Dixon v. Sullivan, 792 F. Supp. 942 (S.D.N.Y. 1992)
May 8, 1992
David DIXON, et al., Plaintiffs,
the State of New York, Plaintiff-Intervenor,
Louis SULLIVAN, as Secretary of Health and Human Services, Defendant.
United States District Court, S.D. New York.
*943 Scott A. Rosenberg, Asst. Director of Litigation, Civil Appeals & Law Reform Unit, The Legal Aid Soc. (Matthew Diller, of counsel), Kathleen A. Masters, Attorney-in-Charge, Robert Belfort, Supervising Atty., Park Place Trial Office (Susan R. Sternberg, of counsel), New York City, Martin Price, Attorney-in-Charge, Brooklyn Office for the Aging, The Legal Aid Soc. (Barbara Quackenbos, of counsel), Brooklyn, N.Y., for plaintiff Class.
*944 Robert Abrams, Atty. Gen. of the State of N.Y., Mary Fisher Bernet, Asst. Atty. Gen., New York City, for plaintiffs-intervenors, Cesar Perales and the State of N.Y.
Otto G. Obermaier, U.S. Atty., S.D.N.Y., Sapna V. Raj, Sp. Asst. U.S. Atty., New York City (Donald Gonya, Chief Counsel for Social Sec., George Lowe, Deputy Associate Gen. Counsel, Dean Landis, Office of the Gen. Counsel, Dept. of Health and Human Services, Social Sec. Div., of counsel), for defendant.
OPINION AND ORDER
WILLIAM C. CONNER, District Judge:
This class action was originally brought before Judge Lasker by applicants for and recipients of benefits based on disability under the federal Supplemental Security Income and Old Age Survivor's and Disability Insurance ("OASDI") programs, administered by the Secretary of Health and Human Services (the "Secretary" or "Defendant") pursuant to the Social Security Act, 42 U.S.C. § 401, et seq., to challenge the Secretary's policy of denying benefits to claimants whose impairments are "not severe" without inquiry into the effect of claimants' ages, educations, and work histories on their ability to work. Defendant contended that the severity regulation is used only to screen out de minimis claims. Plaintiffs replied that the regulation is systematically applied to deny meritorious claims. Plaintiffs also challenged the Secretary's refusal to consider the cumulative effect of different impairments in making severity determinations.
Prior to 1978, the Secretary's regulations provided that a claimant could be denied benefits on medical considerations alone "where the only impairment is a slight neurosis, slight impairment of sight or hearing, or similar abnormality or combination of slight abnormalities." 20 C.F.R. § 404.1502(a) (1968). This "slight impairment" policy was superseded by the "severity regulation," which is step two of a five-step sequential evaluation procedure adopted by the Secretary in 1978, and clarified in 1980. See 20 C.F.R. § 404.1520. In 1982, the Secretary issued Social Security Ruling ("SSR") 82-55 instructing Social Security administrators not to consider the combined effects of impairments which do not individually meet the Secretary's severity standard. SSR 82-55 also listed 20 specific impairments which the Secretary determined are non-severe per se. This ruling bound administrative law judges and the Appeals Council, and was made effective retroactively to August 20, 1980. After Congress passed the Disability Reform Act of 1984, the Secretary invalidated his practice of refusing to consider impairments in combination. See SSR-III-II. Subsequently, the Secretary published SSR 85-28, which clarifies the policy for determining when a person's impairment may be found not severe. Among other things, SSR 85-28 rescinds the list of 20 specific non-severe examples published in SSR 82-55.
By order dated July 26, 1984, Judge Lasker conditionally certified a statewide class in this action consisting of all disability claimants whose benefits had been, or would in the future be, denied or terminated pursuant to the severity regulations or Social Security Ruling 82-55, after July 20, 1983. The Court also granted a preliminary injunction, prohibiting the Secretary from denying or terminating disability benefits *945 on the basis of the severity regulations. Dixon v. Heckler, 589 F. Supp. 1494, 1512 (S.D.N.Y.1984). In addition, the injunction directed the Secretary to identify and notify class members and to reconsider their claims.
On appeal, the Second Circuit affirmed Judge Lasker's Order granting a preliminary injunction. Dixon v. Heckler, 785 F.2d 1102 (2d Cir. 1986). The Supreme Court subsequently granted certiorari, vacated the judgment of the Court of Appeals and remanded the case for further consideration in light of its decision in Bowen v. Yuckert, 482 U.S. 137, 107 S. Ct. 2287, 96 L. Ed. 2d 119 (1987), upholding the facial validity of the severity regulation and its requirement that disability claimants make a threshold showing of impairment, based on medical criteria alone, before vocational factors would be considered. Thereafter, the Second Circuit ordered that the preliminary injunction issued by Judge Lasker be vacated, and remanded the case for further proceedings in light of Yuckert. See Dixon v. Heckler, 827 F.2d 765 (2d Cir.1987). On November 5, 1987, this Court denied plaintiffs' motion to reinstate the preliminary injunction. Dixon v. Bowen, 673 F. Supp. 123 (S.D.N.Y.1987).
Before this case was transferred to this Court, Judge Lasker ruled, in a letter dated September 17, 1987, that plaintiffs had "made a sufficient threshold showing under Bowen v. City of New York, 476 U.S. 467, 106 S. Ct. 2022, 90 L. Ed. 2d 462 (1986), to justify their entitlement to discovery as to whether a clandestine policy of the Secretary existed which could support a finding that the statute of limitation should be tolled."
In December 1989, the Secretary moved for summary judgment. Because plaintiffs had not yet completed discovery, the Court deferred consideration of the Secretary's motion in order to allow plaintiffs to take a number of depositions in this action. See Dixon v. Bowen, 126 F.R.D. 483 (S.D.N.Y. 1989). The Court further noted that "[i]f plaintiffs show that the Secretary applied the severity test in a manner stricter than that set forth in his published regulations, which express a policy of using step two to screen out only de minimis cases, plaintiffs will have met the requirements of City of New York." Id. at 488.
After the completion of discovery, the parties agreed to resolve the remaining matters in this action by trial on a stipulated record. Toward this end, the parties have entered into a stipulation as to the documents and deposition testimony that constitutes the trial record in this case. The parties have also agreed to a number of facts stipulated to be true. Plaintiffs now seek judgment on the following claims: (1) That POMS DI § 2102 and SSR 82-55 violate the severity regulation and the Social Security Act because they do not provide for an individualized functional assessment of severity; (2) that these provisions violate the Second Circuit's requirement that allegations of pain must be fully considered; (3) that POMS DI § 2102 and SSR 82-55 were systematically misapplied by adjudicators; (4) that the severity standard was misapplied even before issuance of POMS DI § 2102; (5) that the Secretary's refusal to consider the combined effects of not severe impairments violated the Social Security Act; and (6) that the SSA's issuance of POMS DI § 2102 and SSR 82-55 violated the notice and comment provisions of the Administrative Procedure Act. Plaintiffs further contend that this Court should toll the statute of limitations as to their claims because of various covert practices on the part of the Secretary and that this Court should grant permanent injunctive relief and should order reopening and readjudication of claims denied as not severe between June 1976 and July 19, 1983.
