MEMORANDUM AND ORDER

DOUGLAS P. WOODLOCK, UNITED STATES DISTRICT JUDGE.

This racial discrimination case was brought by ten black plaintiffs against the Boston Police Department, the City of Boston, and the Commissioner of the Boston Police Department (collectively "BPD"). Plaintiffs challenge the hair drug test administered by the BPD to officers and cadets from 1999 through 2006.

I. BACKGROUND

Beginning in 1999, the BPD required officers and cadets to provide yearly hair samples for drug testing. A small percentage of officers and cadets tested positive for cocaine during this period. Even though approximately two-thirds of the officers and cadets were white, more black officers and cadets than white ones tested positive for cocaine.1 During the eight years at issue, black officers and cadets tested positive for cocaine approximately 1.3% of the time, while white officers and cadets tested positive just under 0.3% of the time. Jones v. City of Boston, 752 F.3d 38, 41 (1st Cir.2014).

The officer's union (the Boston Police Patrolmen's Association) and the BPD included a provision in their collective bargaining agreement implementing these drug tests through Rule 111 of the Boston Police Department Rules and Procedures.2 Rule 111 states that "hair testing procedures are effective pursuant to the collective bargaining agreement." The hair analysis is conducted by a private company, Psychemedics Corporation. When a hair sample test result is positive for cocaine, the BPD consults a physician to see whether the officer had received medication that could result in a positive result. A person who tests positive can also request a "safety-net test" to confirm the positive result, but the levels for the safetynet test were not specified in bargaining with the union. The safety-net test originally used a cutoff of 2 ng/10 mg, but that was reduced to .2 ng/10 mg in 2001. An employee who tested positive for the first time could agree to an unpaid suspension of forty-five days and to drug rehabilitation treatment, otherwise the employee would be terminated.

Seven of the plaintiffs are former officers who were fired, an eighth was a former cadet who was fired, a ninth tested positive but agreed to undergo drug rehabilitation and a forty-five-day unpaid suspension as an alternative to being fired, and the tenth was an applicant to the BPD whose contingent job offer was revoked. Jones, 752 F.3d at 41. Each of the ten plaintiffs tested positive for cocaine using the hair test and suffered adverse consequences. Each of the plaintiffs denied cocaine use, arguing instead that the hair test generated false positives indicating ingestion of cocaine,3 a circumstance that is particularly likely to arise with individuals having African-American hair.

The plaintiffs brought this action in state court in July 2005, and the BPD removed the case to this court. In plaintiffs' First Amended Complaint, they alleged that the hair drug testing had a disparate impact on black officers in violation of Title VII of the Civil Rights Act of 1964 and that the testing and terminations violated the Americans with Disabilities Act (ADA), the Equal Protection Clause through 42 U.S.C. § 1983, and Massachusetts law. Judge O'Toole granted summary judgment to the BPD on all claims, in a decision that concluded that the plaintiffs had not established a prima facie case of racially disparate impact under Count I, the Title VII claim. Jones v. City of Boston, No. Civ. 05-11832-GAO, 2012 WL 4530594, *3 (D.Mass. Sept. 28, 2012) aff'd in part, rev'd in, part and remanded, 752 F.3d 38 (1st Cir.2014). Judge O'Toole did not reach two remaining aspects of the disparate impact claim, business necessity and the existence a less discriminatory alternative. Id. The plaintiffs appealed.

The First Circuit reversed in part and affirmed in part. Jones, 752 F.3d 38. The First Circuit reversed the grant of summary judgment to the BPD on the question whether the plaintiffs had made out a prima facie case, and granted summary judgment on this issue instead to the plaintiffs. In doing so, the court rejected the argument that practical significance, which focuses on determining whether a statistically significant difference is sufficiently large, plays any role in determining a prima facie case. The court held that "a plaintiff's failure to demonstrate practical significance cannot preclude that plaintiff from relying on competent evidence of statistical significance to establish a prima facie case of disparate impact." Jones, 752 F.3d at 53. The court affirmed the dismissal of all other claims. The First Circuit, however, declined to determine certain remaining questions: whether the practice is job related and consistent with business necessity and whether the plaintiffs have offered an adequate alternative with a less disparate effect. The case was remanded to the district court to resolve these questions in the first instance. On remand, the matter was reassigned to me.

Simultaneous with the development of this case, five of the ten plaintiffs in this matter were among a group of officers who appealed their terminations to the Massachusetts Civil Service Commission ("the Commission") under Mass. Gen. Laws c. 31 §§ 41-43. After extensive hearings, which will be discussed in greater detail below, the Commission concluded in its February 28, 2013 decision that the hair drug test could not be used on its own to discharge a tenured officer. The Commission ultimately held that some of the appellants were properly discharged and that others, including the five appellants who are also plaintiffs in this case, must be reinstated. The BPD appealed the decision to Suffolk Superior Court, where Judge Fabricant affirmed the Commission's decision. Boston Police Dept. v. Civil Service Comm'n, Nos. 13-1250-A, 13-1256-F (Suffolk Sup.Ct. October 6, 2014) (the "Superior Court Decision"). The BPD has appealed the Superior Court decision to the Massachusetts Appeals Court, where the appeal is pending. Boston Police Dept. v. Civil Service Comm'n, appeal docketed, No.2015-P-0330 (Mass. App.Ct. Mar. 11, 2015).

II. ANALYSIS

The First Circuit held that the plaintiffs have made a prima facie showing of disparate impact in this case; thus the burden now shifts to the BPD to demonstrate that "the challenged practice is job related for the position in question and consistent with business necessity." 42 U.S.C. § 2000e-2(k)(1)(A)(i). If an employer makes that showing, a plaintiff has the burden to prove an "alternative employment practice," 42 U.S.C. § 2000e-2(k)(1)(A)(ii), that would be a "test or selection device, without a similarly undesirable racial effect" that would still serve the employer's legitimate interest. Jones, 752 F.3d at 54 (quoting Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975)). The plaintiff prevails if the employer has refused to adopt such an alternative employment practice. 42 U.S.C. § 2000e-2(k)(1)(A)(ii).

