Loretta Boiles applied for Supplemental Security Income (SSI) in March 2000, claiming that she was disabled because she suffered from several maladies, including pseudoseizures. Her claim was denied initially, upon reconsideration, and after a hearing before an administrative law judge. The ALJ found that although Boiles had a severe impairment, it did not equal a listed impairment, and that she was capable of working with certain restrictions. Because the ALJ did not adequately support his decision that Boiles's condition was not equal in severity to a listed impairment, we vacate the decision and remand for further factfinding.

At the time of her hearing, Boiles was 34 years old. She has a ninth-grade education and has been employed as a fast food worker/assistant manager, a house cleaner, and a babysitter. None of these jobs lasted more than a few months, and in 1998 she stopped working because she "couldn't keep [a] job because of my mental illness." Two years later, after her disability claim had been denied initially, she worked briefly as a meat packer but left the job because she was "having blackouts and nerve problems."

Boiles's medical history reveals treatment for a number of physical and psychological problems. In the past decade, she has been treated for conditions including pseudoseizures, severe depression, anxiety, post-traumatic stress disorder, heartburn, high blood pressure, back pain, thyroid disease, and migraines. In addition, Boiles was treated for alcoholism in 1990 and for prescription drug abuse as late as 1999. The psychological effects of being sexually abused as a child have been cited frequently as a cause or an exacerbating factor of many of Boiles's health problems. Because Boiles appeals solely on the ground that her pseudoseizures are medically equivalent to a listed impairment, only her history of treatment for that condition is discussed in detail.

Pseudoseizures, also known as psychogenic seizures, nonepileptic seizures, and paraoxysmal nonepileptic episodes (PNES), resemble epileptic seizures but are not attributable to epilepsy or abnormal electric activity in the brain. Ronald P. Lesser, Treatment and Outcome of Psychogenic Nonepileptic Seizures, Epilepsy Currents, Nov. 2003, at 198. No single cause of psychogenic seizures has been identified, but they are typically attributed to an underlying psychological disturbance. Id. Those who have been victims of physical or sexual abuse seem to be at greater risk for developing pseudoseizures. Id. Some symptoms of a pseudoseizure disorder can be treated with medication, but psychological therapy, not medication, appears to be the preferred course of treatment. Id.

According to her testimony and reports to her neurologist, Boiles experienced her first seizure sometime in early 2000, when she arrived at work to find the right side of her car damaged and could not recall how it happened. Not long after, she had a seizure at work and an ambulance had to be called. Boiles also described a seizure that occurred while she was riding in her sister's car; her sister rushed her to the hospital after her eyes rolled back, "foam" came out of her mouth, and she began to bang her head against the car window. After this incident, Boiles was put on Dilantin, an anti-epileptic drug. Then in June 2001, Boiles had another seizure that prompted her boyfriend to call an ambulance. In addition to these daytime incidents, Boiles reported frequent seizures at night.

In the summer of 2001, Boiles sought treatment for her seizures and was referred to a neurologist, Dr. Matthew Wallack. She reported experiencing seizures at night that sometimes woke her up or caused bowel or bladder incontinence. After his initial consultation with Boiles, Dr. Wallack opined that pseudoseizures were a "significant possibility." He prescribed Depakote ER "as a seizure medication." In October 2001 Dr. Wallack saw Boiles again and noted that she reported only four daytime incidents in the prior two months, but that she continued to experience nighttime pseudoseizures, after which she woke up sore and tired. Dr. Wallack prescribed Topamax and noted that if the medication did not alleviate the seizures, he wanted to admit her to an epilepsy monitoring unit. After an appointment in November 2001, Dr. Wallack reported that Boiles had not experienced any daytime seizures since her last appointment, but that she was still having pseudoseizures at night. Dr. Wallack increased the dosage of Topamax but noted that he was "not convinced that these are seizures." Boiles had another appointment with Dr. Wallack the following month, after which he reported that her seizures had become less frequent (about two seizures per week), and typically occurred while she was asleep. In January 2002 Boiles was admitted to the hospital for observation for five days, during which time she had one pseudoseizure. Shortly afterward, Dr. Wallack concluded that she need not be treated with anti-epileptic drugs for her pseudoseizures.