A. Bowen v. Yuckert
In Bowen v. Yuckert, 482 U.S. 137, 107 S. Ct. 2287, 96 L. Ed. 2d 119 (1987), the Supreme Court upheld the severity regulation as valid on its face. The Court, in an opinion joined by six Justices, held that a threshold showing of a severe medically determinable impairment required under the severity regulation is consistent with *946 the language and legislative history of the Social Security Act. Noting the "exceptionally broad authority" of the Secretary to prescribe regulations with respect to proving disability under the Act, the Court held that 20 C.F.R. § 404.1520(c) was consistent with the definition of disability provided in section 223(d) (1) (A) of the Act, 42 U.S.C. § 423(d) (1) (A). Specifically, the Court observed that the Act defined "disability" as an "inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment...." Yuckert, 482 U.S. at 146, 107 S. Ct. at 2293 (citing 42 U.S.C. § 423(d) (1) (A)). The Supreme Court noted further that the regulation adopts precisely this type of functional approach to disability. Under the regulation, as under the Act, "if the impairments are not severe enough to limit significantly the claimant's ability to perform most jobs, by definition the impairment does not prevent the claimant from engaging in any substantial gainful activity." Id.
In a footnote, the Court further stated that "[t]he severity regulation plainly adopts a standard for determining the threshold level of severity: the impairment must be one that `significantly limits your physical or mental ability to do basic work activities'" Id. at 153 n. 11, 107 S. Ct. at 2297 n. 11 (citing 20 C.F.R. § 404.1520(c)). Thus, the Court remarked that the Act expressly places on the claimant the burden of showing a medically determinable impairment on the claimant and that the Secretary has no need to consider the claimant's age, education, and work experience if the claimant is unable to meet this threshold burden of establishing medical severity. Id. at 148, 107 S. Ct. at 2294.
Justice O'Connor, in a concurrence joined by Justice Stevens, noted that "respondent's evidence suggests that step two has been applied systematically in a manner inconsistent with the statute." Id. at 157, 107 S. Ct. at 2299 (O'Connor, J. concurring). Justice O'Connor further stated that, in her view, "step two may not be used to disqualify those who meet the statutory definition of disability." Id. at 158, 107 S. Ct. at 2300 (O'Connor, J. concurring). "Only those claimants with slight abnormalities that do not significantly limit any `basic work activity' can be denied benefits without undertaking this vocational analysis." Id. (O'Connor, J. concurring).
Because the Court of Appeals for the Ninth Circuit had not considered the issue, the Supreme Court did not address Yuckert's contention that the regulation was invalid as applied.
B. Facial Validity of SSR 82-55 and POMS DI § 2102
From April 1981 through February 1985, the Secretary required his adjudicators to apply two administrative directives implementing the severity regulation, Program Operations Manual DI § 2102 (the "POMS") and SSR 82-55 (the "SSR"). Plaintiffs assert that on their face, these directives conflict with the Social Security Act ("the Act") and the severity regulation. Plaintiffs argue that while the Act and the severity regulation call for an individualized functional assessment of the claimant's ability to perform work activities, the Secretary's internal policies dictated that adjudicators base determinations solely on the list of 20 profiles. Plaintiffs contend further that these policies precluded adjudicators from taking into account the fact that impairments may affect people differently and that assessment of the effects of impairments cannot be reduced to a simple formula.
Plaintiffs' argument is flawed. In their discussion of the facial validity of the SSR and the POMS, plaintiffs fail to address directly the significance of Yuckert to their facial invalidity claims. Yuckert permits the Secretary to require disability claimants to make a threshold showing of medical severity without consideration of non-medical factors. Plaintiffs' claim that the SSR and the POMS are facially invalid because they prevent adjudicators from making individualized functional assessments with consideration given to all the material evidence is a thinly disguised attempt to relitigate essentially the same question decided in Yuckert: whether medical severity may be determined without consideration *947 of non-medical factors. Yuckert makes clear that at step two, the type of all-inclusive consideration of the material evidence which plaintiffs seek is not mandated.
Plaintiffs attempt to salvage their argument by portraying the 20 examples as a "negative listing" or as per se adjudicative rules of the sort rejected by the Supreme Court in Sullivan v. Zebley, 493 U.S. 521, 110 S. Ct. 885, 890, 107 L. Ed. 2d 967 (1990). However, Zebley has no application here. Zebley was not concerned with de minimis screening at step two, but focused on step three, which, as applied to child claimants, allowed a claimant to be found disabled only if medical evidence of his impairment matched or was equal to one of a listing of impairments. See id., 110 S. Ct. at 890. Not only is step three not implicated in the present action, but also the list of 20 examples at issue here performs a function that is precisely the opposite of that performed by the step three lists. While at step three claimants had to match or equal one of the listed impairments in order to receive benefits, claimants at step two had to show that their disorder was not one of the twenty representative disorders defined as less than severe. Thus the step two list at issue here was not intended as an exclusive set of examples of impairments entitling a claimant to recovery but merely as a representative list of disorders which should be considered not severe. Plaintiffs' argument refuses to acknowledge the plain language of the SSR, which states:
[W]e have tried to identify medical conditions which demonstrate nonsevere impairments.... These 20 examples are intended to illustrate a level of impairment severity where the evaluation principle for nonsevere impairments applies. In addition, these 20 examples, while not exhaustive, represent a significant number of specific impairments that serve as guidelines of the level of impairment severity which would warrant a finding that an individual's impairment is not severe.
The 20 examples were not intended to be used as any type of "negative listing." Rather, they were an attempt by the Secretary to illustrate the type of conditions that do not constitute not severe impairments. Considered on its face, such a list is certainly consistent with the regulation's requirement of a threshold showing of medical impairment and is wholly dissimilar to step three lists which purport to define and limit the universe of compensable disabilities.
C. The Requirement that Allegations of Pain be Fully Considered
Plaintiffs note that the Second Circuit has long required that once an impairment that can cause pain has been objectively determined to exist, SSA cannot limit its inquiry to consideration of objective evidence of pain. See Gallagher v. Schweiker, 697 F.2d 82 (2d Cir.1983). Plaintiffs aver that the POMS and the SSR violated the requirement that subjective as well as objective evidence be taken into account in assessing allegations of pain because, plaintiffs believe, these policy statements required the denial of claims based solely on clinical findings, without regard to testimony and other evidence concerning the existence and extent of pain in a particular case.
Plaintiffs' claim focuses primarily on language found in the SSR to the effect that "[i]n formulating these examples, the potential for severe and prolonged pain has been considered. These conditions are not expected to produce symptoms of severe and prolonged pain."
*948 This instruction, according to plaintiffs, must be viewed in conjunction with SSA's regulation concerning the assessment of pain. The regulation states, in part, that "[w]e will never find that you are disabled based on your symptoms, including pain, unless medical signs or findings show that there is a medical condition that could reasonably be expected to produce these symptoms." 20 C.F.R. § 404.1529. Thus, plaintiffs contend that the SSR's statement that the examples were "not expected" to produce "severe and prolonged pain" was keyed to regulatory language which required pain to be discounted when it is not "expected."
The Court finds plaintiffs' claim unsustainable. Plaintiffs' argument here would be plausible only if the 20 examples had been formulated as definitive rules. However, as noted above, the 20 examples simply were not intended as a negative listing, but were to be used as illustrative examples of nonsevere impairments. Any allegation of disabling pain by a claimant, whose condition otherwise conformed to one of the examples, would render the listing inapplicable and require further evaluation of the claimant. This is confirmed by unrebutted deposition testimony of two SSA officials.
