Strassburg v. Saul

UNITED STATES DISTRICT COURT EASTERN DISTRICT OF WISCONSIN WADE STRASSBURG, Plaintiff, v. Case No. 19-CV-1010-SCD ANDREW M. SAUL, Commissioner of Social Security, Defendant. DECISION AND ORDER In 2015, Wade Strassburg applied for Social Security benefits, alleging that he is unable to work due to a combination of physical and mental impairments. Following a hearing, an administrative law judge (ALJ) determined, based primarily on testimony from a vocational expert (VE), that Strassburg remained capable of working notwithstanding his impairments. Strassburg now seeks judicial review of that decision. Strassburg argues that the VE’s testimony was unreliable and that the ALJ impermissibly “played doctor” when evaluating his mental-health limitations. The Commissioner contends that the ALJ did not commit an error of law in reaching his decision and that the decision is otherwise supported by substantial evidence. I agree that the ALJ failed to ensure that the VE’s testimony was reliable. Because the ALJ’s step-five finding is not supported by substantial evidence, the decision denying Social Security benefits to Strassburg will be reversed and this matter will be remanded for further proceedings. BACKGROUND On July 28, 2015, Strassburg slipped on a wet floor and injured his knee. R. 279–81.1 A few months later, he applied for disability insurance benefits and supplemental security income from the Social Security Administration (SSA), alleging that he became disabled on the date of the slip-and-fall. R. 362–74. Strassburg asserted that he was unable to work due to a torn meniscus, a sprained ACL and MCL, and torn cartilage in his femoral condyle. R. 409. The medical records submitted in support of his applications showed that he also suffered from mental-health symptoms. See R. 102–04. After his applications were denied at the local level, see R. 98–157, Strassburg appeared before ALJ Brian Steger for an administrative hearing on September 19, 2018, R. 36–97. He was represented by an attorney. R. 38–39. At the time of the hearing, Strassburg was forty-eight years old. He was living in a group home for individuals with drug and alcohol or mental-health issues. R. 44, 69–70. Strassburg testified that he was unable to work due to his physical and mental impairments. R. 44. The ALJ also heard testimony from VE Deana Olah. According to Olah, Strassburg’s past jobs ranged from light to very heavy on the SSA’s exertional scale. R. 77–82. Olah testified that a hypothetical person with Strassburg’s age, education, and work experience could not perform those jobs if he were restricted to sedentary work with additional nonexertional limitations. R. 83–88. However, that person could perform other jobs, including, for example, document preparer (approximately 125,000 jobs in the national economy), final assembler (17,000 jobs), and sorter (19,000 jobs). Id. Olah explained that she estimated the number of jobs available by multiplying the total number of jobs for a particular occupational classification by the percentage of jobs in each classification that had similar skill levels, 1 The transcript is filed on the docket at ECF No. 12-2 to ECF No. 12-52. — 2 — physical demand levels, and job tasks. R. 91–95. Strassburg’s lawyer objected to this explanation, arguing that Olah’s methodology was unreliable. R. 77, 95. The ALJ instructed counsel to submit a post-hearing brief elaborating on his objection. Id. However, counsel never filed a brief in support of his objection. See R. 27. On October 31, 2018, the ALJ issued a decision applying the five-step evaluation process. See 20 C.F.R. §§ 404.1520 (a)(4), 416.920(a)(4). R. 10–35. He determined that Strassburg had not engaged in substantial gainful activity since his alleged onset date (July 28, 2015); Strassburg suffered from eleven “severe” impairments: obesity, lumbar spine disc bulge and degenerative disc disease, internal derangement and osteoarthritis of the right knee statuspost surgery, degenerative changes to the right knee, Hepatitis C, post-traumatic stress disorder, anxiety, attention-deficit hyperactivity disorder, depression, diverticulitis, and history of left thumb