Opinion
Civil Action No. 03-3109.
June 22, 2005
MEMORANDUM
After a psychiatrist attributed his physical symptoms to a panic disorder, Mark Warren applied for Social Security disability insurance benefits. The Commissioner of the Social Security Administration (the "Commissioner") denied Warren's claim based on her finding that his co-morbid alcoholism was a contributing factor material to his alleged disability. Warren appealed from that determination, and the parties' motions for summary judgment are before us.
Factual Background
Our inquiry focuses on whether Warren was disabled between October 1, 1996, the date on which he claims to have become disabled, and December 31, 2000, the last date on which he was insured. To put the evidence from this period in its proper context, we shall review the entire record.
A. Pre-Disability Records
On September 9, 1993, Warren met with Dr. Seth J. Worley, a cardiologist, because he was experiencing chest pain. Dr. Worley noted that Warren "drinks 5 to 6 beers per day," and blood tests revealed a gamma-glutamyl tanspeptidase ("GGT") level of 409, which is much higher than normal. See Rec. 373; see also id. at 374. When further testing failed to disclose any heart problems, Dr. Worley concluded that Warren's "chest symptoms do not seem to be cardiac in origin" and stated that "he will need to stop drinking." Id.; see also id. at 402 (noting that Warren drank "[b]eer daily").
GGT, a liver enzyme, is "often elevated in heavy drinkers, making it of potential value as an indicator of drinking status." Rec. 332; see also id. at 696 (describing GGT as an "alcohol related enzyme" and describing a normal range as "8 to 73").
It is not clear whether Dr. Worley actually told Warren to stop drinking or whether he simply noted in his report that Warren ought to stop drinking.
Nearly a year later, in late July, 1994, Warren elected to receive arthroscopic surgery on his right knee. See Rec. 372. Although the surgery appears to have been successful, see id. at 383, 386, Warren suffered a seizure after he returned home on July 28, 1994. His wife, who is a licensed nurse, immediately brought him to the emergency room at Lancaster General Hospital, and he experienced a second seizure there. In the course of evaluating his condition, Dr. Roy S. Small learned that Warren consumed "at least 12 ounces of alcohol a day" and that his GGT level was 498. Id. at 379 (footnote added). That evidence was "consistent with alcohol abuse" and suggested to Dr. Small that Warren's seizure might have been "alcohol withdrawal related."Id. at 380. Dr. William H. Adams also treated Warren in the emergency room, and he concurred that there was a "likelihood of a possible alcohol withdrawal seizure." Id. at 377. Though Dr. Small appears to have reported that Warren drank at least 12 beers per day, Dr. Adams stated that Warren drank "approximately one case a day of beer." Id.
Another doctor opined that the reference to "12 ounces" ought to be read literally (i.e., Warren consumed the equivalent of 12 ounces of pure alcohol per day) rather than figuratively (i.e., Warren consumed 12 ounces of alcoholic beverages per day). See Rec. 707. Since a can of beer contains approximately one ounce of pure alcohol, the literal reading of this evidence suggests that Warren drank 12 cans of beers per day. See id. We express no opinion on whether the literal or figurative reading is proper.
A "case" contains 24 cans of beer. See Rec. 535.
After Warren was admitted to the hospital, Dr. C.P. Binning, a neurosurgeon, examined him and then opined that his seizures were "typical alcohol related withdrawal seizures." Rec. 389. Out of an abundance of caution, Dr. Binning advised Warren not to drive until he could confirm that diagnosis and that he should curtail his drinking, though he ought "not to stop the beer abruptly." Id. at 384; see also id. at 389. By August 12, 1994, Warren had reduced his alcohol consumption from "up to a case of beer every day" before the seizure to "2 to 3 cans a day," but Dr. Binning "advised him to stop drinking further."Id. at 384-85. Warren was "drinking only one or two cans of beer daily" by the end of August, 1994. Id. at 381. This progress encouraged Dr. Binning enough to recommend that the Department of Transportation allow Warren to resume driving, in spite of his history of seizures. Id. at 381-82. Though Warren had stopped drinking entirely by September 16, 1994, see id. at 400, he again reported drinking "two cans of beer on a daily basis" on December 19, 1994, id. at 399.
Dr. I. Stanley Porter recognized that Dr. Binning believed that "ethanol withdrawal" caused Warren's seizures when he performed an orthopedic consultation. See Rec. 376.
B. Early Treatment of Disability
About one and one half years later, in the summer of 1996, Warren began experiencing shortness of breath and sought treatment with Dr. P. James Navalkowsky. Dr. Navalkowsky ordered a routine blood test, which revealed GGT levels of 96, still above normal, but far below the levels they reached when Warren was drinking heavily in 1994. See Rec 348. Dr. Navalkowsky also referred Warren to Dr. Lloyd G. Goldfarb, a pulmonologist. After performing extensive tests, however, Dr. Goldfarb could not isolate any pulmonary problems. See id. at 340-59. Since Dr. Goldfarb suspected that "diastolic cardiac dysfunction" might be causing Warren's shortness of breath, see id. at 346, Dr. Navalkowsky referred Warren to Dr. Worley, who is a cardiologist. Dr. Worley also performed a battery of tests, but he did not detect any problems with Warren's heart. See id. at 339, 414-16, 367.
On September 30, 1996, Warren told Dr. Worley that he drank alcohol "daily." Rec. 414.
Having ruled out the most probable physical causes of his symptoms, Warren sought treatment from a psychiatrist, Dr. David E. Nutter. In their first meeting, on October 31, 1996, Dr. Nutter noted that Warren "[h]ad [a] problem [with] drinking — going back to divorce." Rec. 482. Warren and Dr. Nutter met four more times in 1996 so that Dr. Nutter could gauge the proper dosages for each of the several medications that he had prescribed. Id. at 483-84. On December 12, 1996, Dr. Nutter informed Warren's employer, Continental Press, that Warren "suffer[ed] from significant panic disorder with agoraphobia as well as depression" and had been "unable to work at least since October 21, 1996." Id. at 366.
Warren was divorced in 1987, at the latest. See Rec. 555 (reporting that he remarried eleven years before 1998).
As the treatment continued throughout 1997 and 1998, Dr. Nutter occasionally noted Warren's drinking. See Rec. 487-503. On March 6, 1997, after learning that he was "drinking 2, 6 packs per day," Dr. Nutter opined that Warren was "self medicating [with] alcohol" and told him that he "need[ed] to [reduce his] alcohol" consumption. Id. at 486. On February 19, 1998, Warren told Dr. Nutter that he was "decreasing his alcohol consumption," and Dr. Nutter again encouraged him to further "[d]ecrease alcohol by 10% a week." Id. at 495. Warren had "decreased his consumption in half" by March 26, 1998. Id. at 496. On April 30, 1998, Warren informed Dr. Nutter that he was drinking "about 1/4 of a keg about every 2 weeks." Id. at 497.
A "quarter keg" or "quarter barrel" contains 7.75 gallons, or 992 fluid ounces, of beer. See http://www.beveragefactory.com/faqs/draftbeer.shtml (last visited June 9, 2005). This volume is equivalent to about 83 12-ounce cans of beer. If Warren consumed that much alcohol in two weeks, his average consumption would be roughly equivalent to 6 cans of beer per day, half of what he had been drinking in March, 1997.
C. Initial Application for Benefits
Warren applied for Social Security disability insurance benefits on March 12, 1998. In his application, he claimed that "significant panic disorder — agoraphobia — depression" had prevented him from working since October 1, 1996. Rec. 322; see also id. at 285-87, 318-23. About two weeks after submitting his application, Warren completed a Daily Activities Questionnaire stating that he had difficulty going out in public.Id. at 311; see also id. at 309-13.