Here, the BPD has again moved for summary judgment pursuant to Rule 56(c) of the Federal Rules of Civil Procedure. Summary judgment is appropriate only where there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law. Fed. R.Civ.P. 56(c); Perry v. Roy, 782 F.3d 73, 78 (1st Cir.2015). A genuine issue of material fact must be built on materials of evidentiary quality—my role is to assess whether there is sufficient evidence for a reasonable jury to return a verdict for the nonmoving party. Id. All facts in the record and all reasonable inferences must be drawn in favor of the nonmovant. Id. at 77. The plaintiffs insist that they are not independently moving for summary judgment themselves, although they note issues—such as their claim of issue preclusion—that may tempt me to use my ability to grant summary judgment in favor of a nonmovant, Fed.R.Civ.P. 56(f).

A. Job Relatedness and Business Necessity

The threshold question of the job-relatedness inquiry is whether the BPD can "show that its program aims to measure a characteristic that constitutes an `important element[] of work behavior.'" Jones, 752 F.3d at 54 (quoting Albemarle, 422 U.S. at 431, 95 S.Ct. 2362). The plaintiffs do not object to the concept of mandatory drug testing of law enforcement officers and applicants for employment as law enforcement, and they do not dispute the value of a drug-free workplace for law enforcement. Plaintiffs do not dispute that drug testing aims to measure a characteristic that is an important element of work behavior and job performance. The job-relatedness requirement therefore is met.

The next step in this inquiry is whether the BPD has shown that the results of the drug testing are business related, meaning "predictive of or significantly correlated with" officers' drug use. Albemarle, 422 U.S. at 431, 95 S.Ct. 2362.

1. Issue Preclusion4

The plaintiffs first argue that due to the prior decision of the Massachusetts Civil Service Commission ("the Commission") in In re Boston Police Department Drug Testing Appeals, Nos. D01-1409 et al. (herein "Commission Decision") subsequently affirmed by the Suffolk Superior Court, Boston Police Dept. v. Civil Service Comm'n, Nos. 13-1250-A, 13-1256-F, the BPD is precluded from arguing that the hair drug testing performed from 1999-2006 could distinguish positive tests caused by contamination from those caused by ingestion.

The plaintiffs in the Commission action were ten tenured black police officers who were discharged after their hair tested positive for the presence of cocaine. They argued that the hair drug test alone was not an adequate basis on which to find "just cause" for termination under Mass. Gen. Laws c. 31 § 41 ("Except for just cause and except in accordance with the provisions of this paragraph, a tenured employee shall not be discharged . . .").

While the Commission initially agreed to postpone hearing the termination appeals while the parties focused instead on this federal civil rights lawsuit, the Commission decided to go forward with a hearing after five years had passed and this matter was still in the discovery stage. The Commission held eighteen days of evidentiary hearings from October 2010 through February 2011. The Commission received two hundred and two exhibits, heard oral testimony from the appellants and expert witnesses and from BPD personnel. The issue before the Commission was whether the BPD had met its burden to prove just cause for termination by a preponderance of the evidence.

The Commission issued a unanimous 132-page decision on February 28, 2013. The Commission affirmed the BPD's right to take steps to identify and terminate police officers using illegal drugs, and found that "the BPD's annual hair testing plan is an appropriate tool to enforce such a policy." Commission Decision at 4. The Commission identified the long-standing debate about how accurately hair tests are able to differentiate between external contamination as opposed to ingestion, the lack of nationally approved standards for hair testing, and inconsistency among laboratory protocols. The Commission concluded that a positive hair drug test result is not conclusive evidence of ingestion because it "does not provide [] 100% irrefutable evidence of drug ingestion," and therefore it is insufficient on its own to provide just cause for termination. Id. at 4. The Commission then looked to the totality of the evidence and reinstated six of the officers, including five of the plaintiffs in this case.

As noted above, the BPD appealed this decision for review to Suffolk Superior Court, Boston Police Dept. v. Civil Service Comm'n, Nos. 13-1250-A, 13-1256-F. The Superior Court affirmed the Commission Decision, modifying it only to the extent that the reinstated officers were awarded full back-pay and benefits as of the date of the officers' discharge. The BPD has since appealed the Superior Court decision to the Massachusetts Appeals Court, and the appeal is pending, Boston Police Dept. v. Civil Service Commission, No.2015-P-0330. Plaintiffs now argue that the BPD is precluded from relitigating the issue of the reliability of the hair drug test because such testing has already been found in the MCAD case to be insufficiently reliable.

State agency decisions are entitled to preclusive effect under certain circumstances. "[W]hen a state agency acting in a judicial capacity . . . resolves disputed issues of fact properly before it which the parties have had an adequate opportunity to litigate, federal courts must give the agency's factfinding the same preclusive effect to which it would be entitled in the State's courts." Univ. of Tennessee v. Elliott, 478 U.S. 788, 799 106 S.Ct. 3220, 92 L.Ed.2d 635 (1986). See also B & B Hardware, Inc. v. Hargis Indus., Inc., ___ U.S. ___, 135 S.Ct. 1293, 1302-05, 191 L.Ed.2d 222 (2015) (reaffirming the capacity of agency decisions to ground issue preclusion in the context of the Lanham Act and a decision of the Trademark Trial and Appeal Board). Massachusetts issue preclusion principles therefore apply. To preclude a party from relitigating an issue, a court must find that "(1) there was a final judgment on the merits in the prior adjudication; (2) the party against whom preclusion is asserted was a party (or in privity with a party) to the prior adjudication; and (3) the issue in the prior adjudication was identical to the issue in the current adjudication." Kobrin v. Bd. of Registration in Med., 444 Mass. 837, 832 N.E.2d 628, 634 (2005) (quoting Tuper v. North Adams Ambulance Serv., Inc., 428 Mass. 132, .697 N.E.2d 983 (1998)). A judgment may be final and subject to preclusive effect, despite the pendency of an appeal. Currie v. Group Ins. Comm'n, 290 F.3d 1, 16 (1st Cir. 2002) (Woodlock, J. dissenting) (citing O'Brien v. Hanover Ins. Co., 427 Mass. 194, 692 N.E.2d 39, 44 (1998)). The issue also must have been essential to the earlier judgment. Id.