reattachment surgery; Strassburg did not suffer from an impairment or combination of impairments that met or medically equaled the severity of a presumptively disabling impairment; Strassburg had the residual functional capacity (RFC) to perform a restricted range of sedentary work; Strassburg was unable to perform any past relevant work; and other jobs existed in significant numbers in the national economy that Strassburg could perform. R. 13–27. Based on those findings, the ALJ concluded that Strassburg was not disabled. R. 28. Strassburg requested review of the ALJ’s decision by the SSA’s Appeals Council. R. 359–61. On May 15, 2019, the Appeals Council denied Strassburg’s request for review, R. 1– 12, making the ALJ’s decision the final decision of the Commissioner of Social Security, see Loveless v. Colvin, 810 F.3d 502, 506 (7th Cir. 2016). — 3 — Strassburg filed this action on July 16, 2019, seeking judicial review of the Commissioner’s decision. ECF No. 1. The matter was assigned to this court in April 2020. All parties have consented to magistrate-judge jurisdiction. See ECF Nos. 22, 23 (citing 28 U.S.C. § 636 (c) and Fed. R. Civ. P. 73(b). The matter is now fully briefed and ready for disposition. See ECF Nos. 15, 19, 21. APPLICABLE LEGAL STANDARDS “Judicial review of Administration decisions under the Social Security Act is governed by 42 U.S.C. § 405 (g).” Allord v. Astrue, 631 F.3d 411, 415 (7th Cir. 2011) (citing Jones v. Astrue, 623 F.3d 1155, 1160 (7th Cir. 2010)). Pursuant to sentence four of § 405(g), federal courts have the power to affirm, reverse, or modify the Commissioner’s decision, with or without remanding the matter for a rehearing. Section 205(g) of the Act limits the scope of judicial review of the Commissioner’s final decision. See § 405(g). As such, the Commissioner’s findings of fact shall be conclusive if they are supported by “substantial evidence.” See § 405(g). Substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Moore v. Colvin, 743 F.3d 1118 , 1120–21 (7th Cir. 2014) (quoting Richardson v. Perales, 402 U.S. 389, 401 (1971)) (other citations omitted). The ALJ’s decision must be affirmed if it is supported by substantial evidence, “even if an alternative position is also supported by substantial evidence.” Scheck v. Barnhart, 357 F.3d 697, 699 (7th Cir. 2004) (citing Arkansas v. Oklahoma, 503 U.S. 91, 113 (1992)). Conversely, the ALJ’s decision must be reversed “[i]f the evidence does not support the conclusion,” Beardsley v. Colvin, 758 F.3d 834, 837 (7th Cir. 2014) (citing Blakes v. Barnhart, 331 F.3d 565, 569 (7th Cir. 2003)), and reviewing courts must remand “[a] decision that lacks — 4 — adequate discussion of the issues,” Moore, 743 F.3d at 1121 (citations omitted). Reversal also is warranted “if the ALJ committed an error of law or if the ALJ based the decision on serious factual mistakes or omissions,” regardless of whether the decision is otherwise supported by substantial evidence. Beardsley, 758 F.3d at 837 (citations omitted). An ALJ commits an error of law if his decision “fails to comply with the Commissioner’s regulations and rulings.” Brown v. Barnhart, 298 F. Supp. 2d 773, 779 (E.D. Wis. 2004) (citing Prince v. Sullivan, 933 F.2d 598, 602 (7th Cir. 1991)). Reversal is not required, however, if the error is harmless. See, e.g., Farrell v. Astrue, 692 F.3d 767, 773 (7th Cir. 2012); see also Keys v. Barnhart, 347 F.3d 990 , 994– 95 (7th Cir. 2003) (citations omitted). In reviewing the record, this court “may not re-weigh the evidence or substitute its judgment for that of the ALJ.” Skarbek v. Barnhart, 390 F.3d 500, 503 (7th Cir. 2004) (citing Lopez ex rel. Lopez v. Barnhart, 336 F.3d 535, 539 (7th Cir. 2003)). Rather, reviewing courts must determine whether the ALJ built an “accurate and logical bridge between the evidence and the result to afford the claimant meaningful judicial review of the administrative findings.” Beardsley, 758 F.3d at 837 (citing Blakes, 331 F.3d at 569 ; Zurawski v. Halter, 245 F.3d 881, 887 (7th Cir. 2001)). Judicial review is limited to the rationales offered by the ALJ. See Steele v. Barnhart, 290 F.3d 936, 941 (7th Cir. 2002) (citing SEC v. Chenery