Pages 318 through 323 of the record constitute a six-page Disability Report (Form 3368), but the pages of the Report are not in numerical order in the record.
As part of its claims processing procedure, the Social Security Administration ("SSA") asked Michael Zhin, one of Warren's friends, to complete a daily activities questionnaire describing Warren's lifestyle. See id. at 293-97. In response, Zhin reported that his friendship with Warren had deteriorated since 1996 because Warren "rarely le[ft] the house." Id. at 294. Zhin also stated that Warren "cannot hold a job" and "does not fit the `mold' of [a] normal working class husband." Id. at 296.
The SSA also asked Dr. Nutter to evaluate Warren's condition.See Rec. 417-21. In his March 31, 1998 response, Dr. Nutter explained that Warren had been working at a "job he liked" when he began to experience "chest tightness, nervous[ness], tens[ion], and shortness of breath" in the summer of 1996. Id. at 418. Warren could sometimes go "for several weeks" without experiencing a panic attack, but he would then "get of flurry" of them. Id. at 420. Dr. Nutter described Warren as "[d]epressed, very discouraged" and noted that he had "poor concentration, forgets a lot, [and] worries about his memory." Id. at 419. Although Dr. Nutter had implemented "a very aggressive anti-depressant, anti-anxiety pharmacotherapy regimen," Warren had shown "no great improvement" and his "overall functioning [was] tenuous." Id. at 418, 421. Sometimes Warren would "get panicky" and miss appointments, but Dr. Nutter described him as "highly motivated to get better." Id. at 421.
After receiving Dr. Nutter's report, the SSA forwarded Warren's file to Dr. William Myers for a psychiatric review. On April 8, 1998, Dr. Myers indicated that Warren suffered from an affective disorder and an anxiety related disorder with "[r]ecurrent severe panic attacks . . . occurring on the average of at least once a week." Rec. 475. In Dr. Myers's opinion, those disorders moderately restricted Warren's activities of daily living, caused slight difficulties in his maintenance of social functioning, and often resulted in failure to complete tasks in a timely manner.See id. at 478. Dr. Myers also performed a mental residual functional capacity ("RFC") assessment of Warren, finding moderate limitations in both his ability to maintain attention and concentration for extended periods and his ability to complete a normal workday and workweek. See id. at 467-68. Dr. Myers concluded that Warren retained the "RFC for simple jobs," id. at 469.
Dr. Myers also stated that Dr. Nutter made "no mention of alcohol." Rec. 472; see also id. at 450 (handwritten notation). Dr. Myers found "no mention of alcohol" in Dr. Nutter's records because he reviewed only Dr. Nutter's March 31, 1998 response to the SSA's inquiries. See id. at 417-21. While Dr. Nutter did not discuss Warren's use of alcohol in his response, the SSA did not ask him about that subject.
Unlike the March 31, 1998 report, Dr. Nutter's handwritten treatment notes, which are part of Exhibit 11F, contain many references to Warren's alcohol use. See Rec. 480-503. Because Exhibit 11F became part of the record all at once, and because it contains a document dated November 12, 1998, id. at 480, Exhibit 11F could not have been become part of the record before November 12, 1998. Thus, Dr. Myers could not have reviewed Exhibit 11F (including Dr. Nutter's treatment notes) when he prepared his April 8, 1998 report.
Based on Dr. Myers's evaluation, the SSA denied Warren's claim on April 15, 1998. See Rec. 112-14. Warren requested reconsideration of the decision on May 18, 1998. See id. at 298-301.
Pages 298 through 301 of the record constitute a four-page Reconsideration Disability Report (Form 3441), but the pages of the Report are not in numerical order in the record.
On June 9, 1998, Dr. Nutter left a voice message with the SSA that summarized his treatment of Warren since March 31, 1998. Although his message for the most part restated the contents of his treatment notes, compare Rec. 447-49 with id. at 496-99, Dr. Nutter failed to disclose to the SSA that Warren had resumed drinking "about 1/4 of a keg about every 2 weeks," id. at 497, instead stating generally that Warren had "been coming back on the use of alcohol which he has been drinking to alleviate his anxiety," id. at 447. Dr. Nutter's message concluded with his diagnosis that Warren was "homebound . . . with significant panic attacks with agoraphobia." Id. at 448.
The SSA later transcribed the voice message and the transcription is part of the record. See Rec. 447-49.
Dr. Roger K. Feitz relied on Dr. Nutter's message in conducting his June 12, 1998 psychiatric review. See Rec. 454-66. Like Dr. Myers, Dr. Feitz believed that Warren suffered from an affective disorder and an anxiety related disorder with "[r]ecurrent severe panic attacks . . . occurring on the average of at least once a week." Id. at 462. He concurred with Dr. Myers's assessment that the disorders moderately restricted Warren's activities of daily living and often resulted in failure to complete tasks in a timely manner. Whereas Dr. Myers believed that Warren had only slight difficulties in maintaining social functioning, Dr. Feitz found moderate difficulties in that regard. Dr. Feitz also identified one or two episodes of deterioration or decompensation in work or work-like settings, while Dr. Myers found no such episodes. See id. at 465. Despite finding moderate limitations in Warren's understanding and memory, concentration and persistence, and adaptation, Dr. Feitz concluded that he "would be able to perform simple, routine tasks." Id. at 456.
Dr. Feitz also noted a "question of alcohol involvement," but found "no indications . . . regarding excessive use." Rec. 456. Of course, only Dr. Nutter's notes would have revealed the full extent of Warren's alcohol consumption, and Dr. Feitz did not have access to them on June 12, 1998. See supra note 10.
Though Dr. Feitz reported more severe limitations than Dr. Myers, the SSA again denied Warren's claim for benefits on June 22, 1998, see Rec. 107-09, so Warren requested a formal hearing before an administrative law judge ("ALJ"), id. at 100.
D. First ALJ Decision
While he waited for the hearing, Warren continued to receive treatment from Dr. Nutter. On July 29, 1998, Dr. Nutter reported that Warren had "cut his alcohol down to 1 keg/9 days." Rec. 501. Dr. Nutter also authored a report, dated November 12, 1998, which diagnosed Warren with "a particularly severe form of panic disorder with agoraphobia that is incapacitating." Rec. 480.
A "keg" (also called a "half barrel") contains 15.5 gallons, or 1984 fluid ounces, of beer. See http://www.beveragefactory.com/faqs/draftbeer.shtml (last visited June 9, 2005). Drinking one keg every nine days is the equivalent of drinking more than 18, 12-ounce cans of beer per day.
Since Dr. Nutter noted on April 30, 1998 that Warren was drinking "about 1/4 of a keg about every 2 weeks," id. at 497, it seems peculiar that he would consider Warren's July 29, 1998 report of drinking 1 keg every nine days to be a reduction in his alcohol consumption.
On November 23, 1998, an ALJ convened a hearing on Warren's application for benefits. At the hearing, Warren testified that he began drinking about a case of beer a day "right before" the onset of his panic attacks in 1996. Rec. 535-36. Dr. Nutter had directed him to reduce his alcohol consumption gradually (but not to stop drinking entirely), so Warren stated that, by the date of the hearing, he had reduced his drinking to "[m]aybe a case a week between [his] wife" and himself. Id. at 535. Warren's wife confirmed that he had been drinking as many as 12 cans of beer a day when his panic attacks began in 1996 and that he was drinking about 6 cans of beer each day at the time of the hearing. Id. at 557-58.
