. To be admissible, evidence must fall within one or the other. “Close” to judicial notice does not count.

The majority has not offered any precedent from the law of evidence to support its reliance on its own factual research. Instead, it tries to downplay the unprecedented step it takes, including its emphasis that it is “not ordering that judgment be entered in Rowe’s favor” and that defendants will be entitled to rebut the majority’s factual research on remand. Ante at 631-32. The majority’s modest demurrer loses sight of the stakes. The issue on summary judgment is whether the evidence in the record would allow a reasonable jury to find in favor of the non-moving party. See Reeves v. Sanderson Plumbing Products, Inc., 530 U.S. 133, 149-50, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 251-52, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). By reversing, the majority is necessarily finding that this record is sufficient to support a jury verdict for Rowe. I disagree.

The majority also points out that “judges and their law clerks often conduct research on cases without disclosure to the parties.” Ante at 628. Such research has long been understood to involve only legal research. The majority’s effort to compare long-accepted judicial research into case law and statutes to its independent factual research shows the majority has entered unknown territory.

To justify this venture, the majority asks a number of rhetorical questions and. invokes the courage of the barons at Runnymede in 1215. Ante at 629. With respect, we are an intermediate appellate court. The Federal Rules of Evidence and Federal Rules of Civil Procedure that we apply are adopted and amended through processes established by the Rules Enabling Act, 28 U.S.C. § 2071 et seq. We simply do not have authority on our own to take the law into this unknown territory.

III. The Practical Problems

The majority points out correctly that prisoners must depend entirely on the government for their health care. If they turn to the federal courts for help, the combination of the constitutional standard under the Eighth Amendment, deliberate indifference to a serious health need, and the system of personal liability under 42 U.S.C. § 1983 can make it very difficult for a prisoner to hold anyone accountable for serious wrongs. See, e.g., Shields v. Illinois Dep’t of Corrections, 746 F.3d 782 (7th Cir.2014). When a prisoner brings a pro se suit about medical care, the adversary process that is the foundation of our *641judicial system is at its least reliable. Few prisoners have access to lawyers or to expert witnesses needed to address medical issues.

These conditions pose important challenges to federal courts doing their best to decide these cases fairly. Yet the majority’s solution — to research available medical information on its own and find a genuine issue of material fact on that basis — raises problems much more serious than a possible error in the resolution of one prisoner’s case.

The majority’s approach turns the court from a neutral decision-maker into an advocate for one side. The majority also offers no meaningful guidance as to how it expects other judges to carry out such factual research and what standards should apply when they do so. Under the majority’s approach, the factual record will never be truly closed. This invites endless expansion of the record and repetition in litigation as parties contend and decide that more and more information should have been considered.

In addition to the abandonment of neutrality, consider the problems from the district judge’s point of view. The majority clearly implies, while denying it is doing so, that the district judge herself should have done the independent factual research the majority has done on appeal, questioning an unchallenged expert affidavit by looking to websites of the drug manufacturer, the Mayo Clinic, the Physician’s Desk Reference, and Healthline.

The practical questions are obvious: When are district judges supposed to carry out this independent factual research? How much is enough? What standards of reliability should apply to the results? How does the majority’s new category of evidence fit in with a district judge’s gate-keeping responsibilities under Rule 702 and Daubert? The majority offers no answers.

Thé majority essentially orders the district judge on remand to find an expert witness on the medical issues, either for plaintiff or as a neutral expert under Rule 706. That might well be helpful, but as the majority concedes, district courts do not have budgets for that purpose. Even if a few experts might be willing to volunteer in unusual cases, the demand of prisoners for free medical or other expert witnesses will far exceed the supply, especially in the rural areas where so many prisons are located and smaller towns where the nearest district courts are located.

The majority’s solution for this problem is to have the district court use Federal Rule of Evidence 706 to order defendants, and only the defendants, to pay for an expert witness for the plaintiff or the court. See ante at 631-32. That approach is not foreclosed by the language of Rule 706, and there is some case law supporting it. See Ledford v. Sullivan, 105 F.3d 354, 360-61 (7th Cir.1997). Nevertheless, the majority’s reliance on this solution in this ordinary ease further threatens the neutrality of the courts. It is worth recalling that damages under 42 U.S.C. § 1983 must be sought from state employees only in their individual capacities. Will v. Michigan Dep’t of State Police, 491 U.S. 58, 109 S.Ct. 2304, 105 L.Ed.2d 45 (1989). Indemnification by their employer is a matter of state law and policy, and sometimes grace. See Ind.Code § 34-13-4-1; Estate of Moreland v. Dieter, 576 F.3d 691, 694-96 (7th Cir.2009). Is it fair to impose on individual guards, prison administrators, staff, nurses, and doctors the cost of finding evidence to build a case against them? At the very least, such one-sided burdens *642should be imposed only in extraordinary cases.2

Further, if the case goes to trial, how is the district judge supposed to present to a jury the information the majority has found? My colleagues and I agree it is not suitable for judicial notice because it is not indisputable, as required under Rule 201(b). Without an expert witness qualified to present the facts and opinions the majority finds persuasive, that information does not come into evidence. On appeal would the majority’s approach lead us to remand for a new trial with instructions to look harder for the right evidence? Or what should we do if the district judge did not find or rely on the information that our research turns up? As long as the factual record remains open for judicial supplements, parties will try to use the quest for the perfect record to keep any .loss in litigation from being final.

