Donald E. Robinson appeals a final order of the Benefits Review Board denying his claim for benefits under the Black Lung Benefits Act. 30 U.S.C. §§ 901 and 945 (1982). The case is back to this court after remand and reconsideration by the Administrative Law Judge ordered in Robinson v. Picklands Mather & Co./Leslie Coal Co., 914 F.2d 35 (4th Cir. 1990) (Robinson I ).

* In Robinson I, we reviewed the ALJ's decision denying benefits on his finding that Robinson's coal workers' pneumoconiosis did not cause his total disability. In reversing his decision, we held that a black lung claimant need only show that statutory pneumoconiosis contributed to his total disability. The issue we now review is whether substantial evidence supports the ALJ's decision on remand that Robinson's pneumoconiosis did not contribute to his total disability. The ALJ, on remand, took no new evidence and proffered only an abbreviated review of the conflicting record evidence. To resolve the issue, we refer to that evidence as it was developed in Robinson I.

Robinson filed his current application for benefits on March 31, 1983. For reasons not entirely clear, a hearing on his claim was not held until June 8, 1988. At that hearing, Robinson testified and the ALJ found that he had been employed in underground coal mines for 35 years. Six medical doctors testified, but the testimonies of only three bear on the issue facing us in this appeal. Doctors George Zaldivar, Peter G. Tuteur, and D.L. Rasmussen agreed that x-ray evidence shows that Robinson suffers from pneumoconiosis, and they agreed that he is afflicted with a permanently disabling respiratory disease.

The doctors' opinions on the cause of Robinson's disability varied. Dr. Rasmussen reported a chronic productive morning cough, wheezing with exertion, and anterioral discomfort with exertion. He reported breath sounds as essentially normal with no abnormal sounds. Ventilatory function studies and maximum breathing capacities were within normal limits, but his single breath carbon monoxide diffusing capacity was only 43% of normal. Rasmussen also reported that gas exchange was markedly impaired, and on exercise, his volume of ventilation markedly increased. Robinson exceeded his anaerobic threshold at about 52% of his predicted maximum oxygen consumption. He noted that Robinson's dead space ventilation was increased and that he was markedly hypoxic after exercise. Dr. Rasmussen concluded:

These studies indicate very severe pulmonary insufficiency as reflected by the reduced single breath carbon monoxide diffusing capacity and the marked and progressive impairment in gas exchange, and hypoxia during exercise.

The pattern of impairment in this case is consistent with an interstitial type lung disease of which coal pneumoconiosis is an example. This patient is totally disabled for any significant, gainful employment as a consequence of his severe respiratory impairment.

Dr. Zaldivar, examining Robinson at the instance of Picklands (the employer), reported on October 6, 1986:(1) radiographic evidence of simple pneumoconiosis, (2) normal ventilatory study, (3) normal resting blood gas with impairment, and (4) severe impairment of diffusing capacity. He concluded:

The chest x-ray is compatible with coal worker's pneumoconiosis. However the pattern of the ventilatory study is much more easily explainable by the presence of pulmonary fibrosis [scarring of the lungs] which is an entity not usually related to coal worker's pneumoconiosis. The pattern of normal ventilatory study with a severe abnormality of diffusing capacity is specifically found in entities which cause pulmonary fibrosis. Without a lung biopsy, a definite diagnosis cannot be made. Therefore as it stands with information at hand, Mr. Robinson has coal worker's pneumoconiosis and pulmonary fibrosis which is probably not resulting from it. The pulmonary fibrosis has impaired his capacity to work to the level above the light or moderate.

Dr. Tuteur, a Washington University medical professor, did not examine Robinson but was retained by Picklands to offer an opinion based on the reports of the other doctors. He did this in his report of February 17, 1988. After his review of the other doctors' reports, Dr. Tuteur stated:

Mr. Donald Robinson ... worked in the coal mine industry for 35 years predominantly as an underground miner, when he discontinued employment in 1982. Clearly, he was exposed to sufficient amounts of coal mine dust to produce coal workers' pneumoconiosis in a susceptible host. In addition, he smoked cigarettes regularly, often as much as one and one-half packs per day and possibly more for a total of 40 years. This, too, puts him at increased risk for the development of cigarette smoke associated health problems such as chronic obstructive pulmonary disease (chronic bronchitis/emphysema), arteriosclerotic heart disease and/or lung cancer. Unequivocally, this is consistent with clinically significant, physiologically significant, and radiographically significant coal workers' pneumoconiosis. In fact, pending any other explanation for the radiographically demonstrable interstitial pulmonary process and the gas exchange impairment during exercise, because of the long history of coal mine dust exposure, this is the most likely diagnosis.