This level of consumption corresponds to less than 4 cans of beer per day for Warren alone.
On January 13, 1999, the ALJ decided that Warren was not disabled, see Rec. 67-83, and Warren promptly requested review of that decision, see id. at 118. The Appeals Council vacated the ALJ's decision on May 25, 2000 and remanded the case for further development of the record. See id. at 119-22.
E. Second ALJ Decision
In anticipation of another hearing, Dr. Nutter submitted a report dated August 10, 2000. See Rec. 509-12. The report stated that Warren continued to suffer from panic disorder with agoraphobia that had not improved since 1998. With regard to Warren's RFC, Dr. Nutter explained that Warren was not capable of attending work or work-like activities because of "very severe attacks of panic that can occur suddenly without warning and essentially paralyze his intellectual and emotional functioning."Id. at 511. Dr. Nutter also noted that, "[e]ven on 4 medications at the same time[,] he has not made significant recovery of his illness." Id. Significantly, Dr. Nutter opined that Warren's mental condition "caused" or "le[d] to" his use of alcohol. Id. at 510.
We attribute authorship of the August 10, 2000 report to Dr. Nutter, even though Warren's attorney drafted the typed portions of the report, because the attorney based his draft on a conversation with Dr. Nutter, because Dr. Nutter had the opportunity to correct and/or supplement the draft, and because Dr. Nutter verified that the statements in the report were true and correct. See Rec. 654 (confirming that the report reflects Dr. Nutter's opinions).
On August 16, 2000, a different ALJ convened the second hearing on Warren's application for disability benefits. See Rec. 574-636. Warren testified that, since the first hearing when he was drinking about 6 cans of beer each day, he had stopped drinking every day and, when he did drink, he only drank at meals. Id. at 589-90. Essentially confirming that account, Warren's wife stated that he did not drink every day "anymore" and that he only drank a "few" cans of beer a "couple" of times a week. Id. at 591.
Dr. Robert S. Brown, Jr. — who is board certified in internal medicine, psychiatry, and forensic psychiatry — also testified at the August 16, 2000 hearing. Relying on the DSM-IV, Dr. Brown criticized Dr. Nutter's diagnosis of panic disorder with agoraphobia for two reasons. First, Dr. Brown believed that Dr. Nutter "ignor[ed]" the effects that alcohol "can" have on the severity of panic disorders. Rec. 610, 611, 612. Alcohol consumption can "make [a panic disorder] worse" when a person is consuming alcohol and when he is sobering up, id. at 607-09; see also DSM-IV 177; DSM-IV-TR 193, but Dr. Nutter's records did not explicitly mention those effects.
American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders (4th ed. 1994) [hereinafter DSM-IV]. A more recent edition of the DSM contains identical information about all issues relevant to this case. See American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders (4th ed. text revision 2000) [hereinafter DSM-IV-TR].
Dr. Brown stated elsewhere that "anxiety disorders are produced by the substance abuse of alcohol . . . [a]t both intoxication and withdrawal." Rec. 609-10 (emphasis added). This statement is not identical to the statement in the text. The statement in the text means that a person with an anxiety disorder would experience more severe symptoms when using alcohol, but the statement in this note means that alcoholabuse can actually cause anxiety disorders in people who do not have psychological problems.
Dr. Brown's second criticism of Dr. Nutter was that he should not have diagnosed "panic disorder with agoraphobia . . . until [he] eliminated the idea that alcohol . . . caused" Warren's symptoms. Id. at 627; see also id. at 630 ("I am not allowed under the [DSM] to diagnose panic disorder with agoraphobia, if there's a substance abuse problem going on that may have caused the . . . substance induced anxiety disorder, or substance induced affective disorder."); DSM-IV 402 (explaining that a diagnosis of panic disorder with agoraphobia requires a finding that the "Panic Attacks are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition"); DSM-IV-TR 441 (same). According to Dr. Brown, the "only way" to distinguish panic disorder with agoraphobia from substance induced anxiety disorder is to evaluate the patient's symptoms six months after he has stopped drinking. Rec. 629-30; see also id. at 628, 630-31. Since Warren did not stop drinking for at least six months, Dr. Brown did not believe that Dr. Nutter could make a valid diagnosis of panic disorder with agoraphobia.
But see DSM-IV 193 ("As a rule of thumb, symptoms that persist for more than 4 weeks after the cessation of acute Intoxication or Withdrawal should be considered to be manifestations of an indpendent non-substance-induced mental disorder or of a Substance-Induced Persisting Disorder. Clinical judgment is necessary in making this distinction. . . .") (emphasis added); DSM-IV-TR 210 (same).
It bears noting that, while he certainly criticized Dr. Nutter's opinion, Dr. Brown did not offer any diagnosis of his own. For example, Warren's attorney asked Dr. Brown whether Warren's "depression [was] materially contributed to by" drinking five or six beers a day, and Dr. Brown responded that he could not "say [that such consumption] always [has that effect], but most often it" does. Rec. 626. Indeed, Dr. Brown explicitly declined to state whether Warren's alcohol use caused his disorder (or was making it worse), deferring instead to the ALJ's "judicial decision" on that issue. Id. at 612. When Warren's attorney asked Dr. Brown to diagnose Warren, Dr. Brown simply recognized Dr. Nutter's opinion and then deferred to the ALJ's determination of whether "other diagnoses should be considered."Id. at 617. Although Dr. Brown declined to diagnose Warren, he did opine that, if Warren "stop[ped drinking] for the rest of his life, . . . his panic disorder is going to be made better or going to go away." Id. at 611.
When Warren's attorney continued to press Dr. Brown on this point, Dr. Brown stated that he thought that Warren's condition was "alcohol related." Rec. 617. Alcohol Related Disorders include an extremely broad class of disorders, see infra note 27, so we do not consider Dr. Brown's identification of an "alcohol related" condition to represent a precise diagnosis.
Dr. Brown did not say how much "better" Warren's panic disorder would get or that Warren would not be disabled if he stopped drinking.
Since Dr. Nutter was not present at the August 16, 2000 hearing, the ALJ convened another hearing on November 15, 2000 to allow him to testify. See Rec. 637-68. At this hearing, Dr. Nutter reported seeing Warren on March 29, 2000 and November 1, 2000. Id. at 644. As of November 1, 2000, Warren had not stopped drinking, but he had decreased his alcohol consumption.Id. at 657. Dr. Nutter continued to believe that Warren suffered from both "panic attacks with agoraphobia" and an "alcohol abuse disorder." Id. at 644, 659. His diagnosis of panic attacks with agoraphobia was "independent of the alcohol problem." Id. at 655. In fact, Dr. Nutter suggested that Warren "may self medicate with alcohol" because "people[,] when they're getting anxious and nervous, . . . will start to try to tranquilize themselves with alcohol." Id. Unlike Dr. Brown, who believed that Warren's drinking intensified his panic disorder, Dr. Nutter stated that Warren's panic disorder would "get worse, at least temporarily[,]" if he stopped drinking. Id. at 656. Though he recognized that "[i]t is difficult to tell when somebody is drinking . . . how severe the panic disorder is,"id. at 662, Dr. Nutter concluded that Warren was "disabled whether or not he [continued to] drink alcohol. In other words, . . . the panic attacks with agoraphobia . . . is in itself disabling." Id. at 655; see also id. at 661. Thus, according to Dr. Nutter, Warren would be "as limited as [he was in November of 2000], if he were not drinking alcohol." Id. at 666.