Corp., 318 U.S. 80 , 93–95 (1943); Johnson v. Apfel, 189 F.3d 561, 564 (7th Cir. 1999); Sarchet v. Chater, 78 F.3d 305, 307 (7th Cir. 1996)). ANALYSIS Strassburg contends the ALJ failed to lay the foundation for the VE’s job-number estimates and impermissibly evaluated his mental-health limitations. — 5 — I. Reliability of the VE’s job-number estimates At step five of the sequential evaluation process, the Commissioner had the burden of demonstrating the existence of significant number of jobs in the national economy that Strassburg could perform. See 20 C.F.R. §§ 404.1560 (c)(1), 416.960(c)(1). To obtain a job- number estimate at Strassburg’s hearing, the ALJ followed the common path of seeking the assistance of a vocational expert. After classifying Strassburg’s past work, the VE testified that a person with Strassburg’s age, education, work experience, and RFC could still work as a document preparer, a final assembler, and a sorter. R. 87. According to the VE, those jobs existed in significant numbers in the national economy. She estimated that across the country there were 125,000 document preparer jobs; 17,000 final assembler jobs; and 19,0002 sorter jobs. Id. On cross-examination, Strassburg’s counsel asked the VE how she had arrived at those figures. R. 91. The VE explained that the job numbers were representative of jobs that were similar in SVP level,3 jobs tasks, and physical demand: So all the jobs that I give, if they’re under a particular DOT code, those jobs all have the same -- they’re all in the same OES group. So when we’re looking for these jobs and giving these DOT titles, again, we look at all similar job codes in the OES, and we pull those out with the same SVP level, physical demand level, and job tasks. Id. The VE indicated that the job numbers came from a variety of government resources, 2 The transcript actually says 919,000 sorter jobs, but this seems to be a mistake. During follow-up questioning the ALJ clarified that the VE said 19,000, see R. 91, and that’s the figure the ALJ relied upon in making his stepfive finding, see R. 27. This also was the number cited by Strassburg in his brief to the Appeals Council. See R. 539. 3 SVP stands for Specific Vocational Preparation; it “is defined as the amount of lapsed time required by a typical worker to learn the techniques, acquire the information, and develop the facility needed for average performance in a specific job-worker situation.” See Dictionary of Occupational Titles (DOT), App’x C, § II, available at https://occupationalinfo.org/appendxc_1.html (last visited May 14, 2020). SVP levels range from 1 (short demonstration only) to 9 (over 10 years). — 6 — including the Standard Occupational Classification (SOC), the Occupational Information Network (ONET), and data from SkillTRAN’s Occupational Employment Survey (OES). Id. The ALJ then asked how the VE arrived at her numbers in this case. Id. She responded, So, like I said, what we do is we take the total number of jobs for a particular SOC code that we find in SkillTRAN. We verify that code through ONET. Then what we do is we multiply the percentage of jobs from the OES group that have the same SVP level and physical demand level, and we times the number of persons employed in this job to get the estimated number of jobs in the national economy. R. 91–92. After repeating the steps for Strassburg’s counsel, see R. 92–93, the VE provided an example for a mail clerk position, DOT code 209.687-026. See R. 94–95. That DOT code is found within SOC code 43-9051. SOC code 43-9051 contains 14 different DOT job titles. Of those 14 job titles, 6 jobs had similar SVP levels, physical demand levels, and job tasks. Thus, to estimate the number of mail clerk jobs available, the VE would multiply the total number of jobs for SOC code 43-9051 by 43% (i.e., 6 ÷ 14 (rounded up to the nearest %)). Lastly, the VE would go back to ONET to verify that those jobs still existed in the national economy. Strassburg’s counsel objected, arguing that the VE’s method sounded “awfully like the equal distribution method” the Seventh Circuit found unreliable in Chavez v. Berryhill, 895 F.3d 962 (7th Cir. 2018). R. 95. The ALJ took counsel’s