Dr. Tuteur also considered, however, the significance of the reported difficulty Robinson had with his knees-particularly his left knee. He stated:

When one linked the bilateral lower extremity problems with the history of both pleurisy and pneumonia, one raises an important diagnosis that must be differentiated from coal workers' pneumoconiosis as the clinically significant, physiologically significant, and radiographically significant process. This becomes even more important recognizing that simple coal workers' pneumoconiosis, even with only mild radiographic changes such as Category 1, may be associated with substantial impairment of gas exchange and clinical symptoms. This is unusual, but it does occur.

Although no objective evidence showed a blood clot arising from Robinson's condition in his lower extremities, Dr. Tuteur went on to explore the possibility that his respiratory symptoms resulted from a blood clot migrating to the lung causing pulmonary emboli. He considered each of the symptoms reported after the physical examinations and laboratory and pulmonary function tests to conclude that the reported conditions could be symptomatic of either coal workers' pneumoconiosis or pulmonary emboli. In his deposition testimony, he stated:

What I'm talking about is that he clearly was exposed to sufficient amounts of coal mine dust in his thirty-five plus years as an underground miner to put him at significant risk for the development of coal workers' pneumoconiosis if he is a susceptible host.

....

I think with reasonable medical certainty, the problem responsible for his difficulties is either pulmonary embolism or coal workers' pneumoconiosis. I can't tell you which one.1

Dr. Tuteur suggested two tests to establish a certain diagnosis: a pulmonary arteriogram or, alternatively, a radionuclide study, known as a ventilation-profusion lung scan. In his cross-examination, in response to the question, "In your report, though, you do make the determination that coal workers' pneumoconiosis is the most likely diagnosis," he answered, "Yes." Nonetheless, Dr. Tuteur explained, he was unable to make a conclusive diagnosis between coal workers' pneumoconiosis and pulmonary emboli. On cross-examination, however, he added:

In my opinion, when a coal worker of thirty-five years has an interstitial pulmonary process and has no other identifiable or suggested cause for this, then one should conclude with reasonable medical certainty, ninety-five percent certainty, that he has coal workers' pneumoconiosis, recognizing there is a small chance that he does not. But I think that it is a very inefficient way of practicing medicine to have to get definitive one hundred percent certainty diagnoses on every patient.

Dr. Zaldivar, in his deposition testimony, repeated the diagnosis in his report and explained that in his opinion, limited diffusing capacity such as Robinson's was not normally found in simple coal workers' pneumoconiosis. Rather, limited diffusing capacity was consistent with pulmonary fibrosis. He also opined that Robinson's condition was not related to his history of cigarette smoking because those symptoms would be indistinguishable from coal workers' pneumoconiosis. Dr. Zaldivar stated that he could not make a definite diagnosis without a biopsy:

I would be at a loss to find out the exact cause of the impairment without a lung biopsy, that is true. But one thing I can say is that it's not due to simple pneumoconiosis, but there are many other things in there that could be causing it.

He admitted that Robinson had a mild degree of "air trapping" due to a combination of coal dust exposure and smoking, but apparently did not find this disabling.

Dr. Rasmussen disagreed with Dr. Zaldivar's assertion that Robinson's impaired diffusing capacity could not be a symptom of simple coal workers' pneumoconiosis. Contrary to Dr. Zaldivar, Dr. Rasmussen stated that impaired diffusing capacity was indeed a symptom compatible with coal workers' pneumoconiosis. Moreover, both Drs. Tuteur and Rasmussen disagreed with Dr. Zaldivar's opinion that interstitial fibrosis is not normally found in patients with coal workers' pneumoconiosis. Dr. Rasmussen's final statement after reviewing Dr. Zaldivar's report was:

There is no medical reason to conclude that Mr. Robinson suffers from something other than coal workers' pneumoconiosis. He does, however, have a history of significant cigarette smoking, which could play an additive role in his overall impairment.

As we pointed out in Robinson I, under 20 C.F.R. § 718.2, a miner must prove that he has pneumoconiosis, that the disease arose out of his coal mine employment, and that he is totally disabled due to the disease. 914 F.2d at 36. The ALJ determined that Robinson had established the first two of these requirements, but that he had failed in demonstrating his disability was linked to the disease. In Robinson I, we held that a claimant need only show that statutory pneumoconiosis was a contributing cause of his totally disabling respiratory impairment. Id. at 38. We reversed the ALJ's decision primarily because it applied the old rule requiring total causation.2 Also we were concerned with the ALJ's heavy reliance on Dr. Zaldivar's testimony because it was unclear whether Dr. Zaldivar considered the broad statutory definition of pneumoconiosis in 20 C.F.R.s 718.201. Section 718.201 states in part:

This definition includes, but is not limited to, coal workers' pneumoconiosis, anthracosilicosis, anthracosis, anthrosilicosis, massive pulmonary fibrosis, progressive massive fibrosis, silicosis or silicotuberculosis, arising out of coal mine employment. For purposes of this definition, a disease "arising out of coal mine employment" includes any chronic pulmonary disease resulting in respiratory or pulmonary impairment significantly related to, or substantially aggravated by, dust exposure in coal mine employment.