The ALJ denied Warren's claim on February 22, 2001. See Rec. 86-97. On October 27, 2001, however, the Appeals Council vacated the ALJ's decision and remanded the case for further development of the record. See id. at 221-24. The Appeals Council was particularly concerned that the ALJ's "assessment of Dr. Nutter's medical opinion cannot be found substantially supported." Id. at 223.
F. Third ALJ Decision
During 2001 and early 2002, Warren again tried to curtail his alcohol consumption. For example, on April 11, 2001, Warren reported that he was "drinking — not on a daily basis," but he "sometimes ha[d] one or two beers." Rec. 518. Referring to Warren's ethanol consumption, Dr. Nutter on December 12, 2001 noted that "Etoh pt is going to try to completely stop." Id. at 516. Warren apparently made modest progress because "on New Years Eve[,] people offered him shots of alcohol, but he refused them."Id. at 515. By January 16, 2002, he had "gone to `Lite Beer.'"Id.
Despite that encouraging report, Warren's August 14, 2001 test results revealed a GGT level of 581, even higher than it was in 1994, when he admitted to drinking about a case of beer per day. Id. at 514; see also id. at 379.
The ALJ convened a fourth hearing on Warren's claim for disability insurance benefits on February 5, 2002. See Rec. 669-726. Reiterating that "substance abuse can worsen . . . whatever anxiety disorder is there," Dr. Brown testified that the "substance abuse issue has to be eliminated . . . before any definite thing can be said about . . . what's left of the anxiety disorder." Id. at 677. He was "not able to diagnose panic disorder with agoraphobia . . . until [he] eliminated the possibility that it is a substance abuse caused anxiety disorder." Id. at 684. According to Dr. Brown, Warren would have to stop drinking for "as long as six months" before he could diagnose a panic disorder. Id. at 704.
For Dr. Brown, "the issue . . . is [`]what started first[?']."Id. at 712. He concluded from the record that "the alcoholism, fully established[,] started first" and that "now we have, whatever we have, . . . with regard to anxiety disorders." Id. In other words, Dr. Brown concluded that, if Warren "ha[d] [an anxiety disorder], . . . it is substance abuse related, and/or caused or worsened" and that, "if he had any remaining anxiety disorder after all of the substance abuse ceased, it would not be terribly impairing and he'd be able to work." Id. at 676.
After the hearing concluded, the ALJ withheld decision so that Warren's attorney could supplement the record with additional medical evidence. Dr. Nutter submitted a report dated October 14, 2002 in which he stated that "Warren does meet the criteria for anxiety related disorders as defined by Social Security" and opined that "Warren would be disabled even if he stopped using alcohol." Rec. 520.
On October 4, 2002, Dr. Leo Dorozynsky performed an independent psychiatric evaluation of Warren and diagnosed him with "panic disorder with agoraphobia, as well as, alcohol abuse." Rec. 521. Dr. Dorozynsky noted that Warren "continue[d] to drink daily approximately 6 to 10 drinks" and "has had no significant periods of sobriety recently." Id. Echoing Dr. Brown, Dr. Dorozynsky recognized that "alcohol abuse can intensify or percipitate [sic] mood disorders including panic attacks[,] anxiety and depression." Id. Nevertheless, Dr. Dorozynsky, like Dr. Nutter, stated that Warren's alcohol use was "not the sole cause" of what he believed to be a "pre-existing panic disorder with agoraphobia." Id. Dr. Dorozynsky predicted that Warren "would continue to have some degree of impairment [if he stopped drinking]," but he could not "state whether Mr. Warren would still be disabled if he stopped using alcohol." Id. at 522.
On November 26, 2002, the ALJ again denied Warren's claim.See Rec. 14-30. Although the ALJ found that Warren was "unable to sustain any competitive employment," he found that Warren was not disabled because "alcohol abuse is a contributing factor material to the determination of his disability." Id. at 18. The Appeals Council denied Warren's request for review on March 28, 2003, see id. at 7-9, so the ALJ's November 26, 2002 decision became the final decision of the Commissioner, Sims v. Apfel, 530 U.S. 103, 107, 120 S. Ct. 2080, 2083 (2000). Warren has appealed that final decision to this Court. Analysis
A. Standard of Review
Pursuant to 42 U.S.C. § 405(g), we review the record to determine whether "substantial evidence" supports the ALJ's decision. Substantial evidence "means such relevant evidence as a reasonable mind might accept as adequate." Ventura v. Shalala, 55 F.3d 900, 901 (3d Cir. 1995). "A single piece of evidence will not satisfy the substantiality test if the [Commissioner] ignores, or fails to resolve, a conflict created by countervailing evidence. Nor is evidence substantial if it is overwhelmed by other evidence — particularly certain types of evidence (e.g., that offered by treating physicians) — or if it really constitutes not evidence but mere conclusion." Kent v. Schweiker, 710 F.2d 110, 114 (3d Cir. 1983). Though we must scrutinize the record carefully, we are not "empowered to weigh the evidence or substitute [our] conclusions for those of the fact-finder." Williams v. Sullivan, 970 F.2d 1178, 1182 (3d Cir. 1992). With these principles in mind, we shall consider whether substantial evidence supports the ALJ's conclusion that Warren was not disabled within the meaning of 42 U.S.C. § 423.
B. Determination of Disability
An individual is disabled if he is unable "to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment . . . which has lasted or can be expected to last for a continuous period of not less than 12 months." 42 U.S.C. § 423(d)(1)(A) (2005). Courts regularly refer to a "sequential five-step analysis" that the Commissioner uses to guide her disability determinations. See, e.g., Rutherford v. Barnhart, 399 F.3d 546, 551 (3d Cir. 2005); see also 20 C.F.R. § 416.920(a)(4) (2005).
In the first step of this analysis, the Commissioner considers whether the claimant is engaged in any substantial gainful activity. If the claimant is not engaged in any substantial gainful activity, the Commissioner proceeds to the second step where she considers whether the claimant has a severe medically determinable physical or mental impairment or combination of impairments. At the third step, the Commissioner assesses whether the claimant's severe impairments meet or equal one of the listings in appendix 1 to subpart P of part 404 of the Social Security regulations. If the claimant's impairments meet or equal at least one of the listings, then he is usually considered disabled. On the other hand, if the claimant's impairments do not meet or equal any of the listings, the Commissioner must proceed to the last two steps of the sequential analysis. In the fourth step, the Commissioner considers whether the claimant's residual functional capacity permits performance of past relevant work. If the claimant cannot perform past relevant work, the fifth and final step of the analysis focuses on whether, in light of his age, education and work experience, the claimant has the residual functional capacity to adjust to other work. If the claimant cannot adjust to other work, then he is usually considered disabled.
If the claimant is engaged in substantial gainful activity, then he is not disabled.
If the claimant does not have a severe medically determinable physical or mental impairment or combination of impairments, then he is not disabled.
If the claimant can perform past relevant work, then he is not disabled.
If the claimant could adjust to other work, then he is not disabled.
Though the SSA and the courts regularly treat the five-step process as if it summarized the entire analysis that the Commissioner must undertake, matters are not quite that simple in cases where a claimant may suffer from "alcoholism." In 1996, Congress amended the statutory definition of "disability" so that "[a]n individual shall not be considered to be disabled . . . if alcoholism . . . would . . . be a contributing factor material to the Commissioner's determination that the individual is disabled." Contract with America Advancement Act, Pub.L. No. 104-121, § 105(a)(1), 110 Stat. 847, 852 (1996) (codified at 42 U.S.C. § 423(d)(2)(C)).