When asked, with specific reference to example 1-a of the POMS, to define the scope of inquiry as to a claimant's allegations of pain, Albert Harrison, Deputy Director of the Division of Medical and Vocational Policy in SSA's Office of Disability, responded:
What is intended here by the term "symptoms of pain" means minimal pain, minimal pain that doesn't impose any functional limitations.
If there is an allegation of disabling pain or severe pain, prolonged pain that would impact functional capacity, then it would not fit this example.
Harrison Tr. at 220. When asked whether this was articulated in the example, Harrison explained:
It doesn't say that explicitly.
But underlying the concept of not-severe impairment is no more than a minimal limitation of function caused by the impairment.
So, if there was an allegation of more than minimal limitations caused by the apparently caused by the pain, then it would the pain just couldn't be disregarded, it would have to be explored to see just what the limitations were.
Harrison Tr. at 221. Harrison concluded that "an allegation of disabling pain or severe pain, prolonged pain that would cause functional limitations, would have to be explored." Id.
Similarly, Dr. Henrietta Moritz, Associate Deputy Director at SSA's Office of Medical Evaluation for Policy, testified that the 20 examples would not apply if the claimant alleged pain or any other symptom which would not ordinarily be found as part of the disorder. See Moritz Tr. at 166-68, 71.
Plaintiffs offer little to rebut the testimony of Harrison and Moritz to the effect that allegations of severe or disabling pain would remove a claimant from the domain of the 20 examples. While it is true that the SSR contained language indicating that the examples were not expected to cause severe and prolonged pain and the regulation specified that pain be discounted when it is not expected, it is not the case that this, without more, violates the Second Circuit's requirement that pain be fully considered. Considered facially, the SSR and the regulation provide adjudicators with the necessary leeway to consider pain in accordance with the Second Circuit's requirements. As Harrison's testimony confirms, allegations of severe pain, even when such pain would not be expected, would have had to be explored under the SSR. While there may have been misunderstanding among adjudicators with regard to this issue, this may be best understood as another facet of the type of misapplication of the severity standard discussed below and not as a defect sufficient facially to invalidate the SSR.
D. Validity of the POMS and SSR under the APA
Plaintiffs also challenge the validity of the POMS and the SSR under the Administrative *949 Procedure Act ("APA"). Plaintiffs note that under the APA the Secretary is required to subject its internal instructions to public notice and comment. Specifically, plaintiffs claim that the list of 20 examples contained in the POMS and SSR was substantive in nature and SSA was required to publish the list in accordance with the APA, 5 U.S.C. § 553. Plaintiffs base this argument on the conclusion that the 20 examples were medical standards that SSA required its adjudicators to apply rigidly in determining severity. However, as discussed above, the 20 impairments listed in the POMS and SSR were intended as illustrative of the types of impairments that would be deemed nonsevere. Only on the rare occasion when an impairment exactly met the characteristics of one of the listed impairments, with no additional impairments or symptoms alleged, was an individual to be deemed not severely disabled based on an example alone.
In Donovan v. Red Star Marine Services, Inc., 739 F.2d 774, 783 (2d Cir.1984), the Second Circuit held that rules which do not change "existing rights and obligations" are considered interpretative and formal rulemaking procedures are not required. Accord McKenzie v. Bowen, 787 F.2d 1216, 1222 (8th Cir.1986); United States v. Walter Dunlap & Sons, 800 F.2d 1232, 1238 (3d Cir.1986). Here, the 20 examples were intended to be guidelines for applying the severity regulations and were not intended to add or otherwise change the regulations. Considering the issue of whether the list of 20 impairments was substantive or interpretive in nature, the Third Circuit, in Bailey v. Sullivan, 885 F.2d 52 (3d Cir.1989), concluded that:
a careful examination of SSR 82-55 leads us to conclude that the [list of impairments] contains merely examples of the application of the severity regulations, which themselves were subject to notice and comment, and that the SSR represents the type of interpretation which the Secretary has discretion to announce.
Id. at 62. See also Jackson v. Heckler, 580 F. Supp. 1077 (E.D.Pa.1984) (upholding SSR 82-55 against APA challenge). In Bailey, the Court viewed the list of 20 impairments as interpretive and, therefore, as a clarification of SSA's policy, rather than a substantive change of the policy. Because there is no meaningful distinction between the instant case and Bailey with respect to the validity of the list of 20 examples, the Court finds both the POMS and the SSR valid under the APA.
E. Misapplication of the POMS and the SSR
In addition to challenging the facial validity of the examples in the POMS and the SSR, plaintiffs charge that adjudicators systematically applied the POMS and the SSR in a manner that is inconsistent with the Social Security Act. Plaintiffs contend that this misapplication took two basic forms. First, plaintiffs assert that adjudicators applied the examples to deny claims even when impairments did not match the terms of the examples. Second, plaintiffs contend that adjudicators interpreted the instructions as raising the severity standard generally so that even claimants whose claims were not disposed of by the examples received adjudications under an elevated severity standard, resulting in thousands of improper denials.
To support their contentions, plaintiffs cite numerous documents in the Stipulated Joint Appendix. Much, but not all, of the evidence that plaintiffs cite has been reviewed by the court in Wilson v. Sullivan, 734 F. Supp. 157 (D.N.J.1990). There, as here, plaintiffs alleged that the Secretary had used a harsher standard to weed out baseless claims than Yuckert and the Social Security Act permit. The Wilson court was unpersuaded by plaintiffs' evidence, finding that plaintiffs had failed to make a colorable showing of systematic misapplication of the SSR. While mindful of the conclusions reached in Wilson, this Court must, of course, conduct its own independent review of this evidence before it.
Plaintiffs first offer Harrison's testimony. Harrison stated that "program experience indicated that the examples were not used properly" and explained that while it was rare to see a case that was exactly *950 analogous with one of the examples, his experience indicated that situations that were close to the examples were often treated as being identical. Harrison Tr. at 420. Illustrating this point, Harrison explained that in the use of examples there is a tendency toward overgeneralization and noted with respect to the colostomy example in the POMS and SSR, that "there is a tendency to use maybe any colostomy situation as a not-severe example and that is improper." Id.
In drafting SSR 85-28, which replaced SSR 82-55, Harrison wrote that the list of
20 specific examples created an unintended disposition to (erroneously) reach conclusions of not severe (despite the presence of findings not contemplated by the examples)....
Exh. 50, at 3. At his deposition, Harrison further emphasized that a consensus emerged at SSA that the examples "were more likely to be misapplied than to give guidance." Harrison Tr. at 423. This view ultimately led the agency to rescind the examples. See id. at 403-06; 431-32.
Plaintiffs also rely upon the deposition testimony of Patricia Owens, SSA's former Associate Commissioner for Disability and head of the Office of Disability from 1982 to September 1986. Like Harrison, Owens testified that the examples "could be used too literally, perhaps; that they raised more questions than they answered; that there would be a tendency to use examples rather than the intent of the regulation." Owens Tr. at 63. Owens further stated that she viewed it as "likely" that the examples created "inconsistency" in application of the severity regulation, id. at 64, meaning, essentially, that the regulation was "misapplied." Id. at 52.
Defendant insists that the testimony of Harrison and Owens shows, at most, that SSA became aware at some point that the 20 examples were not always being applied as intended and that adjudicators sometimes used the examples to deny claims in cases where the examples were not applicable. This, Defendant insists, shows that the misuse of the examples occurred only on a case-by-case basis. The Secretary also emphasizes that the Wilson court found little probative value in the testimony of Harrison and Owens. As the Wilson court remarked: "That adjudicators misapplied the examples does not tend to show that the Secretary implemented a policy to use the examples to increase the number of denials of benefits." Wilson, 734 F. Supp. at 169.