Then consider the problems parties and their lawyers will face. If we permit such independent factual research by district judges — even expect such research from them — parties will need to plan for it. Responding to the evidence actually offered by the other side is often the biggest challenge and expense in a lawsuit. Now parties need to anticipate the evidence the judge might turn up on her own and prepare to meet it. The time and expense devoted to such preventive measures will be substantial and should be unnecessary. And if the district judge does her own research and gives the parties an opportunity to respond to it, the majority’s approach here is an open invitation for parties to add to the record on appeal. The parties will also need to anticipate on appeal that our court will undertake its own factual research, opening up opportunities to save any losing case by offering new evidence on appeal.3

From the larger perspective of our judicial system, the independent factual research the majority endorses and even requires here is not something that federal courts can carry out reliably on a large scale. History is probably the academic field closest to the practice of law and judging. Yet historians regularly scoff at the phenomenon called “law-office history.” See Velasquez v. Frapwell, 160 F.3d 389, 393 (7th Cir.1998) (Posner, J.) (“[J]udges do not have either the leisure or the training to conduct responsible historical research or competently umpire historical controversies. The term ‘law-office history’ is properly derisory and the derision embraces the efforts of judges and law professors, as well as of legal advocates, to play historian. * * * Judges don’t try to decide contested issues of science without the aid of expert testimony, and we fool ourselves if we think we can unaid*643ed resolve issues of historical truth.”), vacated in part, 165 F.3d 593 (7th Cir.1999).

Law-office or judicial-chambers medicine is surely an even less reliable venture. The internet is an extraordinary resource, but it cannot turn judges into competent substitutes for experts or scholars such as historians, engineers, chemists, psychologists, or physicians. The majority’s instruction to the contrary will cause problems in our judicial system more serious than those it is trying to solve in this case.

IV. How Reliable Is Our Research?

Thus far I have avoided debating the details of the majority’s research, but they deserve closer attention. The specific details highlight the more general criticisms I have directed at such factual research by judges.

First, on the websites the majority relies upon, we find important disclaimers that emphasize the need for filtering their information through qualified medical advice, which no member of this court is qualified to provide. The Physician’s Desk Reference site says it is to be used “only as a reference aid. It is not intended to be a substitute for the exercise of professional judgment. You should confirm the information on the PDR.net site through independent sources and seek other professional guidance in all treatment and diagnosis decisions.” www.pdr.net (last visited August 19, 2015, as were all websites cited here). The Mayo Clinic and Zantac websites have similar disclaimers advising readers to talk to a physician or other health care provider before acting on the information on the websites. See www. mayoclinic.org/about-this-site/termsconditions-use-policy; www.zantacotc.com/ zantac-maximumstrength.html# faqs.

Second, after we get past the disclaimers, the content of the majority’s websites simply does not give clear support to the majority’s views (a) that Dr. Wolfe was wrong in saying that the 150 mg pills Rowe was receiving twice a day could be equally effective even if not given shortly before meals, let alone (b) that Dr. Wolfe was so thoroughly and obviously wrong that a jury could infer that prison staff were deliberately indifferent to Rowe’s health needs. The majority’s websites instead show that some degree of medical judgment is needed to decide when best to administer which size pills for patients with different needs, especially patients like Rowe with chronic conditions.

The Mayo Clinic site says that patients taking prescription strength Zantac twice a day should take one in the morning and one at bedtime. The majority discounts that advice because Rowe was taking an over-the-counter dosage of 150 mg pills rather than the prescription dosage of 300 mg pills. Ante at 626. Yet that explanation overlooks the advice from both the manufacturer and the Mayo Clinic that a patient should not take the over-the-counter pills for more than two weeks unless directed by a doctor. For patients like Rowe, taking Zantac long-term to treat GERD, the Mayo Clinic offers more specific guidance. It advises that adult patients with GERD take the 150 mg pill two times a day without specifying that the pills should be taken shortly before meals. www.mayoclinic.org/drugssupplements/ histamine-h2-antagonist-oral-routeinjection-route-intravenousroute/properuse/drg-20068584. That advice from the Mayo Clinic seems identical to Dr. Wolfe’s view.

Similarly, the PDR advises that for treatment of GERD, “Symptomatic relief commonly occurs within 24 hours after starting therapy with ZANTAC 150 mg twice daily,” again without indicating any need to take the pills before meals, www. pdr.net/full-prescribing-information/ *644zantac-150-and-300-tablets?druglabelid= 241# section-standard-1.