While the parties disagree about whether each of the elements required for issue preclusion was met in this case, the most significant aspect of this debate concerns whether the issue in the prior matter was identical to the issue in this matter. The plaintiffs concede that the legal standards are different. The question before the Commission was the existence of "just cause," whereas the issue here is business necessity. Plaintiffs urge me to look beyond the conclusions tied to assessment of whether the evidence has met the legal standard, and to focus instead on the factual findings related to the reliability of the hair drug test that underpin that conclusion. The plaintiffs point, for example, to language from the Commission decision, based on review of the scientific evidence before it, that "[t]he present state of hair testing for drugs of abuse . . . does not meet the standard of reliability necessary to be routinely used as the sole ground to terminate a tenured public employee in Massachusetts." Commission Decision at 107. The plaintiffs argue that the Commission's evaluation of the science of the hair drug test was done separately from the application of the legal standard to the cases before it, so the question of the hair tests' reliability in the face of a risk of external contamination has been conclusively established.

Issue preclusion can occasionally apply even where the legal standards differ between two proceedings. "A difference in the legal standards pertaining to two proceedings may defeat the use of collateral estoppel. But this is so only where the difference undermines the rationale of the doctrine." Bath Iron, Works Corp. v. Director, Office of Workers' Compensation Programs, U.S. Dept. of Labor, 125 F.3d 18, 22 (1st Cir.1997). The plaintiffs submit that the underlying rationale concerned the reliability of the hair drug tests and that therefore the change in legal standard does not defeat the applicability of issue preclusion.

The legal conclusion at the heart of the Commission Decision and the Superior Court Decision was whether the BPD lacked just cause to terminate the officers based solely on hair tests. Boston Police Dept. v. Civil Service Commission, Nos. 13-1250-A, 13-1256-F, at 1. Factually, the "primary focus of the hearing before the Commission was on the reliability of the hair testing performed by Psychemedics, particularly its capacity to distinguish between voluntary ingestion and environmental exposure." Id. at 5-6. The Commission identified scientific disagreement about the accuracy of the hair drug test, the lack of uniform national standards, and other factors, ultimately to conclude that hair test results do not "meet the standard of reliability necessary to be routinely used as the sole grounds to terminate a tenured public employee under just cause standards governing civil service employees under Massachusetts law." Id. at 6-7. The standard that the Commission used to determine whether the hair drug test could serve as the sole ground for termination was whether it left any room for doubt. Noting that the positive hair test results were "not necessarily conclusive of ingestion," the Commission and the Superior Court determined that the BPD was required to consider other evidence beyond the hair drug test itself in making a final termination decision.

While the Commission and Superior Court expressed concern that the hair drug tests do "not provide the 100% irrefutable evidence of drug ingestion that the BPD (and the police union) believed it did when the policy permitting such testing was negotiated and implemented," id. at 7, the Superior Court noted that "[t]he commission did not conclude that the hair tests supplied no evidence; its conclusion, rather, was that hair tests alone were not sufficiently reliable to outweigh a credible denial." Id. at 25.

The Commission and Superior Court decisions make clear that the accuracy of the hair drug tests is less than 100%, but neither made a finding of how much less. At one point, the Commission noted that "potential variations [in instrumentation error] of as much as 20% (or even 5%) are not acceptable when they would be used to make the difference in an employer's decision whether an employee is retained in employment or terminated." Commission Decision at 112. While the Commission did not permit the BPD to rely solely on the hair drug test, it noted that "[f]or all of its flaws, the BPD hair test has a legitimate place in narrowing down which of its few officers may reasonably be suspected of abusing illicit drugs." Id. at 114. In fact, the Commission used the evidence from the hair drug tests as evidence of drug use in its own conclusions about which of the plaintiffs were properly terminated and which were improperly terminated.

A fair review of the Commission and Superior Court decisions makes clear that the factual conclusions of those decisions fall significantly short of, and even undermine, the conclusions that the plaintiffs seek to have drawn from those decisions. While the prior decisions did find that the hair drug test was not 100% accurate, they also found that it may be used as a significant piece of evidence of drug use and that it was helpful in identifying which officers may be drug users. The Commission's factual findings, while supportive of the conclusions of the Commission and affirmed by the Superior Court, do not support a finding of no business necessity. The specific factual findings made by the Commission cannot be divorced from the legal standard that it applied and interpreted. I have concluded that the issue decided in the Commission proceeding is not identical to the one before me and issue preclusion does not prevent the BPD from relitigating the question whether the hair drug tests are predictive of or significantly correlated with officers' drug use.

2. Summary Judgment on Business Necessity

For an employment selection device to be valid, it must be shown that the examination is "predictive of or significantly correlated with important elements of work behavior," Jones, 752 F.3d at 54-55 (quoting Albemarle, 422 U.S. at 431, 95 S.Ct. 2362), which in this setting is a drug-free workplace. The First Circuit chose not to reach the business necessity analysis but noted some skepticism that the evidence plaintiffs had adduced would be sufficient to counter the BPD's evidence. The court wrote that plaintiffs "have presented little if any evidence that could allow a jury to conclude that the drug test is so unreliable that its results have no significant correlation with drug use. Indeed, even their own evidence, if believed, would offer cause to question the accuracy of only some of the reported results, without indicating that there is a relatively large number of false positives compared to the size of the police force."5 Jones, 752 F.3d at 54. The court noted, however, that the burden rests squarely with the BPD to make the business necessity showing. Id.

a. Evidence Related to Business Necessity

From 1999 to 2006, the BPD conducted 15,057 tests, out of which eighty-five test results were reported to be positive. Jones, 752 F.3d at 44-45. Of 4,222 black officers tested during 1999-2006, fifty-five tested positive for cocaine during that period. Id. It is some portion of these fifty-five tests that plaintiffs identify as being potential false positives due to a racially disparate effect. The BPD contends that the number should be lower than fifty-five, because numerous officers during that period signed a rehabilitation agreement and took a 45-day suspension, admitting that they had actually used illicit drugs. The plaintiffs challenge that conclusion, noting that many officers felt they had no choice but to sign the rehabilitation agreement in order to keep their jobs despite maintaining that they did not use illicit drugs. The First Circuit rejected the BPD's effort to use the rehabilitation agreements to shrink the size of the pool of possible false positives, Jones, 752 F.3d at 47, and in the absence of additional evidence bearing on the significance of signing the rehabilitation agreement, I do so as well.