The SSA's regulations explain that it will assess whether alcoholism is a contributing factor material to the determination of disability only if it "find[s] that [the claimant is] disabled and ha[s] medical evidence of [his] alcoholism." 20 C.F.R. § 404.1535(a) (2005). The "key factor" in the assessment is "whether [the Commissioner] would still find [the claimant] disabled if [he] stopped using . . . alcohol." § 404.1535(b)(1). When considering whether the claimant would be disabled if he stopped using alcohol, the Commissioner "evaluate[s] which of [his] current physical and mental limitations, upon which [she] based [her] current disability determination, would remain if [he] stopped using . . . alcohol and then determine[s] whether any or all of [his] remaining limitations would be disabling." § 404.1535(b)(2).
The SSA's internal operating manual clarifies how ALJs ought to incorporate their consideration of a claimant's alcohol use into the regular five-step disability analysis. See Social Security Administration, Hearings, Appeals and Litigation Law Manual I-5-3-14A (Aug. 24, 2000), available at http://www.ssa. gov/OP_Home/hallex/I-05/I-5-3-14-A.html [hereinafter "HALLEX I-5-3-14A" or "the manual"]. According to the manual, an ALJ must first "[d]ecide whether the individual would be disabled considering the effect of [alcoholism]." HALLEX I-5-3-14A(V)(B)(1). Presumably, the ALJ should use the traditional five-step analysis to make that decision. If the claimant is found to be not disabled, then the ALJ need not consider the effects of alcohol on his condition. See HALLEX I-5-3-14A(V)(B)(2).
On the other hand, if the ALJ concludes that the claimant would be disabled (taking into account the effects of alcohol), then the ALJ must "decide whether there is `medical evidence of [alcoholism].'" HALLEX I-5-3-14A(V)(B)(3). Medical evidence of alcoholism includes evidence from an acceptable medical source that is "sufficient and appropriate to establish that the claimant has a medically determinable substance use disorder." HALLEX I-5-3-14A(V)(C)(1). If there is no medical evidence of alcoholism, then the ALJ should find the claimant to be disabled.
The manual defines "substance use disorders" as "medical conditions described as `substance dependence' and `substance abuse' disorders in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (the DSM-IV)." HALLEX I-5-3-14A(V)(C)(3). Because the manual incorporates the DSM by reference, we shall summarize the way in which the DSM categorizes "substance related disorders."
The DSM divides Substance Related Disorders into "two groups: the Substance Use Disorders . . . and the Substance-Induced Disorders." DSM-IV 176; see also DSM-IV-TR 191. As the SSA manual recognizes, Substance Dependence and Substance Abuse are the only two kinds of Substance Use Disorders that the DSM discusses. See DSM-IV 176-82; see also DSM-IV-TR 192-98. While the DSM identifies many different kinds of Substance-Induced Disorders, we need only mention Substance Intoxication, Substance Withdrawal, and Substance-Induced Anxiety Disorder. See DSM-IV 183-94, 439-44; see also DSM-IV-TR 199-212, 479-83. One may further classify Substance-Induced Anxiety Disorder as occurring With Onset During Intoxication, With Onset During Withdrawal, and/or With Panic Attacks. See DSM-IV 191, 440-41; see also DSM-IV-TR 208, 480-81.
In addition to the general information about Substance Related Disorders cited above, the DSM contains specific information about each of eleven different classes of substances, including alcohol. When a clinician can identify the class of substance that a patient is using, she or he "should use the code that applies to the class of substances" and "record the name of the specific substance" in the diagnoses. See DSM-IV 187; see also DSM-IV-TR 204.
The record in this case could conceivably be read as including diagnoses of the following Alcohol Related Disorders: Alcohol Dependence (303.90), Alcohol Abuse (305.00), Alcohol Intoxication (303.00), Alcohol Withdrawal (291.8), and/or Alcohol-Induced Anxiety Disorder (291.8). See DSM-IV 194-204; see also DSM-IV-TR 212-23. Of course, evidence of Alcohol Intoxication, Alcohol Withdrawal, and Alcohol-Induced Anxiety Disorder is not medical evidence of alcoholism, as defined by the manual, because those conditions are not Alcohol Use Disorders. See HALLEX I-5-3-14A(V)(C).
When there is medical evidence of alcoholism, the ALJ must determine whether the claimant's alcoholism is a contributing factor material to his disability (the "material determination"). See HALLEX I-5-3-14A(V)(B)(3). The ALJ may make a finding that alcoholism is material "only when the evidence establishes that the individual would not be disabled if he/she stopped using . . . alcohol." HALLEX I-5-3-14A(V)(D) (emphasis added). Before making any finding, however, the ALJ must identify which of the claimant's "current physical and mental limitations" would remain if he stopped using alcohol (the "remaining limitations"). Id. Only then can the ALJ determine whether those remaining limitations, by themselves, would be disabling.
When considering whether remaining limitations are disabling, the ALJ must again perform the sequential five-step analysis.See Social Security Administration, Questions and Answers Concerning DAA from the 07/02/96 Teleconference, No. EM-96200, at 37 (Aug. 30, 1996), available at http://policy.ssa.gov/public/reference.nsf/0/209ddfe637161f898525 6d5200664c41?OpenDocument [hereinafter QA]. Unlike the first analysis, which considered all of the claimant's impairments (including those that alcohol caused or worsened), the second analysis focuses only on the limitations that the claimant would continue to experience even if he stopped using alcohol. If the ALJ concludes in the second iteration of the five-step analysis that the claimant's remaining limitations (after excluding the effects of alcohol) would not be disabling, then the claimant is not disabled. See 20 C.F.R. § 404.1535(b)(2)(i) (2005). If the remaining limitations "would still be disabling," however, then the ALJ should find that the claimant is disabled regardless of his alcoholism. HALLEX I-5-3-14A(V)(D); see also § 404.1535(b)(2)(ii).
"37" refers to Question 37 and the answer thereto.
The record sometimes will not permit an ALJ to identify which of the claimant's physical and mental limitations would remain if he stopped using alcohol, much less whether the remaining limitations would be disabling. In these cases, SSA policy requires the ALJ to "find that [alcoholism] is not a contributing factor material to the determination of disability." QA 27;see also id. at 29. In other words, the ALJ must find that the claimant is disabled.
To sum up, when a claimant may be an alcoholic, the Commissioner's disability determination must proceed in four discrete stages. First, she must consider all of the claimant's limitations and then use the usual five-step sequential analysis to decide whether the claimant is disabled. Second, the Commissioner must determine whether there is medical evidence of an Alcohol Use Disorder, as defined in the DSM. When such medical evidence exits, the Commissioner must identify which of the claimant's limitations would remain if he stopped using alcohol. Finally, the Commissioner must return to the five-step analysis to evaluate whether the claimant's remaining limitations would be disabling.
C. ALJ's Decision
Having explained the analysis that the ALJ should have undertaken here, we now consider whether his actual decision is supported by substantial evidence. Put another way, we shall review whether he conducted all parts of the necessary analysis and whether substantial evidence supports each of the findings that he made.
1. Initial Five-Step Analysis
At the first step of his initial sequential analysis, the ALJ found that Warren "has not engaged in substantial gainful activity since his alleged onset date." Rec. 18; see also id. at 29, ¶ 1. Warren does not challenge this finding, and, in any event, there is substantial evidence to support it. See id. at 289-90.