objection under advisement. He instructed counsel to submit a post-hearing brief regarding his objection and stated that he would address the objection in his decision. Id. In his written decision, the ALJ adopted the job-number estimates provided by the VE at the hearing. See R. 26–27. The ALJ noted that Strassburg’s counsel did not file a brief in support of his objection. Nevertheless, the ALJ determined that the objection was without merit. He explained, — 7 — The vocational expert provided detailed testimony relative to her methodology and resources. She also explained that the job numbers she provided were estimates and were representative of other jobs with similar SVP levels and job tasks found using the described methodology along with her education and her experience as a vocational expert. R. 27. The ALJ further noted that, even if the job-number estimates were exaggerated by hundreds, “the actual remaining totals would still constitute significant numbers for the vocational analysis set forth above.” Id. On appeal, Strassburg argues the ALJ failed to ensure that the VE used a reliable method to estimate the number of jobs available in the national economy. See ECF No. 15 at 12–19; ECF No. 21 at 4–9. He maintains that the numbers provided did not “match up” with the VE’s proposed methodology and that the ALJ did not adequately resolve his objection. The Commissioner asserts that Strassburg forfeited his objection concerning the VE’s methodology by not complying with the ALJ’s directive to file a post-hearing brief. See ECF No. 19 at 5–7. “In the context of job-number estimates, . . . the substantial evidence standard requires the ALJ to ensure that the approximation is the product of a reliable method.” Chavez, 895 F.3d at 968 (citing Donahue v. Barnhart, 279 F.3d 441, 446 (7th Cir. 2002); McKinnie v. Barnhart, 368 F.3d 907, 910 (7th Cir. 2004)). Thus, “[a] finding based on unreliable VE testimony is equivalent to a finding that is not supported by substantial evidence and must be vacated.” Chavez, 895 F.3d at 968 (quoting Britton v. Astrue, 521 F.3d 799, 803 (7th Cir. 2008)). “If the basis of the vocational expert’s conclusions is questioned at the hearing, . . . then the ALJ should make an inquiry . . . to find out whether the purported expert’s conclusions are reliable.” Donahue, 279 F.3d at 446 . “[A]ny method that the agency uses to estimate job — 8 — numbers must be supported with evidence sufficient to provide some modicum of confidence in its reliability.” Chavez, 895 F.3d at 969 . I am not persuaded by the Commissioner’s forfeiture argument. At the hearing, Strassburg’s counsel cross-examined the VE about her methodology for estimating job numbers and, citing Chavez, explicitly objected to the reliability of her conclusions. These actions were sufficient to trigger the ALJ’s duty to ensure that the VE’s estimates were the product of a reliable method. See Donahue, 279 F.3d at 446 ; see also Courtney v. Berryhill, 385 F. Supp. 3d 761 , 763–64 (W.D. Wis. 2018) (dismissing Commissioner’s argument that Plaintiff waived challenge to VE’s testimony by not objecting at the hearing because “the claimant does not need to make a formal objection. . . . He needs only to cross-examine the expert and elicit statements that call into question the reliability of his conclusions.”). Though a posthearing brief may have been helpful, counsel minimally explained the basis for his objection at the hearing and provided the VE ample opportunity to defend her methodology. Finding forfeiture under these circumstances would effectively and impermissibly shift the burden to Strassburg at step five. Moreover, while the ALJ’s decision noted counsel’s failure to file a brief in support of his objection, ultimately the ALJ addressed the merits of the objection. So will I.4 Here, the VE estimated the job numbers by using a variation of the “equal distribution method,” which has been repeatedly criticized by the Seventh Circuit. See Chavez, 895 F.3d at 969 (“We have seen the method applied in other Social Security cases and—for the exact 4Although the Commissioner provides a few sentences of argument supporting the merits of the ALJ’s