In broad terms, the parties disagree about the extent to which hair drug testing is widely accepted by the scientific community. The plaintiffs note that the federal government has chosen not to use hair drug testing in its workplace testing programs. They also present evidence that despite the federal government's issuance of Proposed Revisions to Mandatory Guidelines for Federal Workplace Drug Testing Programs in 2004, the federal government chose in 2008, after four years of study, not to adopt standards for hair drug testing, noting that "significant issues have been raised by Federal agencies during the review process which require further examination, and may require additional study and analysis." Plaintiffs note further that the Society of Forensic Toxicologists has not revised its 1992 consensus opinion that "results of hair analysis alone do not constitute sufficient evidence of drug use for application in the workplace." There are no national standards or national certification programs for hair testing laboratories and no federally-certified laboratories for hair drug testing.

In contrast, defendants point to the fact that Psychemdics has forensic drug testing accreditation for hair drug testing from the College of American Pathologists, and that Psychemedics is one of the few laboratories selected as a reference laboratory by the international professional scientific society dedicated to hair drug testing, the Society of Hair Testing. Defendants note that Psychemedics has received and maintained certification from the Clinical Laboratory Improvement Amendments, a federal government laboratory certification, as well as from numerous states including New York and Florida. There is no evidence in the record that Psychemedics has ever failed any certification or proficiency exam related to hair drug testing.

The plaintiffs' objections to the reliability of hair drug testing focus on the risk of external contamination, not the ability of the test properly to identify the presence of cocaine or cocaine metabolites. It is undisputed that the Psychemedics test is able reliably to detect the presence of cocaine and cocaine metabolites at the cut-off level used by the Psychemedics test.6 Even the plaintiffs' expert, Dr. Walsh, testified at his deposition that he has no reason to believe that the BPD test cannot test accurately and reliably at the 5ng/10 mg cut-off level, although he expressed concern that the cocaine identified could come from contamination rather than ingestion.

The parties agree that contamination by cocaine theoretically could result in a false positive. This is particularly true for individuals who are exposed to cocaine in their home lives or through their work.7 There is a greater likelihood that BPD officers would encounter cocaine in their daily work than people with most other occupations. The BPD's position is that Pyschemedics' decontamination procedures are sufficient to remove the concern about external contamination and that the Psychemedics hair drug test only identifies ingested cocaine. Psychemedics' decontamination procedure includes a 3.75 hour, five-part wash procedure with a fifteenminute wash with isopropyl alcohol and vigorous shaking to remove external contaminants, then three aqueous phosphate buffer washes each lasting a half hour with vigorous shaking, followed by two additional one-hour phosphate buffer washes. The last wash is retained and analyzed, and any drug amount found in the last wash is multiplied by five and deducted from the ultimate results in what Psychemedics contends is a conservative estimate of the drug amounts that may remain on the hair due to external contamination.

The BPD identified several studies that they claim demonstrate the effectiveness of these decontamination procedures. For their part, the plaintiffs have identified reasons that these tests would not translate to a real-world scenario, for example due to the brief period of time between contamination and decontamination in the study. There is some evidence in the record that the Psychemedics wash procedures are particularly effective relative to other wash procedures.8

Psychemedics also looks not only at the amount of cocaine present in the hair sample, but also at the ratios between the cocaine and cocaine metabolites. Cocaine metabolites are produced by the human body as it expels cocaine, and therefore the presence of metabolites in particular ratios is used by Psychemedics as an indicator of ingestion rather than mere exposure. Plaintiffs present expert testimony that casts doubt on the proposition that the presence of metabolites necessarily indicates ingestion rather than exposure. For example, plaintiffs' expert, Dr. Kidwell, noted that some metabolites are produced during the washing decontamination process itself. Some metabolites may be present in cocaine on the street due to impurities, or they may develop through degradation of cocaine once it is incorporated into hair through exposure. A 2009 Department of Justice report further expressed concern about the use of ratios to distinguish exposure from ingestion, noting that metabolites can be present in cocaine as a result of the manufacturing process in unknown ratios.

Despite these problems, two of plaintiffs' experts acknowledge that the decontamination procedures remove a significant portion of cocaine contamination from exposure. Dr. Walsh has stated that "over the last few years many researchers have demonstrated that while wash procedures can remove some of the cocaine present in hair, they cannot remove it all." Dr. Kidwell credited Psychemedics' procedures with removing between 83 percent and 97 percent of external contamination.

In addition to general concerns about the ability of Psychemedics' testing to distinguish between exposure and ingestion, the plaintiffs' primary contamination-based objection to the hair test relates to contamination that passes the outer layer of the hair, the cuticle, and reaches the cortex, the inner portion of the hair that contains melanin, the substance that determines hair color. Cocaine binds to melanin, so individuals with higher proportions of melanin, such as those with darker hair, are likely to have higher levels of cocaine in their hair due to environmental exposure.9

In order for cocaine to reach the melanin in the cortex, the cuticle would have to be damaged. Dr. Kidwell's affidavit suggests that this is more likely to be the case with officers with African-American hair than other officers. Dr. Kidwell's affidavit explains that cosmetic treatment often leads to damage to the cuticle, which permits cocaine to enter the cortex and bind to melanin. This is exacerbated by applying topical cosmetic products containing oil and glycerol that are often used on cosmetically treated hair. Oil absorbs and concentrates cocaine from the environment, and glycerol swells the hair and provides the opportunity for drug transfer. Black officers are more likely to treat their hair cosmetically and to use certain cosmetic products for cultural reasons. Defendants counter that the effects of hair treatment on drug absorption due to environmental exposure are speculative and have not been shown in studies to happen with any significant regularity.