The second step required the ALJ to determine whether Warren had a severe medically determinable impairment or combination of impairments. In this regard, the ALJ stated that the "objective medical evidence establishes that the claimant has ongoing alcohol dependence and anxiety, medically determinable impairments that could reasonably be expected to produce the claimant's symptoms." Rec. 21. He also found that "the claimant suffers from a substance abuse disorder and that it is severe."Id. Unfortunately, the ALJ made no formal finding about whether Warren's anxiety, either alone or in combination with his "alcohol dependence," constituted a severe impairment. Despite this ambiguity, the decision, when read as a whole, confirms that the ALJ believed that Warren had a severe medically determinable impairment or combination of impairments. See id. at 29, ¶ 3.
In the third step, the ALJ found that Warren had "severe impairments that meet the criteria of Listings 12.06 [(Anxiety Related Disorders)] and 12.09 [(Substance Addiction Disorders)]." Rec. 26. Though the ALJ stated elsewhere that "the claimant's medically determinable impairments meet Listings 12.04 [(Affective Disorders)] and 12.09," id. at 29, ¶ 4, we believe that the reference to Listing 12.04 is a typographical error because there is not substantial record evidence that Warren suffered from an Affective Disorder. Moreover, only three brief paragraphs after the error, the ALJ again referred to Listing 12.06, not Listing 12.04. See id. at 29, ¶ 7. Regardless of the exact listing(s) to which the ALJ intended to cite, he clearly found that Warren's severe impairments (including the impairments that alcohol use caused and/or worsened) met at least one of the listings. That finding obviated any need to proceed to steps four and five of the initial sequential analysis. If he could ignore Warren's alcohol use, the ALJ would have found him to be disabled.
2. Medical Evidence of Alcohol Use Disorder
After concluding that Warren would have been disabled, the ALJ should have considered whether there was medical evidence of an Alcohol Use Disorder. HALLEX I-5-3-14A(V)(C). The ALJ did not explicitly discuss this issue, but he repeatedly referred to record evidence of Warren's alcohol use and characterized his drinking as "alcohol abuse." See, e.g., Rec. 24. We shall, therefore, treat the ALJ's decision as implicitly finding that there was medical evidence of an Alcohol Use Disorder and review that implicit finding for substantial evidence.
At the outset, we acknowledge that the record is brimming with references to Warren's use of alcohol. He drank five to six cans of beer per day in 1993. Rec. 373. By the summer of 1994, Warren was drinking between twelve and twenty-four cans of beer daily.Id. at 377, 379. When he stopped drinking in July, 1994, so that he could have knee surgery, Warren suffered alcohol withdrawal seizures. On Dr. Binning's recommendation, Warren stopped drinking altogether by mid-September of 1994, see id. at 400, but he resumed drinking two cans of beer each day by the end of that year, see id. at 399. There is no record evidence regarding how much alcohol Warren consumed in 1995, if any.
Before his panic attacks began in the summer of 1996, Warren's alcohol consumption had returned to pre-seizure levels of about twelve to twenty-four cans of beer each day. See Rec. 535-36, 557-58. After noting on March 6, 1997 that Warren was drinking twelve cans of beer per day, see id. at 486, Dr. Nutter encouraged him to reduce, but not eliminate, his alcohol consumption. By April of 1998, Warren's drinking had declined to the equivalent of about six cans of beer per day. See id. at 496-97. The decline was relatively short-lived; Warren was again consuming the equivalent of about eighteen cans of beer daily by July 29, 1998. See id. at 501. At the November 23, 1998 hearing, Warren's wife testified that he was drinking about six cans of beer daily. See id. at 557-58. There is no evidence about the extent of Warren's alcohol usage in 1999.
On August 16, 2000, Warren and his wife testified that he had stopped drinking every day but he still drank more than once a week. See Rec. 589-91. That moderate level of drinking apparently continued at least through April 11, 2001. See id. at 518. The record contains almost no evidence about Warren's alcohol consumption after April, 2001.
Emphasizing repeatedly that Warren drank as much as a case of beer a day, the ALJ's decision overlooks the substantial evidence that the level of Warren's drinking ebbed and flowed over the years. During the relevant period (1996 to 2000), he drank less, often much less, than the case of beer a day that he consumed in early 1994. Moreover, the ALJ ignored that Warren's GGT level had fallen from 498 when he was drinking a case of beer a day to 95 on July 25, 1996, about when his panic attacks began. Compare Rec. 379 with id. at 348. This evidence confirms Warren's testimony that he had reduced his alcohol consumption dramatically, though he continued to drink at lower levels.
More problematic than the gloss that he put on the evidence of Warren's drinking, however, is the ALJ's insistence on relying on that kind of evidence at all. Rather than simply counting the number of beer cans strewn throughout the record, the ALJ should have considered whether there was medical evidence of an Alcohol Use Disorder. HALLEX I-5-3-14A(V)(B)(3), (C)(1)-(3). References to the quantity of beer that Warren drank are evidence about his alcohol consumption, but they cannot substitute for a medical diagnosis of an Alcohol Use Disorder.
At the August 16, 2000 hearing, Dr. Brown criticized Dr. Nutter's diagnosis of panic disorder with agoraphobia and suggested that a diagnosis of substance-induced anxiety disorder might have been more appropriate, but he did not make any definitive diagnosis of his own. See Rec. 607-30. Dr. Brown again recognized the possibility that Warren might have a "substance abuse caused anxiety disorder" at the February 5, 2002 hearing. Id. at 684. Even if we were to treat such indefinite references to Alcohol-Induced Anxiety Disorder as Dr. Brown's firm diagnosis, Dr. Brown's testimony still would fail to constitute substantial evidence of an Alcohol Use Disorder because only Alcohol Dependence and Alcohol Abuse are Alcohol Use Disorders. Alcohol-Induced Anxiety Disorder is an Alcohol-Induced Disorder, not an Alcohol Use Disorder. See supra note 27;see also DSM-IV 195; DSM-IV-TR 212.
Indeed, Dr. Brown ultimately agreed with the ALJ's suggestion that Warren met "the diagnostic criteria for substance induced anxiety disorder." Rec. 684-85; see also DSM-IV 439-44; DSM-IV-TR 479-483.
Although Dr. Brown's testimony is not medical evidence of an Alcohol Use Disorder, there is other medical evidence in the record. For example, Dr. Nutter testified that Warren has an "alcohol abuse disorder." Rec. 659. Dr. Dorozynsky also diagnosed Warren with "alcohol abuse" in addition to "panic disorder with agoraphobia." Id. at 521. Since Alcohol Abuse is an Alcohol Use Disorder, the opinions of Dr. Nutter and Dr. Dorozynsky constitute substantial medical evidence from which the ALJ could have found that Warren suffers from an Alcohol Use Disorder.
3. Remaining Limitations
Since there is substantial evidence that Warren suffers from an Alcohol Use Disorder, the ALJ should have identified the limitations that Warren would continue to experience if he stopped drinking entirely. 20 C.F.R. § 404.1535(b)(2) (2005); HALLEX I-5-3-14A(V)(D). Indeed, the ALJ made the following findings about what Warren's remaining limitations would be:
I find that if the claimant were to stop drinking altogether, he could resume some degree of premorbid functioning.
I find that the claimant would have the following mental health limitations if alcohol abuse is removed from consideration. He would have no more than mild limitations in his activities of daily living. Prior to the worsening of his alcohol abuse and development of a panic disorder, the claimant worked, he got together with friends, he went on vacation every once in a while, and he socialized with friends and family. I find, in giving him the greatest benefit of the doubt, that he would have no more than moderate limitations in social functioning. Although his premorbid functioning does not indicate social limitations, as that has been the most problematic aspect of his combined alcohol and panic disorders and has made him fearful of leaving his house, I find that he may continue to have some limitations in that regard even once alcohol abuse is removed from consideration. I find that he would have no more than mild to moderate limitations in concentration, persistence, and pace and would be able to perform semi-skilled and unskilled tasks. There is no evidence of any extended episodes of decompensation absent substance abuse and I do not find that he would have the degree of agoraphobia that he currently has. I find that he would have some ability to function independently outside the area of his home as he was able to do so prior to 1996.