Boiles applied for SSI in March 2000. She claimed that she was disabled due to unspecified "mental disorders," arthritis, depression, "nerve problems," memory loss, asthma, and bronchitis. Her claim was denied, and later that year she sought reconsideration, reporting that her condition had grown worse because she had started having "blackouts/seizures." In February 2001 her claim was denied upon reconsideration, and Boiles requested a hearing before an ALJ.

In February 2002 Boiles had a hearing before an ALJ at which she testified about her condition. Boiles described being disoriented and sore after seizures, and testified that she had been hospitalized "a few times." Her boyfriend also testified, describing the seizure that led him to call an ambulance. He also corroborated Boiles's testimony that she sometimes experienced bladder and bowel incontinence in bed during seizures.

In response to a request from the ALJ, Dr. Wallack provided a written evaluation of her condition and answered interrogatories regarding the frequency of Boiles's pseudoseizures and her prognosis. He stated that Boiles "suffers from pseudoseizures and chronic pain which are secondary to a history of abuse." He noted that although the cause of pseudoseizures cannot be identified, they result in genuine suffering and are often accompanied by "other debilitating symptoms." Dr. Wallack estimated that the seizures occurred twice per week, and characterized her prognosis as "terrible." He stated that Boiles's condition was "untreatable," that it was his "firm belief" that Boiles could not work, and that he "strongly" supported the disability application. He added that the reason for her disability was not the seizures in particular but rather the "underlying cause of the seizures," namely her sexual abuse as a child, which had "devastated her life." Two non-treating physicians consulted by the ALJ testified at the hearing. Dr. Stump, an internist, distinguished pseudoseizures from epilepsy. An epileptic seizure can be diagnosed by an EEG, he explained, but a pseudoseizure is "another form of seizure altogether," and thus a negative EEG does not mean that no seizure took place. He added that patients who experience pseudoseizures cannot be treated with anti-seizure medication and therefore benefit little from going to the hospital during an episode. Dr. Stump also testified that it would be unfair to "penalize" someone suffering from pseudoseizures by not finding her disabled, because like epileptics, "it's very difficult for these people to get jobs." Dr. Stump emphasized that while pseudoseizures are not caused by epilepsy, they are "real."

The consulting psychologist, Dr. Pitcher, agreed with Dr. Stump, noting that the cause of Boiles's pseudoseizures was unknown, but "there is nothing at all to suggest that she is malingering or faking any seizure." In response to a question from the ALJ, Dr. Pitcher testified that alcohol or drug use did not appear to be a material factor in Boiles's "current position," although "there may have been a period of time when it was." Dr. Pitcher also stated that based on Boiles's testimony and the toxicology reports in the record, there did not appear to be any current substance abuse. Both Dr. Pitcher and Dr. Stump testified that pseudoseizures could occur in the absence of drugs or alcohol.

The ALJ then asked a vocational expert (VE) to determine whether there were jobs in the national economy that Boiles could perform. The ALJ asked the VE to consider an individual with Boiles's past work experience who required seizure precautions and limited contact with other people, and the VE determined that there were jobs, such as assembly work, that such a person could perform. Replying to questions from Boiles's attorney, the VE agreed that a person who missed as much work due to illness as Boiles had at her last job would have difficulty maintaining employment.