reliance on the VE, the briefing on this issue is focused almost exclusively on the issue of forfeiture. Arguably, therefore, the Commissioner has itself forfeited any argument supporting the methods used by the VE here. “We will not assume or supply the Commissioner's positions or arguments.” L.B.M. ex rel Motley v. Astrue, No. 1:08-CV-1354-WTL-DML, 2010 WL 1190326 , at *16 (S.D. Ind. Mar. 23, 2010). — 9 — reason its application here troubles us—have questioned its use in at least four opinions.”). The VE testified that her estimates were not specific to the DOT codes provided; rather, they were “also representative of other jobs.” R. 91. For this method, she multiplied the total number of jobs available in an occupation group by the percentage of job titles in that group that had a similar SVP level, physical demand level, and job tasks as the DOT title provided. Like the equal distribution method, this method assumes without an empirical foundation that all job titles within a group exist in equal numbers. The only difference is that the ALJ here also reduced the broader group to similar types of jobs. However, among those jobs, the numbers are assumed to be equally distributed. An example may help illustrate this point. The VE testified that the hypothetical person could work as a sorter, with DOT code 521.687-086. R. 87. The specific DOT job title for that code is “nut sorter,” which falls within SOC code 51-9061 (“Inspectors, Testers, Sorters, Samplers, and Weighers”). See ECF Nos. 15-1 at 5, 15-2 at 2. Of the hundreds of jobs listed within that code, 13 have the same SVP level (2) and strength level (sedentary) as a nut sorter: “cigarette-making-machine catcher,” “weight tester,” “dowel inspector,” “gauger,” “lens-block gauger,” “film touch-up inspector,” “touch-up screener, printer circuit board assembly,” “button reclaimer,” “zipper trimmer, hand,” “check weigher,” “shadowgraph-scale operator,” “table worker,” and “type-copy examiner.” See ECF No. 15-1 at 5–6. It seems highly unlikely that these jobs exist in equal numbers in the national economy, or even that there is even any kind of approximate equivalence. Chavez, 895 F.3d at 969 (“What most — 10 — concerns us is that the method rests on an assumption about the relative distribution of jobs within a broader grouping that lacks any empirical footing.”) The Commissioner stresses that in Chavez the Seventh Circuit explicitly refused to “enjoin use of the equal distribution method.” ECF No. 19 at 7 (quoting Chavez, 895 F.3d at 970 ). That is true. But the court also explained that, when confronted by a claimant’s challenge to a VE’s job-number estimate, the ALJ “must require the VE to offer a reasoned and principled explanation” before accepting that estimate. Chavez, 895 F.3d at 970 . For example, the VE could draw on her “knowledge of labor or market conditions and occupational trends, gleaned from reviewing data sources or placing workers in jobs.” Id. Despite questioning from counsel and the ALJ, the VE here offered no such explanation. The ALJ therefore did not have a reasoned and principled basis for accepting the job-number estimates. See Chavez, 895 F.3d at 969 (“The ALJ needed to do more than just ask questions; she needed to hold the VE to account for the reliability of his job-number estimates.”). The Commissioner also argues that the ALJ sufficiently resolved Strassburg’s objection in his written decision. See ECF No. 19 at 7–8. I disagree. First, while the VE described the sources and mathematical formula she used to arrive at the numbers provided, she never explained why she was confident that her methodology produced sufficiently reliable jobnumber estimates. See Chavez, 895 F.3d at 969 (“What is entirely lacking is any testimony from the VE explaining why he had a reasonable degree of confidence in his estimates.”).5 Second, while the VE testified that she relied on her education and experience to resolve conflicts with the DOT, she never cited or explained how these attributes informed her job-estimate 5 The VE also made it difficult to “test” her formula, as her example involved a DOT job title (mail clerk) that didn’t apply to this case. — 11 — testimony. See