Rec. 26-27 (emphasis added). Though extensive, these findings all rest on the ALJ's assumption that, if Warren stopped drinking, he would recover almost completely and would be able to function as he could before 1996. There is no record evidence to support that assumption.
While the ALJ accurately described what the record reveals of Warren's pre-1996 capabilities, there is no evidence that he would return to that level of functioning if he stopped consuming alcohol. Even Dr. Brown did not state that his condition would resolve completely. Thus, the evidence of Warren's pre-1996 capacities does not support the ALJ's findings about which of Warren's limitations would persist if he stopped using alcohol.
Put another way, the ALJ appears to have believed that Warren's alcohol use caused his disabling limitations. There is some evidence in the record that alcohol played a role in causing Warren's disabling limitations because Dr. Brown testified that alcohol can intensify anxiety disorders. Still, Dr. Brown stopped short of ruling out that Warren suffered from a panic disorder with agoraphobia, preferring to reserve judgment on that question until Warren stopped using alcohol for six months.
At any rate, the evidence suggesting that alcohol caused Warren's disability is not relevant. The ALJ should have focused on which limitations would remain if Warren stopped using alcohol, not on which limitations might have been caused and/or worsened by alcohol. Though perhaps subtle, the distinction is important because SSA regulations and policy permit an individual to receive disability insurance benefits even if alcohol use caused his disability, so long as he would remain disabled if he stopped using alcohol. Thus, the ALJ's emphasis on cause and effect was misplaced; he should have concentrated on identifying which of Warren's limitations would remain if he stopped drinking.
The ALJ found the reports of Dr. Myers and Dr. Feitz "helpful in trying to determine what the claimant's functioning is without substance abuse since substance abuse did not play much of a role at all in their analysis." Rec. 25. As the ALJ noted, however, Dr. Myers and Dr. Feitz did not review the entire medical record in this case because they made their determinations in 1998, before Dr. Nutter's handwritten notes were included in it and two years before the last date when Warren was insured. Moreover, there is no evidence that Dr. Myers and Dr. Feitz considered what effect Warren's alcohol use had on his limitations or opined what limitations would exist if he stopped using alcohol. Since these reports are based on appraisals of an incomplete record and do not purport to discuss what Warren's remaining limitations would be if he stopped drinking, the ALJ should not have accorded them any weight when he assessed Warren's remaining limitations.
Dr. Brown testified that, if Warren stopped drinking "for the rest of his life, . . . his panic disorder is going to be made better or going to go away," Rec. 611, but he failed to specify how much "better" Warren's panic disorder would get. At another hearing, Dr. Brown again opined that, if Warren "had any remaining anxiety disorder after all of the substance abuse ceased, it would not be terribly impairing and he'd be able to work," without indicating specifically what Warren's limitations would be. Id. at 676. While these statements suggest that Warren's condition would improve if he stopped drinking, they do not address the central issue: what would Warren's limitations be if stopped drinking? Thus, Dr. Brown's testimony does not support the ALJ's findings about Warren's remaining limitations in any way.
Similarly, Dr. Dorozynsky's statement that Warren "would continue to have some degree of impairment" if he stopped using alcohol does not support the ALJ's findings about Warren's remaining limitations because the doctor could not predict "the extent of improvement." Rec. 522.
On October 14, 2002, Dr. Nutter drafted a report that deserves quoting in full:
Mark Warren does meet the criteria for anxiety related disorders as defined by Social Security. He has an irrational fear of being around strangers, in crowded places or traveling by himself out of his house, and he avoids numerous activities, places and situations. He also gets recurrent severe panic attacks. These have significantly resulted in a loss of social, occupational and normal daily living. In addition he can not [sic] function independently outside of his home even with anti-anxiety medication.
It is also my opinion that Mr. Warren would be disabled even if he stopped using alcohol.
Rec. 520. The first sentence of this report makes reference to Listing 12.06, and the remainder of the first paragraph provides the factual basis for Dr. Nutter's conclusion that Warren's impairments met the Listing. See 20 C.F.R. ch. III, pt. 404, subpt. P, app. 1, Listing 12.06(A), (C) (2005).
Still, the October 14, 2002 report is not without some ambiguity. While the first paragraph describes Warren's condition (taking the effects of alcohol into account), the second paragraph states conclusorily that Warren "would be disabled even if he stopped using alcohol." It is possible that Dr. Nutter intended for the brief second paragraph to refer back to the more detailed first paragraph, implying that Warren's impairment would still meet Listing 12.06 even if he stopped using alcohol. On the other hand, the second paragraph does not explicitly refer to the first paragraph, so one could fairly read the report as not expressing any opinion on how limited Warren would be if he stopped drinking. The ALJ did not indicate which of these meanings he ascribed to the October 14, 2002 report.
While this latter reading still suggests that Dr. Nutter believed that Warren would be "disabled" even if he stopped drinking, Dr. Nutter's opinion on the subject of disability — as distinguished from the issue of what limitations Warren would experience if he stopped drinking — is beside the point. Only the Commissioner can decide whether a claimant would be disabled.See 20 C.F.R. § 404.1527(e)(1) (2005).
Assuming that the ALJ could disregard the report, Dr. Nutter testified on November 15, 2000 that Warren would be "as limited as" he was at that time if he stopped drinking alcohol. Rec. 666. Since the ALJ found that Warren's impairments at that time met Listing 12.06, see id. at 26, this testimony is evidence that, even if Warren stopped consuming alcohol, his limitations would still be severe enough to meet Listing 12.06. In other words, when read in conjunction with the ALJ's findings, Dr. Nutter's testimony, if credited, would suggest that Warren's panic disorder was independently disabling.
Even if the ALJ could properly discount Dr. Nutter's opinions, an assumption that we will interrogate shortly, the ALJ's findings about Warren's remaining limitations would still not be supported by substantial evidence. Putting aside Dr. Nutter's October 14, 2002 report and his November 15, 2000 testimony, there is simply no medical evidence about what limitations Warren would have if he stopped using alcohol. Dr. Myers and Dr. Feitz did not consider the issue. Dr. Dorozynsky could not predict what limitations would remain. Dr. Brown suggested that Warren would improve, but he did not explain how much improvement would occur. In short, if one ignores Dr. Nutter's opinions, there is no medical evidence at all about what limitations would persist if Warren stopped drinking.
Despite this total absence of medical evidence, the ALJ made the detailed factual findings about Warren's remaining limitations that we quoted at the beginning of this section. Because no evidence supported those findings, we hold that the ALJ's findings about the extent of Warren's limitations that would persist if he stopped drinking are not supported by substantial evidence. The ALJ based his five-step disability determination on the insufficiently supported factual findings about Warren's remaining limitations, so his ultimate conclusion that Warren's "alcohol abuse is a contributing factor material to his disability" is also unsupported by substantial evidence. Rec. 30, ¶ 16.
D. Remedy
When the Commissioner's decision is not supported by substantial evidence, we may either remand the case for further consideration or award benefits directly. Remand is the more common approach, but we may award benefits "when the administrative record of the case has been fully developed and when substantial evidence on the record as a whole indicates that the claimant is disabled and entitled to benefits." Podedworny v. Harris, 745 F.2d 210, 221-22 (3d Cir. 1984). The administrative record here consists of over seven hundred pages developed over nearly five years, so we have no difficulty finding that it has been fully developed.