Plaintiffs respond by stressing the difference between the factual record in this case and in Wilson. Plaintiffs emphasize two differences in particular. First plaintiffs note that the Wilson court did not have before it or did not utilize the 1984 comments from SSA's adjudicating components (Exhs. 13 and 40) which plaintiffs allege show deficiencies in the examples both facially and in application. Plaintiffs quote directly from the comments of Dr. Gordon Luk, a physician on SSA's Medical Consulting Staff, who concluded, after reviewing comments on the examples from agency adjudicating components: "These examples constitute one of the most often abused and/or misinterpreted set of guidelines in disability evaluations. Much of the confusion could be eliminated by simply deleting the entire section, as suggested by many reviewers." Exh. 40, at p. 143. Plaintiffs also cite the Regional Commissioner for Seattle to the effect that "the examples of `not severe' impairments have created problems: ... the concept may be proving detrimental to equitable adjudication because it is highly judgmental and has processing biases that encourage its inappropriate use...." Exh. 13 at 63.
Plaintiffs next argue that in rejecting the claim of a pattern and practice of misapplication of the examples, the Wilson court either lacked or misunderstood the statistical evidence that is before this Court. The Wilson court stated that it was unable to locate the statistic that SSA's own 800-case study had concluded that 37% of step two denials were erroneous. 734 F. Supp. at 169. The statistic is readily discernible on the record here and can be derived from exhibit 24, a memorandum from Owens to Regional Commissioners dated March 14, 1986, and exhibit 49, a memorandum from *951 Aaron Krute to Hugh Meade dated May 22, 1985. These exhibits reveal that of the 37% of step two denials that were erroneous, 22% of cases were found to be actually "severe," and in an additional 15% of cases documentation was inadequate to determine severity.
At her deposition, Owens was asked about exhibit 49:
Q: Did these study findings indicate that, in the view of the Office of Disability, in 37 percent of the study cases the state agencies had adjudicated the cases incorrectly?
A: ... you're asking me whether the non-severe step was correctly applied, not whether the final outcome of the decision
Q: That is correct.
A: That indicates that the non-severe step was not correctly applied.
Q: Is that a high percentage of cases?
A: I consider that a high percentage of cases.
Q: Do you consider that an unacceptably high percentage of cases?
Owens Tr. 51.
The magnitude of this error rate is made clear by comparison with SSA's accuracy target level of 97% that is used to assess the performance of state agencies. 20 C.F.R. § 404.1643(b). Thus, the study found an error rate more than twelve times greater than SSA's target rate and almost four times greater than the agency's own standard of minimal acceptability. The Court recognizes that the comparison with the overall error rate is somewhat misleading because, as noted above, only 22% of the state-adjudged non-severe cases were determined under the survey to be severe while an additional 15% were found to have insufficient documentation to determine severity. But even if the actual percentage of severe cases erroneously adjudged non-severe were "only" slightly higher than 22%, the Court finds this error rate to be grossly excessive and unacceptable.
The Wilson court's finding that the 20 examples were not systematically misapplied was made without consideration of the highly probative statistics contained in SSA's 800-case study. Among other things, the Wilson court also did not consider the comments from SSA components objecting to the 20 examples and expressing widespread concern over their application. Nor did the Wilson court consider that the systematic tendency to misapply the 20 examples was exacerbated by SSA's failure to reaffirm the de minimis nature of step two screening anywhere in the SSR. Moreover, as is discussed more fully below in the next section, the Wilson court tended to examine the evidence before it fragmentally, failing to acknowledge the overwhelming nature of plaintiffs' evidence when considered as a whole.
The crucial statistics contained in SSA's 800-case study and the alarming remarks of SSA components criticizing the SSR, when considered in light of the Harrison and Owens testimony, provide this Court with more than a sufficient basis for a finding that the examples of not-severe impairments were misapplied by adjudicators.
*952 F. Misapplication of the Severity Regulation Prior to April 1981
Plaintiffs argue that even prior to the issuance of POMS DI § 2102 in April 1981, SSA adjudicators systematically misapplied the severity regulation. In fact, plaintiffs contend that the record shows that the issuance of the 20 examples of not severe impairments in the form of binding policy directives was only the culmination of SSA's heightening of the severity standard.
According to plaintiffs, the first instructions to apply a heightened severity standard came in the form of SSA quality assurance returns to state adjudicating agencies. These returns provide guidance to state agencies concerning the application of SSA policies. In 1976, the New York Office of Disability Determinations ("ODD") noticed a change in SSA's policy concerning denials of impairment on medical grounds alone. In his affidavit, Marvin Lachman, former Chief Disability Analyst and Director of Quality Assurance at ODD, explains that in June, 1976 his office began receiving a greatly increased number of returns with instructions that the claims should be denied as "slight." Lachman Aff. at ¶¶ 16-25 (hereinafter "Exh. 60"). According to Lachman, these returns included increasing numbers of examples in which impairments were deemed not to warrant full review. Significantly, Lachman did not merely state that the quality assurance returns increased in volume; he further asserted that the returns embodied a higher threshold standard. Exh. 60 ¶¶ 19-20; 34, 36.
Plaintiffs point to numerous exhibits in support of Lachman's testimony that, beginning in 1976, SSA applied a heightened severity standard in quality assurance returns. Thus plaintiffs assert that Lachman's testimony is supported by the fact that (1) ODD officials, such as Sidney Houben, New York ODD Director, wrote to SSA's regional office in late 1976 to question the greater number of returns (Exh. 31)); (2) that SSA's regional office had also observed the trend toward increasing returns with concern (Exh. 29, 32); (3) that other SSA components around the country expressed similar concerns (Exh. 28); (4) that SSA's Regional Commissioner for New York has conceded that SSA has a practice of implementing policy changes through the quality assurance process (Exh. 59); (6) that SSA's Not Severe Impairment Work-group Report concluded that the agency's application of the not-severe concept changed in 1975 and that its application of the concept may have defeated Congressional intent by denying potentially meritorious claims (Exh. 48); and (7) that the denial rate at step two climbed precipitously in the late 1970s (Exh. 27).
*953 In the face of this evidence, the Secretary relies heavily upon the decision in Wilson. That court considered and found little merit in much of the evidence outlined above. With respect to Lachman's testimony, the Wilson court noted that Lachman was not cross-examined and, in the court's view, did not set forth "facts to support his conclusion that the SSA applied an increasingly strict threshold standard." 734 F. Supp. at 167. On the whole, the court found Lachman's testimony conclusory. See id.
The court also considered the letter dated December 27, 1976, from Sidney Houben, the New York ODD Director, to Elmer Smith, Associate Commissioner, Officer of Program Policy and Planning, which expresses concern over the increasing number of federal returns. In the letter, Houben states:
[O]ur staff as well as staff in other State agencies have been expressing concern over the ever increasing returns of cases with suggestions a denial based on slight impairment is appropriate. We have had difficulty in relating these individual cases to the current regulations and the provisions in the State Manual.
Exh. 60, att. C, at 1-2. The letter further states that the proposed changes in the definition, which were eventually adopted in 1978, "make it much clearer as to why these returns were forthcoming" and then notes that "we believe this supports our feeling that this change is more than a technical clarification of language." The Wilson court dismissed this letter saying that the "deliberative" nature of the document weighed against its probative value. 734 F. Supp. at 168.