R. 76–96. Third, the Commissioner points out that the ALJ noted that even if the final totals were significantly wrong, “the actual remaining totals would still constitute significant numbers for the vocational analysis.” R. 28. In other words, because the VE identified 161,000 jobs (a large number), even a significant methodological error would still leave us with plenty of available jobs. This seems to be the same kind of “fudge factor” argument rejected by the court in Chavez, however: We recognize that the VE identified three suitable jobs for Chavez and then estimated that, in total, nearly 500,000 of those jobs existed in today’s economy. The observation leads nowhere, however, as each of the VE’s job estimates was the product of the equal distribution method, and nothing in the administrative record allows us to conclude with any reliability that the estimates reasonably approximate the number of suitable jobs that exist for Chavez. The substantial evidence standard does not permit the shortcut, and too much is at stake for Chavez for us to take it. Her case must go back. 895 F.3d at 970 . Because the ALJ failed to ensure that the VE’s job-number estimates were the product of a reliable method, his step-five finding is not based on substantial evidence. “[T]hough the VE explained the mathematical operations he performed to arrive at the numbers he provided, he failed to explain why he had any amount of confidence that these mathematical operations yielded sufficiently accurate job-number estimates.” Baird v. Berryhill, 2018 WL 4874757 , at *4 (N.D. Ind. Oct. 9, 2018). II. Mental-health limitations Strassburg also argues that the ALJ’s RFC assessment is not supported by substantial evidence with respect to his mental-health limitations. See ECF No. 15 at 20–23; ECF No. 21 at 9–10. In assessing Strassburg’s mental RFC, the ALJ gave little weight to the opinions of the state agency psychological consultants. R. 24 (citing Exhibits 3A; 4A; 5A; 8A). The ALJ explained that, “[w]hile Strassburg’s part-time work demonstrates an ability to perform simple — 12 — tasks and work activities, a combination of his symptoms caused by post-traumatic stress disorder, anxiety, depression, and attention-deficit hyperactivity disorder results in additional social limitations.” R. 24. The ALJ included specific mental limitations in his RFC assessment that went beyond the state consultants’ opinions. See R. 19. Strassburg maintains that the ALJ impermissibly translated his worsening mentalhealth symptoms into specific limitations and that, in his view, the ALJ should have sought an updated medical opinion. I disagree. Imposing limitations beyond any medical opinion is not per se unreasonable. Indeed, assessing a claimant’s RFC is the province of the ALJ; it does not need to be based on any medical opinion. See 20 C.F.R. §§ 404.1527 (d)(2), 416.927(d)(2); see also Schmidt v. Astrue, 496 F.3d 833, 845 (7th Cir. 2007). Strassburg also has not demonstrated that the purportedly new mental-health evidence was too technical for the ALJ to consider without the benefit of a medical expert. The ALJ considered Strassburg’s mentalhealth diagnoses and acknowledged that he suffered from moderate mental limitations. Finally, despite claiming that his symptoms worsened, Strassburg has not identified any limitations that the ALJ omitted and should have included. He has therefore failed to show that the ALJ committed reversible error in assessing his mental-health limitations. CONCLUSION For all the foregoing reasons, I find that the ALJ failed to ensure that the VE’s jobnumber estimates were reliable. Based on this record, however, I cannot determine whether Strassburg was disabled as of July 28, 2015. Accordingly, it is necessary to remand this matter to the Commissioner for a new step-five hearing. — 13 — The Commissioner’s decision is REVERSED, and this action is REMANDED pursuant to sentence four of section 205(g) of the Social Security Act, 42 U.S.C. § 405 (g), for further proceedings consistent with this decision. The Clerk of Court shall enter judgment accordingly. SO ORDERED this 27th day of May, 2020. ye C. DRIES United States Magistrate Judge — 14 —