We have already explained why substantial evidence does not support the ALJ's conclusions about what limitations Warren would have if he stopped drinking. Still, the question remains whether there is substantial evidence in this record from which any conclusion about Warren's remaining limitations could be drawn. In this regard, we reiterate that Dr. Myers and Dr. Feitz did not consider the issue and Dr. Dorozynsky and Dr. Brown did not specify how much Warren would improve if he stopped drinking. Only Dr. Nutter offered a definite opinion that Warren would remain "as limited" as he was when he did drink. Rec. 666. Rather than seriously analyze Dr. Nutter's opinion, however, the ALJ preferred to manufacture reasons to discount it.
Though Dr. Nutter's October 14, 2002 report could also be read to support that conclusion, the report is too ambiguous to constitute substantial evidence of what Warren's remaining limitations would be.
Though many of those "reasons" were illegitimate, the
The ALJ refused to defer to Dr. Nutter because "he virtually glossed over the claimant's alcohol usage through many of his opinions." Rec. 23. When the ALJ first leveled that criticism in his February 22, 2001 decision, see id. at 93, the Appeals Council warned that the record "evidence either contradicts or is difficult to reconcile with the decision's conclusions about . . . whether [Dr. Nutter] had glossed over or obfuscated the claimant's drinking," id. at 223. Indeed, the Appeals council vacated the February 22, 2001 decision principally because it, like the November 26, 2002 decision, unjustifiably minimized Dr. Nutter's repeated references to Warren's drinking. See id. at 447-48, 482, 486, 495, 796, 497, 501, 510, 515, 516, 518, 520. The ALJ even tried to enlist Dr. Brown to support his pet theory that Dr. Nutter had not thoroughly documented Warren's alcohol use, but Dr. Brown refused to "comment on . . . whether [Dr. Nutter] did or didn't record it enough." See id. at 680. In short, the ALJ could not give less weight to Dr. Nutter's opinions merely because he would have "expect[ed]" Dr. Nutter to document Warren's alcohol use more completely. See id. at 23.
The ALJ also stated that Warren's "drinking history . . . makes portions of Dr. Nutter's reports inconsistent with the Record as a whole." Rec. 25. Dr. Nutter's reports, however, constitute almost all of the relevant medical evidence in this case. To be sure, there are other reports from 1993 and 1994, but they are not directly relevant to Warren's limitations between 1996 and 2000. Dr. Brown and Dr. Dorozynsky also offered medical opinions, but their conclusions were based largely on Dr. Nutter's reports. Since Dr. Nutter's reports are the record, the ALJ could not discount them based on an alleged — and unspecified — inconsistency with other unidentified evidence.
Another reason that the ALJ gave for not relying on Dr. Nutter's opinions was that they were "based in large part on [Warren's] subjective statements," which the ALJ considered not fully credible. Rec. 26. Of course, Dr. Brown's reports are also based on Warren's statements, albeit indirectly, because Dr. Brown only reviewed the record evidence (which contained Warren's statements) and never examined Warren personally. See id. at 673-74. Moreover, to the extent that the ALJ discredits Dr. Nutter for relying on what his patients tell him, he reveals a fundamental distrust of psychiatry generally. See id. at 663 (explaining that, in treating "most psychiatric disorders," doctors must rely on a patient's "description of what he's feeling"). The ALJ should not have dismissed Dr. Nutter's opinions merely because they report things that Warren said.
The ALJ faulted Dr. Nutter for not performing "much in the way of treatment for [the alcohol] phase of an obviously dual diagnosis." Rec. 23. Dr. Nutter, however, repeatedly encouraged Warren to reduce his drinking and recommended that he participate in "some sort of intensive outpatient treatment program." See id. at 664; see also id. at 486, 535, 553, 598. It is true that Warren never stopped drinking completely and that he did not enter any program, but the ALJ should not have used Warren's failure to take Dr. Nutter's advice as a reason to discredit Dr. Nutter (especially when Warren did not enter a program because he could not afford to do so, see id. at 553, 597-99, 664). Neither the ALJ nor any doctor suggested other treatment options that Dr. Nutter failed to explore, so the ALJ, on this record, could not legitimately discredit Dr. Nutter for not doing more.
ALJ did offer one valid reason for declining to credit Dr. Nutter fully. Specifically, the ALJ found Dr. Brown's conclusion (i.e., that "one cannot really state" the degree of limitations) to be more "consistent with the Diagnostic and Statistical Manual of mental disorders for (DSM-IV) [sic]." Rec. 24; see also id. 23 ("Dr. Brown, not Dr. Nutter, has more heavily relied on the current medical literature to support his conclusions."). Dr. Brown did indeed purport to base his opinions on the DSM, and Dr. Nutter provided no explanation for how he reached his conclusions. Because Dr. Brown's opinions were better reasoned, the ALJ's decision to credit them over Dr. Nutter's opinions is supported by substantial evidence.
Although the ALJ could refuse to rely on Dr. Nutter's opinions, he would have had to make findings about Warren's remaining limitations on other record evidence. Dr. Brown testified that abstaining from alcohol would improve Warren's condition, but he never explained how much improvement one could expect. Dr. Dorozynsky also failed to offer any definite opinion. The vagueness of this evidence would have made it impossible for the ALJ to make any finding about precisely which of Warren's limitations would remain if he stopped drinking.
When an ALJ "cannot project what limitations would remain if [a claimant] stopped using . . . alcohol," SSA policy requires the ALJ to "find that [alcoholism] is not a contributing factor material to the determination of disability." QA 27. Thus, even though the ALJ could ignore Dr. Nutter's opinion, the fair-minded application of SSA policy to the remaining record evidence would have required the ALJ to find that Warren's alcoholism was not a contributing factor material to the disability determination. In short, the ALJ would have had to find Warren disabled. Since substantial evidence indicates that Warren is disabled, we shall award disability insurance benefits from October 1, 1996 rather than remand the case to the Commissioner.
Conclusion
Despite the extensive record in this case, there is not substantial evidence to support the ALJ's findings about which of Warren's limitations would remain if he stopped consuming alcohol. Had the ALJ followed SSA policy, he would have had to conclude that Warren was disabled because the record does not contain any evidence specifically describing what Warren's remaining limitations would be (except for Dr. Nutter's opinions, which the ALJ could find inadequately supported). Since we can be sure that the ALJ would have had to find Warren disabled, we shall grant Warren's motion for summary judgment, reverse the Commissioner's November 26, 2002 decision, and award disability insurance benefits from October 1, 1996.
An appropriate Order follows.
ORDER
AND NOW, this 22nd day of June, 2005, upon consideration of the administrative record in this matter, plaintiff's renewed motion for summary judgment (docket entry # 43), defendant's renewed motion for summary judgment (docket entry # 44), and plaintiff's reply, and in accordance with the accompanying Memorandum, it is hereby ORDERED that:1. Plaintiff's renewed motion for summary judgment is GRANTED;
2. Defendant's renewed motion for summary judgment is DENIED;
3. The Commissioner's decision of November 26, 2002 is REVERSED;
4. Plaintiff is AWARDED disability insurance benefits from October 1, 1996; and
5. The Clerk shall CLOSE this civil action statistically.
JUDGMENT
AND NOW, this 22nd day of June, 2005, in accordance with the accompanying Memorandum and Order and Fed.R.Civ.P. 58, it is hereby ORDERED that JUDGMENT IS ENTERED in favor of plaintiff Mark Warren and against defendant Jo Anne B. Barnhart.