Similarly, the Wilson court dismissed SSA's Not Severe Impairment Work Group's conclusion that, given the increase in the number of denials, the Secretary either misapplied the standard before 1975 or was misapplying it after 1975. The court stated that this conclusion does not offer probative evidence that the Secretary used a harsher standard than permitted under the regulation.
With respect to the statistics before it showing the precipitous rise in denials after 1976, the court stated that these statistics offered no evidence that the Secretary caused the increase in denials. The court attributed the increase in denials to a number of factors including a downturn in the economy, vigorous quality assurance, and use of the Secretary's noncombination policy. Id. at 168-69.
In response, plaintiffs argue that Wilson, and the related arguments of the Secretary, are distinguishable based largely on their approach to the evidence. Plaintiffs contend that the Wilson court erroneously examined each piece of evidence of systematic misapplication in isolation, looking for a "smoking gun" rather than considering each document in context. Thus, plaintiffs contend, when the totality of the evidence is considered, plaintiffs' explanation of the evidence is more likely to be true than that advanced by the Secretary. The Court agrees.
When all the evidence amassed by plaintiffs, including evidence not before the Wilson court, is considered together, it is difficult to escape the conclusion that even prior to the issuance of the POMS and SSR, SSA adjudicators systematically misapplied the severity regulation. As Lachman explained, *954 beginning in 1976, his office began receiving a greatly increased number of quality assurance returns with instructions that the claims should be denied as "slight." Exh. 60, at ¶¶ 16-25. These returns indicated that SSA expected ODD to reject claimants with more serious impairments on medical grounds alone than had previously been the case. Id. In particular, case returns contained more and more examples of impairments that were henceforth to be considered "slight." As Lachman explained, this caused ODD to respond as if a change in the definition of "slight" impairment had occurred:
The more returns ODD received from SSA's Central Office, the more examples ODD received of what the Secretary considered a "slight" impairment. For example, federal returns during this period began to indicate that loss of one eye, hypertension, colostomy and mental retardation with an I.Q. of between 80 and 89 were all examples of impairments that SSA considered "slight." This represented a significant change in SSA policy from that in effect prior to 1976 when SSA policy would not have considered these impairments "slight" and persons with these conditions would have been accorded a full sequential evaluation of their impairment. However, when ODD received federal returns and SSA instructions stating these impairments were to be determined "slight," ODD had no choice but to find similar impairments in other persons' cases as "slight" as well....
Id. at ¶ 20.
Taken alone, the Lachman affidavit would probably not be enough to sustain plaintiffs' case. But the Lachman affidavit must be considered in conjunction with the Houben letter, discussed above, and its observation that SSA's efforts at clarification of the regulation only strengthened ODD's "feeling that this change is more than a technical clarification of language." Exh. 60, att. C at 2. Moreover, a second letter dated December 23, 1976, from Houben to Regional Representative Henry Riley, also stressed that based on the increased number of returns, he believed that changes in the "slightness" language in the regulation, to what became the severity regulation in 1978, were effecting a "redefinition of terms rather than a technical clarification of language...." Exh. 60, att. B at 2.
In response to Houben's letters and others, SSA's New York Regional Commissioner, Joseph Kelly, responded by stating that he too was concerned about the lack of clarity in SSA's slightness policy. Exh. 32. On January 4, 1977, Kelly sent a memorandum to SSA's central office echoing the views of ODD:
The DDS's have been questioning us with increasing frequency about the issue of what constitutes a "slight" impairment. Based on the feedback from BDI quality reviews we must share their concern. Existing DISM guidelines are being interpreted more broadly by [central office] reviewing staff in case-related situations. ...
Exh. 60, att. D.
The Lachman affidavit as well as the Houben and Kelly exhibits all document a broadening of the "slight" impairment concept *955 as early as June, 1976. In addition to this evidence, the Court must also consider SSA's Not Severe Impairment Work Group's conclusion that:
Whatever SSA's actual conception of the minimum impairment level as for policy purposes between 1975 and the present, its application of the concept in deciding claims suggests a change of position. Yet at the same time there was no corresponding change in the statute, and in fact SSA itself stated that the regulations [sic] changes did not constitute a change in the standard. SSA may have correctly applied the minimum impairment standard intended by Congress in 1975, or may be doing so now, but absent a total change in the nature of the claimant population, of which there is no evidence, both cannot be correct.
Exh. 48 at 11 (emphasis original). While the Wilson court found this conclusion ambiguous, this Court finds the Workgroup's conclusion that there had been a change in SSA's application of the minimum impairment level after 1975 to be corroborative of the change in policy suggested by the evidence examined above.
Any remaining doubts as to SSA's misapplication of the severity regulation by 1976 are resolved by the overwhelming statistical evidence before the Court. Until the mid-1970s, very few claims were denied on medical grounds alone as constituting no more than a "slight" impairment and very few claims were returned by SSA's quality assurance components rejecting ODD allowances on the ground that the impairment was "slight." Exh. 60 at ¶¶ 12-15. Consequently, in 1975 only 8.4% of total OASDI disallowances were based on non-severe impairments. By 1977, when the Secretary had embarked upon a new application of the "slightness" test through increased quality assurance returns, total disallowances based on nonsevere impairments had risen to 24.8%. By 1979, when the DISM was in effect, denials had risen to 41.6%, a level commensurate with that seen after the enactment of the POMS in 1981 and the SSR in 1982. Thus, the statistics reveal a precipitous rise in nonsevere denials well before the enactment of the POMS and the SSR.
In his memoranda, the Secretary struggles to explain away this statistical trend by hypothesizing several other possible explanations for the statistics. First the Secretary asserts, without any factual support from the record, that the changed denial rates may have been due to the formalizing of the sequential evaluation process. However, this argument cannot explain why, after the issuance of SSR 85-28, the step two denial rate fell to only half the rate at which it had been running in the late 1970s. Nor can the theory explain why the rate is currently below 5% in New York State. See Exh. 116. This theory is also undercut by the fact that the rise in step two denials was accompanied by a corresponding drop in the overall allowance rate for disability claims, indicating that step two was not being used simply to deny claims that were previously denied on other grounds but was actually being used to deny claims previously found to constitute compensable disability. See Exh. 27.
Second, the Secretary asserts that the high step two denial rate in the late 1970s in New York may have been due to New York State policies which require public assistance recipients to apply for SSI if they appear to be disabled. This theory is untenable, however, because it ignores the fact that the rise in step two denials was a national phenomenon. See Exh. 27. Moreover, the statistics to which the Secretary refers concern the SSI program not the OASDI program and state policies requiring public assistance recipients to apply for SSI cannot possibly explain the increase in OASDI denial rates.
Nor can the increase in the denial rate be explained as a function of an influx of claims or changes in the claimant population resulting from a downturn in the economy. The data does not reveal any increase in OASDI applications when the severity denials soared. Further, SSA's own Workgroup specifically rejected the contention that change in the step two denial rate was due to changes in the claimant population. The Workgroup concluded that *956 "there is no evidence" that the change in step two denial rates was attributable to "a total change in the nature of the claimant population" that might account for the change in denial rates. Exh. 48 at 11.