The ALJ determined that Boiles was not disabled. Following the five-step analysis used to evaluate disability, see 20 C.F.R. § 404.1520(a)-(g), the ALJ found that: (1) Boiles had not performed substantial work since the onset of her medical problems; (2) her pseudoseizures were a severe impairment; (3) her pseudoseizures did not meet or equal a listed impairment; (4) she could not perform her past relevant work and had no transferable skills; and (5) there were jobs available in Indiana that she could perform. Of significance for this appeal, the ALJ found at step three that Boiles's pseudoseizures were not equal in severity to epilepsy as described in Listing 11.02, which applies to major-motor seizures, and Listing 11.03, which applies to minor-motor seizures. When a claimant's condition does not meet the criteria of any of the listed impairments that give rise to a presumption of eligibility for benefits, she may still establish presumptive disability by showing that her impairment is accompanied by symptoms that are "at least equal in severity and duration to the listed findings." 20 C.F.R. § 416.926(a); see Barnett v. Barnhart, 381 F.3d 664, 668 (7th Cir.2004). The ALJ also found that Boiles's description of her physical limitations was not "totally credible" and that she had no exertional limitations, but that she should avoid work at unprotected heights and around dangerous machinery or open bodies of water or flame, and have no more than superficial contact with supervisors, co-workers, and the general public. The Appeals Council declined review, and the ALJ's decision thus became the final decision of the Commissioner of Social Security. The district court then affirmed the decision, and Boiles appeals.

Boiles argues that the ALJ erred in finding that her pseudoseizures were not equal in severity to a listed impairment because he substituted his own judgment for that of the treating and consulting physicians without citing contradictory medical evidence. Listing 11.02 applies to epilepsy that is "documented by detailed description of a typical seizure pattern, including all associated phenomena; occurring more frequently than once a month, in spite of at least 3 months of prescribed treatment," with either daytime episodes or "nocturnal episodes manifesting residuals which interfere significantly with activity during the day." 20 C.F.R. § 220, App. 1. The ALJ decided that Boiles's pseudoseizures were not severe enough to equal Listing 11.02 because there was no EEG evidence of an abnormal brain wave pattern, the frequency of the seizures was "open to question given the lack of emergency room medical care for such episodes," and there was "no evidence" of residual symptoms from nighttime seizures that would interfere with Boiles's activities during the day. Boiles asserts that the ALJ improperly "played doctor" and substituted his own judgment for the medical evidence that her pseudoseizures prevented her from being able to work.

We will uphold the ALJ's decision if it is supported by substantial evidence in the record. Barnett, 381 F.3d at 668. Evidence is "substantial" when it is "sufficient for a reasonable person to accept as adequate to support the decision." Jens v. Barnhart, 347 F.3d 209, 212 (7th Cir.2003) (internal quotations omitted). If the findings of the ALJ are supported by substantial evidence, they are conclusive, and we will not substitute our judgment for that of the ALJ. Id. The ALJ is required to "articulate, at some minimum level, [his] analysis of the evidence." Dixon v. Massanari, 270 F.3d 1171, 1176 (7th Cir.2001). An ALJ may not substitute his own judgment for a physician's opinion without relying on other medical evidence or authority in the record. Clifford v. Apfel, 227 F.3d 863, 870 (7th Cir.2000).

In this case, the testimony of Dr. Wallack and the two non-treating physicians did not support the reasons stated by the ALJ for finding that the pseudoseizures did not equal a listed impairment. First, the ALJ did not explain the relevance of the lack of EEG evidence for his finding that the pseudoseizures were not severe enough to equal a listing. The ALJ did not cite any evidence to contradict Dr. Stump's opinion that a negative EEG: (1) was perfectly consistent with Boiles's type of seizure disorder and (2) did not mean that her seizures were any less "real" than those that could be measured by electric output; thus, it was improper for the ALJ to use the lack of EEG evidence as support for his decision.

The ALJ also improperly relied on the lack of emergency room visits as evidence that Boiles's seizures were not frequent enough to be equal in severity to impairments described in Listing 11.02. Again, the ALJ did not point to anything in the record that contradicted Dr. Stump's testimony that hospitalization was futile for someone suffering from pseudoseizures, which are essentially untreatable.

Nor does the record support the ALJ's third stated reason for concluding that Boiles's condition was not severe enough to equal a listing: that there was "no evidence" that her daytime functioning was impaired due to nocturnal pseudoseizures. In fact, the record contains evidence that Boiles was severely fatigued and in pain after a pseudoseizure. Dr. Wallack told the ALJ that Boiles suffered from "chronic pain" and Dr. Pitcher, the psychologist, testified that Boiles "has the postictal, too," referring to the residual symptoms that follow a seizure or convulsion. Boiles's cousin, who completed a third-party questionnaire, described Boiles as confused, weak, and tired after a pseudoseizure. Boiles testified that after she has a nighttime episode, "I feel beat, I'm real drained for like two or three days, I feel like I just cannot get up." She also stated that she knows she has had a nighttime seizure when "it feels like somebody that's beating me real hard.... I hurt." Boiles reported that she could not drive, take baths by herself, or "stay alone" due to her seizures. The ALJ's conclusion that there was "no evidence" that Boiles was impaired during the day was thus unsupported by the record.

