EASTERBROOK, Circuit Judge.
Eugene Villain died in 1997 of colon cancer at the age of 66. His other afflictions included coal workers’ pneumoconio-sis — -or so the Department of Labor concluded during Eugene’s life, for it had ordered Zeigler Coal Co. to pay him disability benefits under the Black Lung Benefits Act. After Eugene’s death, his widow Phyllis applied for survivor’s benefits. To obtain them, she had to show that pneumo-coniosis caused Eugene’s death. See 30 U.S.C. § 901(a); 20 C.F.R. § 718.205(c). She proposed to establish that Eugene had pneumoconiosis by relying on the adjudication of his disability claim, and causation by proving that breathing problems lessened Eugene’s ability to resist the effects of the cancer, leading him to succumb more quickly. An administrative law judge ruled in Phyllis’s favor on both points, the Benefits Review Board affirmed, and Zeigler now asks us to set aside the administrative decision.
Peabody Coal Co. v. Spese, 117 F.3d 1001 (7th Cir.1997) (en banc), holds that normal rules of preclusion govern administrative proceedings in black lung cases. When deciding that Eugene was disabled by pneumoconiosis, the agency necessarily concluded that he had that dis[334]*334ease — and as this is one element of the widow’s claim too, it makes sense to treat it as established. Although the widow was not a party to the miner’s claim, Zeigler itself was. Treating Zeigler as bound by the outcome is a straightforward application of offensive nonmutual issue preclusion. See Parklane Hosiery Co. v. Shore, 439 U.S. 322, 99 S.Ct. 645, 58 L.Ed.2d 552 (1979).
What Zeigler offers in response is the proposition that timing matters. Eugene was entitled to benefits if he was disabled by pneumoconiosis in 1989, when an ALJ ruled favorably on his claim. Phyllis is entitled to benefits only if Eugene had pneumoconiosis in 1997, when he died. True enough, but is there any medical reason to suppose that people recover from pneumoconiosis? Not all kinds of black lung are progressive; the milder forms of the condition do not get worse over time unless the miner inhales more dust. Yet unless pneumoconiosis sometimes goes into remission, there is no reason to hold a new hearing on the question whether a person who had that condition during life also had it at death. Zeigler does not offer us (and did not introduce before the agency) any medical evidence suggesting that black lung can be cured.
Another potential reason for thinking the passage of time significant is that diagnosis is error-prone. Other pulmonary diseases such as emphysema and asthma may produce symptoms hard to tell apart from pneumoconiosis. Radiologists frequently disagree about the interpretation of x-ray films; only for the most serious forms of the disease are the opacities indicative of pneumoconiosis easy to distinguish from opacities with other causes. Death offers a considerably better source of evidence: analysis of the lung tissue removed in an autopsy. The Benefits Review Board therefore has created an autopsy exception to the rule of issue preclusion. Both a mine operator and a survivor are allowed to introduce autopsy evidence in an effort to show that the determination made during the miner’s life was incorrect. (A mine operator seeks to refute an earlier finding of pneumoconiosis, while a widow will try to show that rejection of the miner’s claim was a false negative.) Again this makes sense. But no autopsy was performed on Eugene. He left behind only the sorts of x-ray, ventilatory-capacity, and blood-gas readings — indicators with high rates of false positives and false negatives — that were available during his life. We agree with the BRB that there is no point in readjudicating the question whether a given miner had pneumoconiosis unless it is possible to adduce highly reliable evidence — which as a practical matter means autopsy results. Otherwise the possibility that the initial decision was incorrect is no reason to disturb it. See Federated Department Stores, Inc. v. Moitie, 452 U.S. 394, 101 S.Ct. 2424, 69 L.Ed.2d 103 (1981). We therefore hold that a grant of survivor’s benefits may rest on findings made during the miner’s life.
Next comes the question whether Eugene’s pneumoconiosis caused his demise. No one thinks that coal dust causes colon cancer. But benefits must be paid if the disease hastened death from another source. See Peabody Coal Co. v. Director, OWCP, 972 F.2d 178, 183 (7th Cir.1992). See also 20 C.F.R. § 718.205(c)(5), approved by National Mining Ass’n v. Department of Labor, 292 F.3d 849, 871 (D.C.Cir.2002). Even the murder of an infant just hastens death, given that death comes eventually to us all. The difference here was four months. Eugene’s oncologist estimated that he had six months to live from the time his cancer was diagnosed; he died in two, and a general practitioner opined that this acceleration was [335]*335attributable to Eugene’s weakened state, which the physician blamed on the pneu-moconiosis. One responsive line of argument might have been that survivors’ benefits are designed to replace lost income, which implies drawing a line between advancing death by a few years and advancing it by a few months, weeks, or days. Phyllis lost no more income as a result of Eugene’s actual death than she would have lost had he lived as long as the oncologist predicted. Requiring Zeigler to provide Phyllis with lifetime benefits because of small changes in the date of her husband’s demise from cancer is jarring. Yet Zeig-ler does not make such an argument, so we do not decide whether there is a legal difference between taking years and days off a miner’s life span.