Lastly, it must be noted that the Wilson court placed great weight on the hypothesis that the rise in step two denial rates may have been due to the Secretary's refusal to consider the combined effects of impairments, rather than to application of a heightened severity standard. See 734 F. Supp. at 167 & n. 7, 169. This theory is questionable for at least two reasons. First, the Wilson court apparently did not have access to SSA's 800-case study that is in the record before this Court. One report on the study specifically indicates that the errors committed by the states in misapplying step two were "not due to the complexity of the presence of multiple impairments." Exh. 49 at 5. The report further states that "[n]ew regulations concerning multiple not severe impairments will not obviate the differences between central office and field ratings," id. at 6, and a different report on the same data even went to so far as to note that "[t]here were no cases with multiple impairments identified as not severe where the combined effect was determined to be severe by the OD physicians." Exh. 24 at 12. While the precise import of these statements is by no means clear, they do appear, at the very least, to offer some evidence that the effect of combined impairments on the rise in step two denial rates was more modest than the Wilson court supposed.
Second, the Wilson court's theory that the rise in step two denial rates may have been due to the Secretary's refusal to consider the combined effects of impairments cannot account for the fact that denial rates had risen dramatically well before the noncombination policy was initiated in December of 1978.
In sum, the evidence in the record is sufficient to support a finding of misapplication even before issuance of the POMS and SSR. In addition to the Lachman affidavit, which is corroborated by the Houben and Kelly letters, the Court finds the statistical evidence on social security denials to be highly probative of the Secretary's misapplication. The Court cannot accept the Secretary's contention that the sharp increase in denials reflected nothing more than occasional mistakes of a few adjudicators who misunderstood instructions.
G. Use of the Secretary's Noncombination Policy
Between December, 1978, and December 1, 1984, the Secretary prohibited adjudicators from considering the combined impact of non-severe impairments on a claimant's ability to perform work-related functions. This policy was first incorporated in the DISM and was included in the SSR. Plaintiffs argue that this "noncombination" policy violated the Social Security Act and resulted in the denial of potentially meritorious claims.
The Second Circuit has long required that the combined effect of all of a claimant's impairments must be considered in determining disability and has repeatedly ruled that the Act requires SSA to evaluate the combined impact on a claimant's ability to work of every impairment, regardless of whether each is considered severe. See, e.g., De Leon v. Secretary, 734 F.2d 930 (2d Cir.1984) (Secretary's decision terminating claimant's SSI benefits reversed in part for failure to consider the combined effect of his four impairments); Cutler v. Weinberger, 516 F.2d 1282, 1285 (2d Cir.1975) (claimant's case remanded for new SSA review of the combined effect of impairments on ability to work).
Every Circuit Court that has considered the issue has likewise held that SSA must consider the combined effect of all of a claimant's impairments in determining disability. See, e.g., Bowen v. Heckler, 748 F.2d 629 (11th Cir.1984); Dellolio v. Heckler, 705 F.2d 123, 128 (5th Cir.1983). This includes every federal appellate court reviewing the issue in the context of a class action. Johnson v. Sullivan, 922 F.2d 346, 350 (7th Cir.1990) (en banc); Bailey v. Sullivan, 885 F.2d 52, 59-61 (3d Cir.1989); McDonald v. Secretary of Health and Human *957 man Services, 795 F.2d 1118, 1126 (1st Cir.1986).
Without citing a single one of the relevant Circuit Court opinions listed above, the Secretary attempts to argue that his noncombination policy was consistent with the Act. Specifically, the Secretary argues that the fact that the Social Security Act defines "disability" as the "inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or 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), requires the presence of at least one impairment which by itself could render the claimant disabled. The Secretary further contends that Yuckert held that Section 423(d) (1) (A) is the controlling provision of the Act with respect to defining disability.
Plaintiffs point to a different section of the Act, 423(d) (2) (A), which mandates that a claimant be found disabled, "if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work, but cannot ... engage in any other kind of substantial work." (emphasis added). Plaintiffs note that this requirement was reiterated by a 1984 amendment to the Act which specifically requires:
[The] Secretary shall consider the combined effect of all of the individual's impairments without regard to whether any such impairment, if considered separately, would be of such severity.
42 U.S.C. § 423(d) (2) (B).
As numerous courts have recognized, the 1984 amendment merely made clear that the Act represented no change in the preexisting statutory requirement that the combined effect of all impairments be considered. Johnson, 922 F.2d at 351 ("The legislation, however was expressly clarifying and did not represent a change in the law."); Bailey, 885 F.2d at 61; McDonald, 795 F.2d at 1127. Thus, the 1984 amendment undercuts the Secretary's argument by tending to show that the singularity of the word "impairment" in 42 U.S.C. § 423(d) (1) (A) was not intended as limiting.
Moreover, as several courts have already determined, the holding in Yuckert was far narrower with respect to Section 423(d) (1) (A) than the Secretary presumes. As the Third Circuit noted in Bailey, "If anything Yuckert reinforced the notion that the SSA concept of disability is functional rather than technical, thus supporting the common-sense realization that the combined effect of non-severe medical conditions can cause a functional impairment that meets the statutory definition of disability." 885 F.2d at 61. And as the Seventh Circuit further emphasized in Johnson, "nothing in Yuckert undercuts our holding in Johnson I [v. Heckler, 7th Cir., 769 F.2d 1202] that the Secretary's policy of refusing to consider the combined effects of non-severe impairments contravened the Social Security Act." Johnson, 922 F.2d at 352. While the Secretary has overlooked these decisions, this Court cannot and therefore concludes that the Secretary's refusal to consider the combined effect of not severe impairments violated the Social Security Act.
H. Equitable Tolling
The Social Security Act requires that a complaint challenging a final decision of the Secretary be filed within 60 days of the claimant's receipt of the decision. 42 U.S.C. § 405(g). This "60-day requirement is not jurisdictional, but rather constitutes a period of limitations." Bowen v. City of New York, 476 U.S. 467, 106 S. Ct. 2022, 90 L. Ed. 2d 462 (1986). Its purpose is "to move cases to speedy resolution in a bureaucracy that processes millions of claims annually." Id. at 481, 106 S. Ct. at 2031. The requirement may be waived by the Secretary, however, according to the "traditional equitable tolling principle" in a "rare case." Id. at 480-81, 106 S. Ct. at 2030-31. "While in most cases the Secretary will make the determination whether it is proper to extend the period within which review must be sought, cases may arise where the equities in favor of tolling the limitations period are `so great that deference to the agency's judgment is inappropriate'" *958 Id. at 480, 106 S. Ct. at 2030 (quoting Mathews v. Eldridge, 424 U.S. 319, 96 S. Ct. 893, 47 L. Ed. 2d 18 (1976)). In addition, equitable tolling is consistent with Congress's intent in enacting Section 405(g) because Congress designed the Social Security Act to be "unusually protective" of claimants. Id., 476 U.S. at 480, 106 S. Ct. at 2030. Thus, where "the Government's secretive conduct prevents plaintiffs from knowing of a violation of rights," federal courts may toll the 60-day limitations period. Id. at 481, 106 S. Ct. at 2030.
Bowen v. City of New York involved an unpublished policy mandating "a presumption applicable at the level of the initial state psychiatric assessment that a failure to meet or equal the [Listing of Impairments] was tantamount to a finding of ability to do at least unskilled work." Id. at 473, 106 S. Ct. at 2026. The policy was "implemented through various internal memoranda and the `returns' process by which SSA sends cases back to the States for correction." Id. at 473-74, 106 S. Ct. at 2026. This manner of enforcement prevented claimants, counsel, social workers, and advisors from discovering the existence of the policy for an extended period of time. Id. at 475, 106 S. Ct. at 2027. Quoting the opinion of the Second Circuit, the Supreme Court noted that:
All of the class members who permitted their administrative or judicial remedies to expire were entitled to believe that their Government's determination of ineligibility was the considered judgment of an agency faithfully executing the laws of the United States. Though they knew of the denial or loss of benefits, they did not and could not know that those adverse decisions had been made on the basis of a systematic procedural irregularity that rendered them subject to court challenge.