b. Standard Required for Business Necessity

The terms "business necessity" and "job related" are not clearly defined in the law, but when Congress enacted the Civil Rights Act of 1991, Pub.L. No. 1-2-166 § 3, 105 Stat. 1071 (1992), it also explicitly approved of the standards articulated by the Supreme Court in Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), and Albemarle, 422 U.S. 405, 95 S.Ct. 2362, while rejecting the less rigorous interpretation in Wards Cove Packing Co. Inc. v. Atonio, 490 U.S. 642, 109 S.Ct. 2115, 104 L.Ed.2d 733 (1989)(holding that the employer only bears the burden of production). 317 Cong. Rec. S. 15,276 (daily ed. Oct. 25, 1991)(interpretive memorandum); see also U.S. v. Mass., 781 F.Supp.2d 1 (D.Mass.2011)(detailing the various standards stated by the Supreme Court and Congress' response in the Civil Rights Act of 1991).

In Griggs, the Court held that the employer must show that a challenged practice is "demonstrably a reasonable measure of job performance," 401 U.S. at 436, 91 S.Ct. 849, has "a manifest relationship to the employment in question," id. at 432, 91 S.Ct. 849, and "bear[s] a demonstrable relationship to successful performance of the jobs for which it [is] used," id. at 431, 91 S.Ct. 849. In Albemarle, the Supreme Court held that "discriminatory tests are impermissible unless shown by professionally acceptable methods, to be `predictive of or significantly correlated with important elements of work behavior which comprise or are relevant to the job or jobs for which candidates are being evaluated." 422 U.S. at 431, 95 S.Ct. 2362.

The standards for job relatedness and business necessity—focusing on reasonableness, probability and correlation—reflect the fact that the law does not require that an employer prove that a test or evaluation is perfect. A test can have an error rate but still be predictive and significantly correlated with a job function. The degree of correlation required between the qualification and an essential job function is reflected on a sliding scale, depending on the size of the test's adverse impact. See, e.g., Bates v. United Parcel Serv., Inc., 511 F.3d 974, 996 (9th Cir.2007) (quoting Clady v. County of Los Angeles, 770 F.2d 1421, 1432 (9th Cir.1985)). A number of courts have permitted more leeway for employment tests that relate to public safety, as this one does. See, e.g., Bates, 511 F.3d at 996 (quoting EEOC v. Exxon Corp., 203 F.3d 871, 875 (5th Cir. 2000)). That an assessment tool could be more accurate need not itself be dispositive, so long as it reliably predicts an employee's suitability for a job based on an important element of job behavior. See Lopez v. City of Lawrence, 2014 U.S. Dist. LEXIS 124139, at *61 (D. Mass. Sept. 5, 2014), appeal docketed, No. 14-1952 (1st Cir. Sept. 17, 2014) ("for assessing validity, the fact that [a civil service examination] could have been better does not mean necessarily that it was not good enough to be deemed sufficient. The key question regarding the content validity of a selection method is whether it reliably predicts a candidate's suitability for the job.").

Plaintiffs argue that the hair drug test's inability to identify all cases of external contamination make it unreliable and therefore it cannot be considered significantly correlated with drug use. That position relies on an all-or-nothing view of reliability. However, the job-relatedness inquiry permits a reasonable error rate so long as there is a significant correlation or predictive nexus between the test and the job requirements. As the First Circuit stated in Jones, plaintiffs' evidence "offer[s] cause to question the accuracy of only some of the reported results, without indicating that there is a relatively large number of false positives compared to the size of the police force." Jones, 752 F.3d at 54. A review of the summary judgment record before me confirms that the plaintiffs have failed to present evidence from which a reasonable jury could determine that the hair drug test is not predictive of or significantly correlated with drug use.

Even if the test only determines exposure to cocaine through either ingestion or environmental contamination over a threshold level, it nevertheless likely would be significantly correlated with drug use, because a person whose hair tests positive for a higher level of cocaine, regardless of the source of the cocaine, is more likely to be a drug user than a person whose hair tests negative. As more precautions are built into the process, the less likely it is that the cocaine would be present in the test results at the cut-off level for reasons other than ingestion, although the test may never reach 100% accuracy. Once a threshold of significant correlation is reached, the question whether the test must be further improved is the subject of the superior alternative inquiry, not a function of the business necessity or job-relatedness inquiry.

Here, the record provides evidence of such precautions and supports the proposition that the number of false positives, if any, would be a small percentage of the samples that did have cocaine on or in them due to environmental exposure. There is no dispute that the hair test reliably identifies cocaine and cocaine metabolites at a level of 5 ng/10 mg. Plaintiffs do not dispute that their hair samples tested positive for cocaine or cocaine metabolites. There is no material dispute that the officers who tested negative properly tested negative. Dr. Kidwell, plaintiffs' expert, agreed that the extensive decontamination procedures by Psychemedics remove a very significant percentage of environmental contamination, from 83 to 97 percent.

Plaintiffs vigorously dispute the ability of the decontamination procedures to remove completely cocaine that has bound to melanin in the inner cortex of a hair specimen due to environmental exposure and the ability of the hair testing to distinguish between environmental cocaine and cocaine that has been processed through a person's body. Evidence in the record does support a dispute of fact on these points. However, the plaintiffs have not presented any evidence that shows this happens in significant numbers in real-world situations. Plaintiffs have shown that cocaine could enter through the cuticle into the cortex, particularly for damaged or treated hair, and have stated that African-American hair would be more likely to have a damaged cuticle, although they have presented no evidence that the plaintiffs in this case had treated or damaged hair.10

Undisputed evidence presented by the BPD suggests that the effect—if any—due to external contamination is a small one after the hair samples go through the Psychemedics decontamination procedure. The eight sworn officer plaintiffs who went through the hair test multiple times never previously tested positive for cocaine. The BPD notes that the BPD's Drug Control Unit, which is responsible for conducting drug investigations and undertaking drug arrests, is made up of eighty-eight officers, including an unspecified number of officers with African-American and other dark hair. None of those officers has ever tested positive. Similarly, none of the officers in the Evidence Management Unit, which regularly manages cocaine and other drug evidence, has ever tested positive on the BPD hair drug test. Id.