Instead Zeigler grumbles that a general practitioner should not opine on matters beyond his expertise. Frederick Ridge, who had been Eugene’s physician, did not treat either cancer or pulmonary disease and referred Eugene to specialists for both. Why, then, should the agency accept Ridge’s view about the cause of Eugene’s death, Zeigler wonders? It relies on Peabody Coal Co. v. McCandless, 255 F.3d 465 (7th Cir.2001), for the proposition that medical diagnoses must be supported by medical reasoning; a history of treatment is no substitute for scientific analysis. True enough, but as the Secretary observed when promulgating § 718.205(c)(5), the proposition that persons weakened by pneumoconiosis may expire quicker from other diseases is a medical point, with some empirical support. See 65 Fed.Reg. 79,920, 79,950 (Dec. 20, 2000). Whether this was true of Eugene is a question that depends on the state of his physical and mental health, a subject about which Ridge professed to be knowledgeable.
One way to respond would be to show that Eugene’s death occurred at the expected time after all. The oncologist’s estimate of six months’ life expectation was simplified.
Free access — add to your briefcase to read the full text and ask questions with AI
EASTERBROOK, Circuit Judge.
Eugene Villain died in 1997 of colon cancer at the age of 66. His other afflictions included coal workers’ pneumoconio-sis — -or so the Department of Labor concluded during Eugene’s life, for it had ordered Zeigler Coal Co. to pay him disability benefits under the Black Lung Benefits Act. After Eugene’s death, his widow Phyllis applied for survivor’s benefits. To obtain them, she had to show that pneumo-coniosis caused Eugene’s death. See 30 U.S.C. § 901(a); 20 C.F.R. § 718.205(c). She proposed to establish that Eugene had pneumoconiosis by relying on the adjudication of his disability claim, and causation by proving that breathing problems lessened Eugene’s ability to resist the effects of the cancer, leading him to succumb more quickly. An administrative law judge ruled in Phyllis’s favor on both points, the Benefits Review Board affirmed, and Zeigler now asks us to set aside the administrative decision.
Peabody Coal Co. v. Spese, 117 F.3d 1001 (7th Cir.1997) (en banc), holds that normal rules of preclusion govern administrative proceedings in black lung cases. When deciding that Eugene was disabled by pneumoconiosis, the agency necessarily concluded that he had that dis[334]*334ease — and as this is one element of the widow’s claim too, it makes sense to treat it as established. Although the widow was not a party to the miner’s claim, Zeigler itself was. Treating Zeigler as bound by the outcome is a straightforward application of offensive nonmutual issue preclusion. See Parklane Hosiery Co. v. Shore, 439 U.S. 322, 99 S.Ct. 645, 58 L.Ed.2d 552 (1979).
What Zeigler offers in response is the proposition that timing matters. Eugene was entitled to benefits if he was disabled by pneumoconiosis in 1989, when an ALJ ruled favorably on his claim. Phyllis is entitled to benefits only if Eugene had pneumoconiosis in 1997, when he died. True enough, but is there any medical reason to suppose that people recover from pneumoconiosis? Not all kinds of black lung are progressive; the milder forms of the condition do not get worse over time unless the miner inhales more dust. Yet unless pneumoconiosis sometimes goes into remission, there is no reason to hold a new hearing on the question whether a person who had that condition during life also had it at death. Zeigler does not offer us (and did not introduce before the agency) any medical evidence suggesting that black lung can be cured.
Another potential reason for thinking the passage of time significant is that diagnosis is error-prone. Other pulmonary diseases such as emphysema and asthma may produce symptoms hard to tell apart from pneumoconiosis. Radiologists frequently disagree about the interpretation of x-ray films; only for the most serious forms of the disease are the opacities indicative of pneumoconiosis easy to distinguish from opacities with other causes. Death offers a considerably better source of evidence: analysis of the lung tissue removed in an autopsy. The Benefits Review Board therefore has created an autopsy exception to the rule of issue preclusion. Both a mine operator and a survivor are allowed to introduce autopsy evidence in an effort to show that the determination made during the miner’s life was incorrect. (A mine operator seeks to refute an earlier finding of pneumoconiosis, while a widow will try to show that rejection of the miner’s claim was a false negative.) Again this makes sense. But no autopsy was performed on Eugene. He left behind only the sorts of x-ray, ventilatory-capacity, and blood-gas readings — indicators with high rates of false positives and false negatives — that were available during his life. We agree with the BRB that there is no point in readjudicating the question whether a given miner had pneumoconiosis unless it is possible to adduce highly reliable evidence — which as a practical matter means autopsy results. Otherwise the possibility that the initial decision was incorrect is no reason to disturb it. See Federated Department Stores, Inc. v. Moitie, 452 U.S. 394, 101 S.Ct. 2424, 69 L.Ed.2d 103 (1981). We therefore hold that a grant of survivor’s benefits may rest on findings made during the miner’s life.