Id. at 480-81, 106 S. Ct. at 2030 (quoting City of New York v. Heckler, 742 F.2d 729, 737-38 (2d Cir.1984)). Since the class members were entitled to believe that the denial of their claims was "the considered judgment of an agency faithfully executing the laws of the United States," and since the "full extent of the Government's clandestine policy" was uncovered only after the class suit was brought, the Supreme Court found that equitable tolling was appropriate. Id.
1. Misapplication Claim
Relying on McDonald v. Secretary of Health and Human Services, 612 F. Supp. 293 (D.Mass.1985), modified on other grounds, 795 F.2d 1118 (1st Cir. 1986), and Pittston Coal Group v. Sebben, 488 U.S. 105, 109 S. Ct. 414, 102 L. Ed. 2d 408 (1988), the Secretary argues that tolling is inappropriate here with respect to plaintiffs' systematic misapplication claim. However, this Court has previously rejected the Secretary's arguments based on these cases, see Dixon v. Bowen, 126 F.R.D. 483, 487-88 (S.D.N.Y.1989), and will not further consider these arguments here.
In addition to relying on case law that this Court has already found inapposite, the Secretary appears to argue that the misapplication of the severity regulation was publicly known, and that plaintiffs' are therefore not entitled to tolling of the statute of limitations. This argument is untenable. While claimants may be expected *959 to know of the severity regulation, they cannot possibly be charged with knowledge that the Secretary was misapplying his own regulation. Thus, claimants had no way of knowing that the decisions in their cases were vulnerable to legal challenge because of the systematic error.
While the Secretary was quietly utilizing the quality assurance returns and, ultimately, the 20 examples to toughen the standard of severity, he maintained publicly that no change had occurred in the severity rules. In proposing the severity regulation in March 1978, SSA stated expressly in the Federal Register that the term "not severe" was "not intended to alter the levels of severity for finding of disabled or not disabled on the basis of medical consideration alone." 43 Fed.Reg. 9284, 9287 (March 7, 1978), Exh. 58. The agency never disclosed, and perhaps never fully understood until the recision of the SSR, that while the standard embodied in the regulations remained the same, SSA no longer adhered to that standard in practice.
As this Court noted in its Opinion and Order dated June 29, 1989:
If plaintiffs show that the Secretary applied the severity test in a manner stricter than that set forth in his published regulation, which expressed a policy of using Step 2 to screen out only de minimis cases, plaintiffs will have met the requirements of City of New York.
126 F.R.D. at 488. Since plaintiffs have made the requisite showing of systematic misapplication, the Court finds that plaintiffs have met the requirements of City of New York and are entitled to equitable tolling for all individuals whose claims were denied under the misapplied severity standard.
2. Noncombination Policy
While the Court recognizes that the class of persons denied benefits under the Secretary's illegal noncombination policy is merely a subset of the class of persons denied benefits pursuant to the Secretary's systematic misapplication of the severity regulation, the Court nonetheless undertakes an separate tolling analysis with respect to the noncombination policy.
As plaintiffs note, the Third Circuit in Bailey v. Sullivan, 885 F.2d 52, 64 (3d Cir.1989), rejected tolling of the statute of limitations with respect to this claim for the period after 1980 on the ground that the Secretary's refusal to consider the combined effects of impairments was stated in a regulation, 20 C.F.R. § 404.1522, which stated, in part:
We cannot combine two or more unrelated severe impairments to meet the 12 month duration test. If you have a severe impairment(s) and then develop another unrelated severe impairment(s) but neither is expected to last for 12 months, we cannot find you disabled even through the two impairments in combination last for 12 months. However, we can combine unrelated impairments to see if together they are severe enough to keep you from doing substantial gainful activity. We will consider the combined effects of unrelated impairments only if all are severe and expected to last 12 months.
Notwithstanding the holding in Bailey and the clear language of the regulation to the effect that combined impairments will be considered only if "all are severe," plaintiffs maintain that this Court should toll the statute of limitations. Plaintiffs aver that, in response to public comments, SSA stated that the regulation concerned the "tacking" of different impairments to meet the twelve-month durational requirement of the Act, rather than the combining of two impairments to meet the severity requirement. See 45 Fed.Reg. 55,574-75 (Aug. 20, 1980). However, the Court is *960 unpersuaded by plaintiffs' argument. While the SSA did state that 404.1522 and the related section 20 C.F.R. 416.922 "concern the duration requirement," SSA nowhere stated that these sections concerned the duration requirement exclusively and thus SSA's remarks are not sufficient to offset the plain language of Section 404.1522.
Plaintiffs also note, however, that the Secretary instituted his noncombination policy prior to the promulgation of Section 404.1522 through the DISM, which was issued in December 1978. With respect to the period prior to the promulgation of Section 404.1522 and after the issuance of the DISM, the Wilson court found equitable tolling applicable. Wilson, 734 F. Supp. at 173-74. This Court agrees. Although claimants could have obtained a copy of the DISM for inspection pursuant to 20 C.F.R. § 422.401(b) (4), "it is unrealistic to believe that publication of regulations that permit inspection of the agency's procedural manuals would alert lay persons to the Secretary's failure to comply with the requirements of the Social Security Act." Wilson, 734 F. Supp. at 174.
For the above stated reasons, the Court finds that: 1) SSA adjudicators engaged in a pattern and practice of misapplication of the severity regulation fostered by SSR 82-55 and POMS DI § 2102; 2) SSA adjudicators engaged in a pattern and practice of misapplication of the severity regulation even before issuance of POMS DI § 2102 in April 1981. This pattern and practice began in June, 1976; 3) SSA's policy of refusing to consider whether different nonsevere impairments, when combined together, result in a significant restriction of any basic work activities, violated the Social Security Act; 4) With respect to plaintiffs' misapplication claim, the statute of limitations should be equitably tolled because claimants were not notified and could not reasonably be expected to know that SSA adjudicators were engaging in a pattern and practice of misapplying the severity regulation; 5) with respect to the Secretary's illegal noncombination policy, the statute of limitations should be equitably tolled for the period after the issuance of DISM 321B and before the promulgation of 20 C.F.R. § 404.1522.
Plaintiffs should submit a proposed order for settlement on 10-days notice.
 As clarified in 1980, the severity regulation reads as follows:
If you do not have any impairment(s) which significantly limits your physical or mental ability to do basic work activities, we will find that you do not have a severe impairment and are, therefore, not disabled, We will not consider your age, education, and work experience.
20 C.F.R. § 1520(c).
 This list of 20 not severe impairments first appeared in Program Operations Manual ("POMS") DI § 2102, issued in April 1981, which was itself a expansion of an original list of 5 not severe impairments that appeared in SSA's Disability State Manual ("DISM") 321B of December 1978.
As discussed more fully below, plaintiffs contend that even prior to the issuance of the DISM in December 1978, Defendant effected a heightening of the severity standard through the use of SSA's "quality assurance review" of initial state determinations.