We instructed the ALJ, on remand, to determine whether Dr. Zaldivar's medical opinion could support a finding that Robinson's disability did not arise out of coal mine employment. He was instructed that if Dr. Zaldivar's opinion was "not consistent with the broad legal definition of pneumoconiosis found in the regulations," and if he (the ALJ) found "that Robinson's disability was caused by his pneumoconiosis as defined above, then benefits should be awarded." Robinson I, 914 F.2d at 39.

On remand, the ALJ acknowledged his responsibility to determine whether Dr. Zaldivar's opinion could support a finding that Robinson's disability did not arise out of coal mine employment. Finding that he could not determine whether Dr. Zaldivar considered the broad legal definition of pneumoconiosis, the ALJ went on to say that "even if Dr. Zaldivar's opinion is disregarded, claimant under Rose [v. Clinchfield Coal Co., 614 F.2d 936, 938 (4th Cir. 1980) ], still must prove that he suffered from coal mine related respiratory disease which contributed to his total disability." He stated: "Based on my analysis of the opinions of Drs. Rasmussen and Tuteur, I conclude that the claimant has not met this burden. For this reason, the claim is denied."

The ALJ's reference to his analysis of the opinions of Drs. Rasmussen and Tuteur was to his first decision, which we reversed in Robinson I.3 In Robinson I, of course, we concluded that the ALJ had denied a disability award primarily on the basis of Dr. Zaldivar's report and testimony. In response, on remand, the ALJ restructured his Robinson I conclusions, relying solely on the opinions of Drs. Rasmussen and Tuteur to find that Robinson did not satisfy his burden of showing that pneumoconiosis was a "contributing cause" of his disability. In our view, his resulting conclusion is not supported by substantial evidence.

Dr. Rasmussen characterized Robinson's medical condition as a "severe respiratory impairment" resulting from an interstitial lung disease, which he equated with coal workers' pneumoconiosis. Critiquing Dr. Zaldivar's opinion on Robinson's limited diffusing capacity, Rasmussen testified that the diffusing capacity test results reflected a disability arising out of coal mine employment. Dr. Zaldivar had opined that the limitation could not have resulted from coal dust exposure because in his view, limited diffusing capacity was never caused by coal workers' pneumoconiosis. He based that opinion on the results of experiments reported in two textbooks. Dr. Rasmussen explained in his critique that the reported experiments relied on by Dr. Zaldivar had been conducted on asymptomatic miners (i.e., those with x-ray evidence of pneumoconiosis but no impairment). In contrast, Robinson was totally impaired in the opinions of all three reporting doctors. Dr. Rasmussen explained further that the textbooks did not address the impact that disabling pneumoconiosis has on diffusing capacity. Finally, Dr. Rasmussen stated:

There is no medical reason to conclude that Mr. Robinson suffers from something other than coal workers' pneumoconiosis. Dr. Tuteur agreed with Dr. Rasmussen and disagreed with Dr. Zaldivar on two highly probative medical opinions. He agreed that the decreased diffusion results were consistent with coal workers' pneumoconiosis and that interstitial fibrosis is consistent with coal workers' pneumoconiosis. Dr. Tuteur also found significant Robinson's 35 1/2 years of underground mining activity. He opined that coal workers' pneumoconiosis was the most likely diagnosis but indicated that this could not be determined with a reasonable degree of medical certainty, which he defined as 95%. Applying this medical standard of reasonable certainty, he assessed that Robinson's disability was caused by either pulmonary emboli or coal workers' pneumoconiosis.

We are persuaded that Robinson has sustained his burden of proving by a preponderance of the evidence that his demonstrated pneumoconiosis contributed to his total disability and that the ALJ's decision to the contrary is not supported by substantial evidence. As we explained in Robinson I, the broad definition of pneumoconiosis in § 718.201 " 'effectively allows for the compensation of miners suffering from a variety of respiratory problems that may bear a relationship to their employment in the coal mines.' " Robinson I, 914 F.2d at 39 (quoting Rose, 614 F.2d at 938). There is no question that Robinson is totally disabled, nor is there any question that he suffers from pneumoconiosis. Dr. Rasmussen's reports as corroborated by Dr. Tuteur's testimony preponderantly establishes a causal link between Robinson's pneumoconiosis and his total disability. As said, the ALJ's interpretation of the facts is simply not supported by substantial evidence.

The decision of the Board is therefore reversed and remanded with directions to award Robinson benefits.