Next comes the question whether Eugene’s pneumoconiosis caused his demise. No one thinks that coal dust causes colon cancer. But benefits must be paid if the disease hastened death from another source. See Peabody Coal Co. v. Director, OWCP, 972 F.2d 178, 183 (7th Cir.1992). See also 20 C.F.R. § 718.205(c)(5), approved by National Mining Ass’n v. Department of Labor, 292 F.3d 849, 871 (D.C.Cir.2002). Even the murder of an infant just hastens death, given that death comes eventually to us all. The difference here was four months. Eugene’s oncologist estimated that he had six months to live from the time his cancer was diagnosed; he died in two, and a general practitioner opined that this acceleration was [335]*335attributable to Eugene’s weakened state, which the physician blamed on the pneu-moconiosis. One responsive line of argument might have been that survivors’ benefits are designed to replace lost income, which implies drawing a line between advancing death by a few years and advancing it by a few months, weeks, or days. Phyllis lost no more income as a result of Eugene’s actual death than she would have lost had he lived as long as the oncologist predicted. Requiring Zeigler to provide Phyllis with lifetime benefits because of small changes in the date of her husband’s demise from cancer is jarring. Yet Zeig-ler does not make such an argument, so we do not decide whether there is a legal difference between taking years and days off a miner’s life span.
Instead Zeigler grumbles that a general practitioner should not opine on matters beyond his expertise. Frederick Ridge, who had been Eugene’s physician, did not treat either cancer or pulmonary disease and referred Eugene to specialists for both. Why, then, should the agency accept Ridge’s view about the cause of Eugene’s death, Zeigler wonders? It relies on Peabody Coal Co. v. McCandless, 255 F.3d 465 (7th Cir.2001), for the proposition that medical diagnoses must be supported by medical reasoning; a history of treatment is no substitute for scientific analysis. True enough, but as the Secretary observed when promulgating § 718.205(c)(5), the proposition that persons weakened by pneumoconiosis may expire quicker from other diseases is a medical point, with some empirical support. See 65 Fed.Reg. 79,920, 79,950 (Dec. 20, 2000). Whether this was true of Eugene is a question that depends on the state of his physical and mental health, a subject about which Ridge professed to be knowledgeable.
One way to respond would be to show that Eugene’s death occurred at the expected time after all. The oncologist’s estimate of six months’ life expectation was simplified. A more complete statement would have included the normal variability, such as six months plus or minus three months, for colon cancer that had metastasized to the liver (Eugene’s state when his cancer was detected). Suppose the actual expectation for a condition similar to Eugene’s was 6±4 months for persons without any other complicating disease. Then it would be hard to see how Eugene’s death could be said to have been hastened by black lung disease. No such evidence is in the record, however.
Another potential response might have come from epidemiologic evidence. Suppose that a large-scale study of persons diagnosed with black lung disease showed that they survive as long after a diagnosis of cancer as other persons do. Then although it may be sensible to think that pneumoconiosis hastens death in conjunction with some other disease, we would know that cancer (or colon cancer specifically) was not one of them. Black lung disease weakens its victims, but maybe not in ways that matter for particular maladies. But Zeigler did not offer such evidence. Perhaps it does not exist — or perhaps it exists but favors miners by showing that black lung does hasten death from cancer. Either way, there is nothing in the record to counter Ridge’s conclusion, which therefore supplies substantial evidence to support the administrative decision.
Zeigler doubtless would dispute our statement that there is “nothing” in the record to counter Ridge’s assessment. It introduced an affidavit from Peter Tuteur, a professor at Washington University School of Medicine, opining that Eugene’s pulmonary problems did not accelerate his death. But this affidavit is worthless, because Dr. Tuteur gives no reason. He did [336]*336not examine Eugene, did not discuss the normal variability in the time it takes colon cancer to kill its victims, did not point to epidemiologic studies, did not cite a single article in the medical literature — in sum, did not help the decisionmaker. “An expert who supplies nothing but a bottom line supplies nothing of value to the judicial process.” Mid-State Fertilizer Co. v. Exchange National Bank, 877 F.2d 1333, 1339 (7th Cir.1989). Neither Ridge nor Tuteur is a cancer specialist; and between the two, Ridge at least had the advantage of observing whether a pulmonary problem sapped Eugene’s ability to withstand the effects of the cancer.
Some federal agencies refuse to make decisions on the basis of reports such as the one Ridge (or for that matter Tuteur) supplied. For example, the Food and Drug Administration would not dream of deciding whether some new treatment postpones death from cancer (an issue equivalent to whether black lung hastens death) without a statistically sound epide-miologic study or a well-conducted double-blind experiment. Reports from physicians in the field along the lines of “I gave drug X and it kept the patient alive for four extra months” would be dismissed as worthless anecdotes. While the fda and other agencies, such as the Environmental Protection Agency, insist on statistically valid results, the Department of Labor in black lung cases (and the Social Security Administration in disability cases) rely on evidence that does not carry any indicators of statistical power. But Zeigler, which took the same approach by proffering only Dr. Tuteur’s unreasoned assertion, is in no position to complain. Unless mine operators show that the Department’s approach to these matters is medically unsound, it is entitled to proceed as it did here.
AFFIRMED-