 Plaintiffs' reliance on City of New York v. Heckler, 578 F. Supp. 1109 (E.D.N.Y.), aff'd, 742 F.2d 729 (2d Cir.1984), aff'd, 476 U.S. 467, 106 S. Ct. 2022, 90 L. Ed. 2d 462 (1986), is similarly misplaced. City of New York was a case involving step three and step four in which claimants were denied benefits because defendant assumed that the failure to meet listings at step three automatically translated into a residual functional capacity to do unskilled work at step four.
 In support of their contention that the SSR and the POMS violate the Act and the severity regulation, plaintiffs cite a number of cases which, to the extent that they consider the SSR at all, consider it as applied. These cases are therefore inapposite in this context.
 Exhibits 49 and 24 were apparently not in the Wilson court's record.
 SSA's "threshold level" of accuracy for state agencies is 90.6%. 20 C.F.R. § 404.1643(d). The threshold level is defined as the "minimal acceptable level of performance." Id. at § 404.1641(c). Thus, any error rate above 9.4% is viewed by SSA as unacceptable. Id. at § 404.1643(d).
 As noted below, the Wilson court also appears to have overemphasized the role of the Secretary's illegal noncombination policy in dismissing SSA's systematic misapplication of the severity standard.
 The Court is in no way dissuaded from this view by the statistics, referred to by the Secretary, which show a very low rate of reversals on "Dixon" readjudications. See Exh. 94. The statistics chosen by the Secretary represent readjudications at the lowest levels of hearings in SSA's four-tiered system of administrative adjudication of claims for disability benefits. At the first two tiers, disability determinations are made by State Disability Determination Services ("DDS") pursuant to contract with SSA. See 20 C.F.R. §§ 404.1601, et seq. In New York State, the Office of Disability Determinations ("ODD") is the state agency that adjudicates claims at the first two tiers. See Stip. Facts at ¶ 6.
In contrast to the statistics selected by the Secretary, the data collected by SSA through May 1987 show a disturbingly high rate of reversals at the third tier, hearings before an administrative law judge ("ALJ"). Of 4070 ALJ hearings conducted pursuant to the Dixon preliminary injunction, prior denials were reversed in 1893 cases. See Exh. 97. These figures show that cases reopened and readjudicated at the ALJ level had an extremely high allowance rate of 46.5%.
Moreover, the statistics relied upon by the Secretary do not show how many of the ODD denials were overturned by ALJs on appeal. Given the 46% allowance rate on cases reopened at the ALJ level, it is possible that literally thousands of ODD denials were subsequently reversed, although this information is not available since the preliminary injunction did not require this information to be produced to plaintiffs. See Order, dated July 24, 1984 at ¶ 17.
Finally, as plaintiffs argue, the disparity between the extremely high rate of reversals at the hearing level and the statistics cited by the Secretary may be explained by the fact that during the period when the preliminary injunction was in effect, ODD continued to use an old rating scale that may have perpetuated the heightened severity standard in a disguised form.
 Plaintiffs also argue, in accord with Lachman's testimony, that SSA's restrictive approach to the severity standard was implemented through internal agency instruction, as well as through the quality assurance process. Plaintiffs note, for example, that in December 1978, SSA issued internal operating instructions in its Disability Insurance State Manual 321B (the "DISM") that implemented the severity regulation, which had just been promulgated. According to plaintiffs, the DISM codified a heightened severity standard in a number of respects: (1) The DISM contained a list of five medical conditions, most of which had previously appeared in quality assurance returns, that adjudicators were required to consider "not severe"; (2) the text of the DISM omitted any statement of the de minimis nature of step 2 or any statement that the severity regulation embodies the same threshold test as the slightness standard; (3) the DISM emphasized how impairments which prevent the return to past work may still be considered not severe. Consequently, plaintiffs assert that the DISM played a significant role in increasing denial. Plaintiffs' contend that this is confirmed by the statement of SSA's Vocational Evaluation Branch ("VEB") that the five examples in the DISM led to a large increase in Step 2 denials (exh. 45) and by Harrison's testimony that the change to "not severe" was viewed by adjudicators as a change in the standard. Harrison Tr. at 433.
Because the Court finds that the DISM played a role in the misapplication of the severity standard that closely approximates the role played by the POMS and the SSR, the Court agrees with plaintiffs' assertions regarding misapplication subsequent to the issuance of the DISM. Consequently, the discussion in this subsection is focussed on misapplication in the period prior to the issuance of the DISM in December in 1978.
 Lachman's observations about the role of quality assurance returns are not unique. As the court recognized in City of New York v. Heckler, 578 F. Supp. 1109, 1114 (E.D.N.Y.), aff'd, 742 F.2d 729 (2d Cir.1984), aff'd, 476 U.S. 467, 106 S. Ct. 2022, 90 L. Ed. 2d 462 (1986), returned cases "function much like remands do in the court system, and create `precedents' for determining future cases."
SSA's Regional Commissioner for New York has further explained:
We believe it is disingenuous to maintain that federal quality returns do not establish policy. Policy statements are of necessity broad and cannot address all potential situations. Individual examiners, and indeed DDSs, properly see quality reviews as the test of their application of broad policy statements to specific cases.
Exh. 59. The Regional Commissioner also described how the quality assurance process is used to implement "unofficial, and perhaps unintended policy." Id.
 The Wilson court did not mention this portion of the Kelly letter. The court appears to have dismissed the letter by simply quoting from the last paragraph, which deals with issues that are not relevant to the instant question. See Wilson, 734 F. Supp. at 168.
 The Secretary's argument appears to be based, to a large degree, on the theory that Congressional inquiries concerning the severity regulation made SSA's practices public well before plaintiffs brought suit. In fact, the Secretary spends an inordinate number of pages in his memoranda reviewing legislative history dealing with the Social Security Act. However, as this Court has previously noted "it would be unreasonable to expect that a claimant, diligently attempting to discern the meaning of the Secretary's regulations, would examine Congressional hearings dealing with the Social Security Act." Dixon v. Bowen, 126 F.R.D. 483, 489 (S.D.N.Y.1989). This Court further noted that it was "not persuaded that there is anything in the Congressional testimony that would place a claimant on notice of a covert heightened severity test" and that "[a]t best, the hearings suggest that an increase in severity denials might be due to increased federal guidance through oral and written instructions, training sessions, and quality assurance reviews." Id. Because the substance of these instructions, and perhaps the very existence of these instructions, was unknown to plaintiffs, they had no means of evaluating whether SSA was misapplying its regulations. See id.
Moreover, the Court notes that the legislative history cited by the Secretary consists largely of staff committee prints, rather than the formal reports of Congressional subcommittees. In discussing such Congressional staff documents, the Supreme Court has stated that they are not entitled to "any significant weight" because they are not approved by the Congressional committee members and typically are circulated only for informational purposes. Sullivan v. Finkelstein, 496 U.S. 617, 110 S. Ct. 2658, 2665 n. 8, 110 L. Ed. 2d 563 (1990). Thus, the Court finds these documents to be of little probative value in the instant action.
 Plaintiffs urge this court to grant a mandatory injunction, directing the Secretary to adhere to his current policy governing the denial of claims on medical grounds alone, as stated in SSR 85-28. Plaintiffs have failed to articulate a coherent rationale for the imposition of such an injunction. Instead, plaintiffs cite several cases for the proposition that mere cessation of allegedly illegal conduct does not moot a case and then argue that because SSA might resume its illegal conduct at any time, an injunction should be imposed. This argument does not withstand scrutiny and therefore the forthcoming order should not contain such an injunction.