Plaintiffs' experts themselves acknowledge that the hair drug testing has legitimate and valuable role to play as a screening mechanism in identifying officers who may use drugs, as evidenced by Dr. Kidwell's suggestion discussed below that hair drug testing be part of a proposed superior alternative. In the face of the BPD's evidence of the general reliability of the hair drug test at detecting cocaine and cocaine metabolites and the efficacy of the decontamination procedures on most types of exposure contamination, the plaintiffs' hypothetical evidence related to the possibility of unremediable contamination is not evidence of a significant enough effect to defeat the BPD's evidence that the hair drug testing generally is reliable and is predictive of and significantly correlated with important elements of work behavior. See Jones, 752 F.3d at 54-55.

B. Superior Alternative Practice

Once the BPD has demonstrated—as it has here—the business necessity and job relatedness of the hair drug test, the plaintiffs still "have one final path to success, by proving the existence of an `alternative employment practice,' 42 U.S.C. § 2000e-2(k)(1)(A)(ii), defined in case law as a different `test or selection device[], without a similarly undesirable racial effect,' which `also serve[s] the employer's legitimate interest. . . .'" Jones, 752 F.3d at 54 (quoting Albemarle, 422 U.S. at 425, 95 S.Ct. 2362). Plaintiffs must show that "the employer refuses to adopt an available alternative employment practice that has less disparate impact and serves the employer's legitimate needs." Jones, 752 F.3d at 55 (quoting Ricci v. DeStefano, 557 U.S. 557, 578, 129 S.Ct. 2658, 174 L.Ed.2d 490 (2009)).

At this stage, the burden of production and persuasion shifts back to the plaintiffs. Albemarle, 422 U.S. at 435, 95 S.Ct. 2362. Given that "disparate-impact plaintiffs bear the burden of production and persuasion at this stage . . . plaintiffs may not rest on speculation regarding the availability, validity, or less discriminatory nature of their proffered alternatives." Johnson v. City of Memphis, 770 F.3d 464, 472 (6th Cir.2014) (citing Allen v. City of Chicago, 351 F.3d 306, 313, 316-17 (7th Cir.2003)).

The First Circuit opinion in this case repeatedly emphasized the high standard, "often requir[ing] extensive data," for proving an alternative practice, particularly where the sample size is large and the racially disparate effect is small, as it is here. Jones, 752 F.3d at 53. "A plaintiff who subjects a defendant's job-related practice to the sensitivity of a large sample analysis can fairly be required to show through statistical evidence, and with equal confidence, that the proffered alternative practice will have a smaller impact . . . ." Id. The plaintiffs' statistical expert, Sharon Kelly, concededly did not analyze the statistical performance of other types of drug tests—her analysis was focused exclusively on the BPD hair test results from 1999-2006. Kelly also did not analyze the results of modifications that the BPD and the union made to Rule 111 and the hair testing procedure in 2007.

Despite the lack of such statistical evidence, I consider below whether the plaintiffs have presented the type of extensive data that the First Circuit identified as necessary to prevail at this stage. The plaintiffs must show an alternative (a) with less disparate impact, (b) that is equally valid, and (c) that the BPD refused to implement.

1. Urinalysis

Plaintiffs first propose that the BPD should have used urinalysis instead of the hair drug testing. The plaintiffs point to a long history of the BPD's use of urinalysis, noting that the department has used urinalysis on a reasonable suspicion basis since the early 1990s. The BPD no longer uses urine testing regularly, since the Supreme Judicial Court ruled in 1991 that random urinalysis was an unreasonable search and seizure under the Massachusetts constitution in the absence of a showing of a drug problem in the BPD, Guiney v. Police Commissioner of Boston, 411 Mass. 328, 582 N.E.2d 523 (1991). It continues to use urinalysis for officers who agree to undergo drug rehabilitation treatment after a positive hair test. The plaintiffs note that urinalysis is less expensive than hair testing and is widely accepted in the scientific community.

It is undisputed that urinalysis has a longer history of use and validation than the hair drug tests. It is the sole approved method for federal employee drug testing and it has been regulated since the mid-1980s with an extensive regulatory and oversight system.

Commissioner Evans testified in a deposition that he noticed an increased disparity in the number of minority officers who were testing positive shortly after the hair drug test was implemented instead of urinalysis. The plaintiffs' and the BPD's experts both addressed the lack of racial disparity or bias in urine testing. In his deposition, Dr. Walsh, the plaintiff's expert, stated that "there was never any indication of any kind of racial bias" in urinalysis.11 The BPD's expert, Dr. Kadehjian, in his deposition stated that "[u]rine testing, I would hold, certainly has no evidence of racial bias." In response to the question, "You personally don't hold the opinion that urinalysis is race-biased, do you?" he answered, "No, I do not."

Dr. Kadehjian's expert report, however, stated that studies he had reviewed "demonstrated similar relative rates of positive results between African-Americans and Caucasians for urine testing, hair testing and self-reported drug use, with African-Americans being identified about twice as often as Caucasians, regardless of the test used. These studies even indicated that urinalyses detected African-Americans relative to Caucasians at a slightly greater rate than hair analyses." After reviewing the various experts' statements and the observations of Commissioner Evans, I find that there is at least a dispute of material fact about whether there is any racial bias or racially disparate effect in urine testing, although the evidence seems largely to support the proposition that there is not. By contrast, it is established, as a matter of law of the case, that there is some racially disparate effect in hair drug testing, at least as practiced by the BPD and Psychemedics, Jones, 752 F.3d at 53 (granting summary judgment to plaintiffs on the prima facie evidence of a racially disparate impact).