The “Ml prescribing information” on the Physician’s Desk Reference website says that for treatment of GERD with the 150 mg and 300 mg pills, “Symptomatic relief commonly occurs within 24 hours after starting therapy with ZANTAC 150 mg twice daily,” again without saying anything about taking pills before meals, www.pdr. net/Ml-prescribinginformation/zantac150-and-300-tablets?druglabelid=24Í. And again, that was Rowe’s diagnosis and those were his pills in 2011.

The majority draws on the PDR website and “common sense” regarding how long the pills remain effective. Ante at 630-31. The PDR website, however, simply does not provide sufficient data on absorption and clearance rates for the medicine to allow us to exercise our own (non-expert) judgment about whether the timing of Rowe’s pills was appropriate. It certainly does not allow us to conclude that the timing could have amounted to deliberate indifference ta his serious health needs or to find that Dr. Wolfe’s uncontradicted affidavit did not support the district court’s entry of summary judgment on this claim.

Of course, the point of this discussion of the websites is not to debate the majority on the medical fine points. The websites the majority relies upon tell us themselves that their information needs to be interpreted by a qualified physician. None of this information is in the record. None was before the district court, nor is it properly before us.

The majority’s interpretation of its internet research is not a reliable substitute for proper evidence subjected to adversarial scrutiny. And while Dr. Wolfe’s affidavit is far less detailed than the information the majority has explored on the internet, I also see no basis for the majority’s harsh criticism of him, especially when Dr. Wolfe has not been given any opportunity to respond or explain.4

In the end, whether Dr. Wolfe’s testimony about the timing for Rowe’s doses was right or wrong in some pure and objective sense, or in a case tried with ample resources and talent on both sides, is not the question for us. For purposes of summary judgment, Dr. Wolfe’s testimony was undisputed. We have no business reversing summary judgment based on our own, untested factual research. By doing so, the majority has gone well beyond the appropriate role of an appellate court. I respectfully dissent from the reversal of summary judgment on Rowe’s claims based on the timing of his medication.

. See, e.g., Layne S. Keele, When the Mountain Goes to Mohammed: The Internet and Judicial Decision-Making, 45 N.M. L.Rev. 125 (2014); Allison Orr Larsen, The Trouble with Amicus Facts, 100 Va. L.Rev. 1757 (2014); Richard A.' Posner, Judicial Opinions and Appellate Advocacy in Federal Courts — One Judge's Views, 51 Duq. L.Rev. 3 (2013); Frederick Schauer, The Decline of "The Record”: A Comment on Posner, 51 Duq. L.Rev. 51 (2013); Elizabeth G. Thornburg, The Lure of the Internet and the Limits on Judicial Fact Research, Litig., Summer 2012, at 41; Brianne J. Gorod, The Adversarial Myth: Appellate Court Extra-Record Factfinding, 61 Duke L.J. 1 (2011); Elizabeth G. Thornburg, The Curious Appellate Judge: Ethical Limits on Independent Research, 28 Rev. Litig. 131 (2008); Coleen M. Barger, On the Internet, Nobody Knows You're a Judge: Appellate Courts’ Use of Internet Materials, 4 J.App. Prac. & Process 417 (2002).

. I share the concerns expressed by the district court in Martin v. Cohn, 1999 WL 325054, at *1 (N.D.Ind. April 5, 1999), about the fundamental fairness of imposing this financial burden on one side solely because the opposing party is indigent. The defendants will end up having to foot the bill for the expert even if they win the case. One partial but creative solution to this problem can be found in Goodvine v. Ankarlo, 2013 WL 1192397, at *2 (W.D.Wis. March 22, 2013) (providing for long-term assessments of plaintiff's prison trust account to pay for court-appointed expert if plaintiff did not prevail).

. If parties on appeal try to supplement the record as the majority does here, they are rebuked and may even be sanctioned. E.g., Hart v. Sheahan, 396 F.3d 887, 894-95 (7th Cir.2005) (stating general rule but finding no violation because appeal was from dismissal on pleadings); Holmberg v. Baxter Healthcare Corp., 901 F.2d 1387, 1392 n. 4 (7th Cir.1990) (striking portions of appellee's brief). Under the majority’s approach, we could not take such steps in response to parties’ invitations to our court to repeat what the majority does here.

. The majority criticizes Dr. Wolfe’s affidavit for not providing an explanation for his opinion about the timing of the Zantac doses. The majority overlooks Federal Rule of Evidence 705, which pérmits conclusory expert testimony unless and until the conclusions are challenged, which Dr. Wolfe's affidavit was not in the district court. He has not yet been called upon to explain his opinion in this case. The fact that he is a defendant does not disqualify him from offering an affidavit; we often affirm summary judgment based on a moving party's testimony. The majority points out that Dr. Wolfe is "a frequent defendant in prisoner civil rights suits,” ante at 625, as if that reflected poorly on his professionalism. Virtually any physician serving large numbers of prisoners will be "a frequent defendant in prisoner civil rights suits.”