Boiles also argues that the ALJ improperly rejected the opinion of her treating physician, Dr. Wallack, who believed that Boiles was "devastated" by the psychological effects of her child abuse and thus disabled. The ALJ stated that he discounted this opinion because Dr. Wallack had no credentials as a psychologist or psychiatrist, and lacked the longitudinal view of Boiles's medical history that the ALJ possessed. Boiles also asserts that the ALJ did not explain why he discredited Dr. Wallack's opinion that Boiles had two pseudoseizures per week, stating instead that the frequency of the seizures was "open to question."

The opinion of a treating physician concerning a patient's condition is "entitled to controlling weight if it is well supported by medical findings and not inconsistent with other substantial evidence in the record." Clifford, 227 F.3d at 870. Here, the ALJ did not explain how other evidence in the record contradicted Dr. Wallack's opinion about the frequency of Boiles's pseudoseizures. See Gudgel v. Barnhart, 345 F.3d 467, 470 (7th Cir.2003). Aside from the puzzling statement that Dr. Wallack "failed to explain the type of seizure involved," the ALJ did not explain why Dr. Wallack's opinion about the frequency of the seizures was not "well supported by medical findings." Dr. Stump, the non-treating medical expert, estimated that Boiles had seizures at least once per month. Although this more conservative estimate (which the ALJ did not mention) might be viewed as evidence that contradicts Dr. Wallack's opinion, it still supports a level of frequency that is commensurate with Listing 11.02. At the very least, the ALJ was obligated to solicit more evidence if he believed that the frequency of the seizures, as reflected in the record, was unclear. See Smith v. Apfel, 231 F.3d 433, 437-38 (7th Cir.2000).

Finally, Boiles challenges the ALJ's consideration of her past drug abuse, arguing specifically that the ALJ "has taken upon himself to decide that Ms. Boiles was abusing her medications." This concern is somewhat borne out by the record. At the hearing, when Boiles's attorney was questioning Dr. Stump about whether her condition was equivalent to the epilepsy listing, the ALJ interjected, "I'll be honest with you Counsel, with the history of drug and alcohol abuse, I would never grant something on a listing on this ... there's no way I'd do it." And later, in his written ruling, the ALJ suggested that he was "more fully aware of the extent of her prescription drug use," and thus his opinion was more informed than Dr. Wallack's.

It is true that the testifying physicians recognized that drug abuse had impacted Boiles's health in the past, yet neither attributed Boiles's pseudoseizures to drug abuse. Moreover, the results of the drug tests in the record support Boiles's testimony that she was not abusing her medications. The ALJ did not acknowledge the drug tests or the physicians' opinions that substance abuse did not cause Boiles's pseudoseizures, nor did he cite evidence to contradict their opinions, see Clifford, 227 F.3d at 870, and therefore he did not properly support his conclusion that Boiles's history of substance abuse was relevant to his determination.

Because of the shortcomings in the ALJ's order, Boiles urges the court to simply reverse the ALJ and award benefits. But the record does not yet support a finding that Boiles's condition is "at least equal in severity and duration" to epilepsy as described in Listing 11.02. In particular, the ALJ made no finding about the frequency of Boiles's seizures. Whether Boiles's pseudoseizures are of equal medical significance to epilepsy will depend in part upon how frequently they occur; thus the record must be more developed on this point. Alternatively, pseudoseizures may be more analogous to an impairment described in a listing other than 11.02, such as one that describes a psychological impairment.

For the reasons stated above, we VACATE the decision of the district court and REMAND to the Social Security Administration for further proceedings.