To defeat summary judgment in favor of the BPD, the plaintiffs must not only show that there is a question of material fact that urinalysis likely has less disparate impact but also that there is a dispute of material fact about whether urinalysis is an equally valid means of testing officers for drug use. The plaintiffs point to the wide acceptance of urinalysis, such as for federal workplace drug testing, as well as the BPD's continued use of urinalysis for individuals who tested positive with the hair drug test and are participating in rehabilitation, as evidence that urinalysis is an equally valid alternative to hair testing. There is undisputed evidence in the record, however, that hair testing and urinalysis are not equally valid alternatives, particularly when the benefits and burdens of urinalysis are considered. See Watson v. Ft. Worth Bank & Trust, 487 U.S. 977, 998, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988)(plurality)("Factors such as the cost or other burdens of proposed alternative selection devices are relevant in determining whether they would be equally as effective as the challenged practice in serving the employer's legitimate business goals.")

Urinalysis reliably identifies ingestion of drugs over only a 48-hour window. A hair drug test, by contrast, covers a 90-day window. Urine can be adulterated, such as through dilution, whereas hair samples are less easily manipulated.12 There are also differences in collection method, with collection and handling of a hair sample being significantly less invasive, particularly if multiple samples need to be collected during the year. Plaintiffs have not addressed how a reasonable suspicion urinalysis program, as required by the Supreme Judicial Court in Guiney, 411 Mass. 328, 582 N.E.2d 523, would be an adequate substitute for the hair drug testing, which is done on a regular, yearly basis. Plaintiffs have presented no evidence of the relative costs of a the comparable urinalysis program. While they note that a single urine test is less expensive than a single hair test, there is no evidence about the number of urine tests that would be necessary to provide evidence that officers are drug free over a larger period of time like that tested through hair samples. The BPD has presented evidence that urinalysis is less sensitive in detecting drugs than the hair drug test and the plaintiffs have not presented evidence to the contrary. I find that despite showing the wide acceptance of urinalysis, the plaintiffs have failed to show that urinalysis could, as a practical matter, be an equally valid alternative to the hair test when benefits and burdens are weighed.

2. Hair Testing Plus

Plaintiffs turn next to alternatives that build on use of the hair drug test but incorporate additional safeguards. Plaintiffs' expert Dr. Kidwell proposed an alternative in this vein, suggesting that BPD could require hair tests several times per year on a random basis, and, if an individual tested positive, then place that individual in a frequent, random urinalysis program. He suggested the BPD only take adverse employment action if the urinalysis results were positive. This would be in contrast to the current system, in which the BPD uses urinalysis only once an officer enters a drug rehabilitation program after testing positive on the hair drug test, after admitting drug use agreeing to a suspension. Plaintiffs also suggest that the BPD could consider the results of independent contemporaneous hair, blood and urine tests, as well as medical records, notes from medical doctors, and employment history, to supplement the hair drug test in making any employment decision.

a. Admissibility

At the outset, the BPD argues that Dr. Kidwell's affidavit is inadmissible13 because plaintiffs did not previously disclose him as an expert pursuant to Rule 26(a)(2) and he was therefore not deposed during discovery. At a status conference on October 27, 2014, the plaintiffs explained that they seek to designate Dr. Kidwell, a Naval Research Laboratory Research Scientist, as an expert for trial if the case survives summary judgment. Dr. Kidwell's affidavit had been a part of the record since June 2003, but he had not previously been designated as an expert because plaintiffs were told that he would not be permitted to act in that capacity. Dr. Kidwell recently informed the plaintiffs that he is available to testify. At the status conference, plaintiffs sought admission of Dr. Kidwell as an expert but noted that they were not seeking to amend the summary judgment record because Dr. Kidwell's affidavit was already part of the summary judgment record that was before Judge O'Toole and that went before the First Circuit. While the BPD opposed permitting supplemental expert designations, they expressed no objection to the state of the summary judgment record as it existed before Judge O'Toole and the First Circuit, and did not object or dispute the plaintiffs' claim that Dr. Kidwell's affidavit was already a part of the summary judgment record.

The BPD contends that the plaintiffs' failure to disclose Dr. Kidwell as an expert under Rule 26 requires that his opinion and affidavit be excluded from consideration on summary judgment. The plaintiffs acknowledge that Dr. Kidwell was not disclosed as an expert, but they argue that preclusion is inappropriate. Preclusion is not a "mechanical exercise;" I have discretion to choose a less severe sanction, Esposito v. Home Depot U.S.A., Inc., 590 F.3d 72, 77-78 (1st Cir.2009). I must consider a wide range of factors in making a decision about preclusion, including the history of the litigation, the centrality of the evidence that may be precluded, any justification for late disclosure, prejudice to the opposing party, and any effect upon the court's docket. Id. at 78.

The plaintiffs argue that they were justified in failing to disclose Dr. Kidwell as an expert earlier because they believed he would not be available to testify, and that the BPD has essentially waived any objection it might have had to inclusion of the affidavit in the summary judgment record by not raising it before Judge O'Toole, the First Circuit, or previously before me. The BPD has never previously objected to inclusion of the affidavit in the summary judgment record. Dr. Kidwell's affidavit is undisputedly important to the plaintiffs' case, particularly as it provides an explanation for the mechanics of external cocaine contamination for African-American hair and provides one of the few sources of proposed alternatives in this case.

While the BPD objects that it never received an expert report pursuant to Rule 26(a)(2)(B) of the Federal Rules of Civil Procedure, the plaintiffs did provide in June 2003 the detailed affidavit that is now at issue. The affidavit contains Dr. Kidwell's qualifications, opinion, and sources of information upon which he relied, the elements required in a Rule 26 expert report. Although the defendants did not seek to depose Dr. Kidwell because they thought he would not be testifying at trial, they did have the opportunity over more than a decade to adduce evidence and testimony that counters the opinions of Dr. Kidwell as contained in this affidavit. While the BPD may be somewhat prejudiced by not having had the chance to depose Dr. Kidwell before aspects of his affidavit were drawn more directly into the core of the remaining dispute following the First Circuit's opinion, permitting the BPD's objection at this late stage after they had given every indication over more than ten years that they had no objection to inclusion of his affidavit in the record would prejudice the plaintiffs because they would have no opportunity to remedy any failures of disclosure. Given the length of time that the affidavit has been considered a part of the record in this case, the BPD cannot fairly claim surprise. Considering the totality of the circumstances related to the non-disclosure of Dr. Kidwell as an expert, I conclude that Dr. Kidwell's affidavit should not be precluded from consideration at summary judgment.14

b. Substantive Merit

Upon consideration, I find Dr. Kidwell does not provide statistical or other data about the effect that adopting these modifications would have on the out-comes of drug testing, and the plaintiffs do not point to any other data supporting adoption of these particular suggestions. Nonetheless, since these proposals build on the hair drug test themselves and add additional procedures, any concern that they could lead to a greater racial disparity is largely absent. The EEOC recognizes that in some circumstances, procedures can be combined into a "significantly more valid procedure," 29 C.F.R. § 1607.3(B). The fact that the cumulative alternative relying on additional confirmation would not lead to an even more racially disparate outcome, however, does not provide extensive data or other support for a claim that it would lead to a less disparate outcome.

In addition, just as the plaintiffs have not presented sufficient evidence from which a reasonable jury could conclude that a urinalysis program alone would be equally valid given the trade-offs in terms of costs and invasiveness of testing, similarly any evidence of the effects of using hair testing and urinalysis in combination is lacking from the record. Dr. Kidwell's only assessment of the equal validity of his proposal is a statement that "[t]he cost of such a policy could be even less than the current hair testing program, and yet provide even a better deterrent to drug use." This speculative and unsupported endorsement of a hybrid program of uncertain dimensions does not provide the "extensive data" from which a reasonable jury could conclude that this is an equally valid alternative.

3. Current Hair Test Procedure

The BPD modified Rule 111 as of January 1, 2007, to include additional and more robust safety-net tests and double confirmation of hair testing. See Commission Decision at 111 ("the use of `true safety nets' and `double confirmation' in hair testing was always available and, indeed, that is the procedure now followed by the BPD.") The plaintiffs point to the BPD's adoption of the revised safeguards as evidence that a less discriminatory alternative to the hair drug test as it had been administered since 1999 was available. See, e.g., Adams v. City of Chicago, 469 F.3d 609, 612 (7th Cir.2006)(looking to subsequent procedure as evidence that an alternative was available at an earlier time); Chrisner v. Complete Auto Transit, Inc., 645 F.2d 1251, 1263 (6th Cir.1981)(subsequent practice is relevant to the existence of a viable alternative employment practice).

The only evidence in the record that the 2007 changes had an effect was the statement of Ms. Horowitz, an attorney for the BPD, who testified before the Commission on November 20, 2010, that since the testing procedures were modified in July 2007, there had been only one confirmed positive test for cocaine. Such a significant decrease in the number of positive tests certainly reflects the fact that the modifications provided more protections to officers, but it does not provide information about any tradeoffs that may exist between ability to identify drug users in the police force and the lower risk of false positives due to additional safeguards. The record presents barely any evidence at all about the nature of the modifications and whether the procedure used today is equally valid at identifying drug users. The record does not contain the evidence required to show an equally valid alternative.

4. Refusal to Adopt Alternative Procedures

The existence of an alternative procedure is only sufficient to overcome the burden of summary judgment if the defendants refused to use a less discriminatory alternative. The plaintiffs point to the fact that no alternative was implemented until 2007 as evidence that the BPD refused to implement a proposed alternative, but the plaintiffs do not point to specific factual allegations concerning proposals to the BPD or equally valid alternatives of which the BPD was aware that they refused to implement.

The plaintiffs note that after the racial disparity was first publicized, Commissioner Evans met with the NAACP and the Massachusetts Association of Minority Law Enforcement Officers (MAMLEO) to address concerns about the disparity, and he stated that "it's been my position . . . that if anybody would provide me with a valid scientific study that showed that [the Hair Test] was discriminatory, that the drug testing would cease." Plaintiffs point to that statement as evidence that the BPD's position was that "those questioning the Hair Test had to prove that it was unreliable before he would consider any alternative." The language of the Commissioner's statement, however, suggests a willingness to modify or end the hair testing procedure, not a refusal to do so, if he saw evidence of a discriminatory effect.

In a section of the plaintiffs' additional statement of material facts titled "Alternatives to the Hair Test Proposed or Available," the plaintiffs note the suggestions in Dr. Kidwell's affidavit and the fact that numerous plaintiffs underwent independent tests of various types. No other facts are alleged about offers or proposals of alternatives. The plaintiffs are correct that EEOC regulations require an employer to investigate "suitable alternative selection procedures" when assessing validity, 29 C.F.R. § 1607.3(B), but here the plaintiffs have not pointed to any feasible alternative selection procedures of which the BPD might have been aware.

Despite the fact that the BPD test and Rule 111 were the subject of extensive collective bargaining negotiations and collaboration between the BPD and the Boston Police Patrolman's Association, the plaintiffs do not present any allegation that the union or plaintiffs proposed alternatives that the BPD refused to consider. In Ricci v. DeStefano, 557 U.S. 557, 589, 129 S.Ct. 2658, 174 L.Ed.2d 490 (2009), in considering whether a proposed alternative test was equally valid to one specified in a contract with a union, the Supreme Court stated that where the choice of an employment-related procedure is "the result of a union-negotiated collective-bargaining agreement, we presume the parties negotiated that [procedure] for a rational reason." Similarly here, given that Rule 111 and the testing procedure were subject to extensive discussion with the union and were agreed to by the union,15 there is a presumption that the procedure chosen was chosen for a rational reason. The fact that the BPD ultimately did come to agreement with the union to modify the hair drug testing procedure through the 2007 amendment to Rule 111 also is evidence of a willingness to adopt an alternative practice, albeit slowly and some years after a disparity was first identified. In this context, summary judgment is appropriate for the BPD in the absence of any compelling evidence of the BPD's refusal to consider and adopt an alternative equally valid procedure.

III. CONCLUSION

For the reasons set forth more fully above, it is hereby ORDERED that the Defendant City of Boston's Motion for Summary Judgment (Doc. No. 218) is GRANTED.