*774In yet another case affected by the wrongdoing of former State chemist Annie Dookhan, see generally Commonwealth v. Scott, 467 Mass. 336, 5 N.E.3d 530 (2014) ; Bridgeman v. District Attorney for the Suffolk Dist., 476 Mass. 298, 67 N.E.3d 673 (2017) ( Bridgeman ), we must reverse a defendant's conviction of trafficking in heroin. See G. L. c. 94C, § 32E(c ). In an effort to cure the taint from Dookhan's association with the case as primary chemist, a police officer testified that he performed a field test of the substance seized from the defendant, which proved that the substance was heroin. The testimony was admitted, over the defendant's objection, *775without establishing the scientific reliability of the field test. We conclude that the admission of this evidence was prejudicial *863error and that the defendant is entitled to a new trial.
Background. We recite the basic facts as the jury could have found them, reserving other facts for later discussion. On April 27, 2009, officers of the Boston police department's drug control unit went to the housing development where the defendant, Juan Carlos Rodriguez, lived to execute three search warrants: one for the defendant's apartment, one for his motor vehicle, and one for his person. Once inside the defendant's apartment, the officers used a key recovered from the defendant's motor vehicle to open a locked bedroom door. In the bedroom's closet, the police found a total of $13,270, a digital scale, and a small pouch that contained nine individually wrapped packages, or "fingers,"1 of a substance that resembled sidewalk chalk. A search of the defendant's person yielded two similar packages.
Officer Robert England took the eleven packages to the police station and conducted a field test using a NarcoPouch 924 test kit manufactured by Safariland. The NarcoPouch 924 test kit is a small, sealed rubber pouch that contains three glass vials filled with chemical solutions. England unsealed the NarcoPouch 924 test kit, placed a small amount of the chalky substance inside the pouch, resealed it, and began "popping" the vials so that the unknown substance interacted with the chemical solutions. He testified, "I field-tested these drugs[2 ] and the preliminary result came back to me. It showed green to me. We believe it was [h]eroin."
The eleven packages were sent to the William A. Hinton State Laboratory Institute in Jamaica Plain (Hinton lab) in April, 2009, for testing. As the primary chemist assigned to the case, Annie Dookhan "received [the packages] from the evidence office ... checked it and [did] all the preliminary testing, which included doing the net weight, doing color tests, [and] perhaps ... other kinds of testing." Della Saunders, the confirmatory chemist, received eleven vials prepared by Dookhan and tested them, concluding that "they were positive for the presence of heroin." Both *776Dookhan and Saunders certified that the packages seized from the defendant's closet and person contained heroin.
In 2013, Dookhan pleaded guilty to twenty-seven counts of criminal misconduct, including tampering with evidence, perjury, and obstruction of justice. The trial judge permitted the defendant substantial leeway in introducing evidence concerning Dookhan's arrest and prosecution for her criminal conduct at the Hinton lab, including the transcript of her guilty plea colloquy,3 testimony from three of Dookhan's coworkers about her misconduct at the lab,4 and State police Captain *864Robert M. Irwin's testimony about the criminal investigation of Dookhan's conduct.
After Dookhan's wrongdoing came to light, the Commonwealth sent the eleven packages seized from the defendant to the laboratory at the Massachusetts State police forensic services group for retesting. Sarah Clark, a chemist at this laboratory, retested the substances in the packages and concluded that they contained heroin.
At trial, the defendant argued that Dookhan's participation irrevocably damaged the Commonwealth's case, and specifically that the Commonwealth could not meet its burden of proving that the packages the Boston police seized from the defendant contained heroin before Dookhan gained access to them. The Commonwealth combatted this defense on two grounds: first, that there was no direct evidence that Dookhan altered the evidence in this case5 and second, that the field test and circumstantial evidence *777proved "it was heroin on that day and it is heroin today." In this vein, the prosecutor argued:
"How do you know this is heroin? You know it because Officer England came before you and told you right after they seized this, back at the station he performed a field test.... And on that date, what happened? It showed the presence of heroin in these drugs. Ladies and gentlemen, it was heroin on that date, and it is still heroin."
Discussion. 1. Field test evidence. a. Scientific reliability of field test results. The defendant contends that the judge erred by allowing the Commonwealth to introduce the results of the NarcoPouch 924 field test for heroin without demonstrating the test's scientific reliability under Commonwealth v. Lanigan, 419 Mass. 15, 26, 641 N.E.2d 1342 (1994). We agree.
The issue arose as follows. Jury selection began on a Friday. On that day the defendant filed a motion in limine "to preclude the Commonwealth from introducing, or referring to, purported field test evidence that the purported drugs in this case are heroin" on the ground that "this opinion does not meet the Daubert [6 ]/ Lanigan, standard because this evidence is not scientifically valid." The judge reserved decision until the following Monday, when she denied the motion, requiring the Commonwealth only to "provide an adequate foundation that the field tester is qualified to do that," but not to demonstrate the scientific reliability of the test.7 In this *865regard, England, a twenty-five year veteran police officer, testified that he was "introduced to NarcoPouch" at a two-week course with the United States Drug Enforcement Administration, where he was "trained and certified," and that he had "recertified with NarcoPouch" since then. He testified that he *778used NarcoPouch "a lot,"8 that "several people" with the Boston police use it too, and that he is often called to crime scenes to perform the test when a trained officer is needed.
Even with a qualified officer testifying, field test evidence may not be offered without a demonstration of its validity or reliability under Lanigan. "[T]o date, no appellate case from Massachusetts has accepted as reliable field test results, regardless of the purposes for which they are offered. Until that occurs, field tests offered to prove the identity of a substance, 'presumptive' or otherwise, must be evaluated according to one of the methods approved in Lanigan." Commonwealth v. Fernandez, 458 Mass. 137, 151 n.20, 934 N.E.2d 810 (2010).9 No Massachusetts appellate case since Fernandez has accepted any field test, including the NarcoPouch 924 field test for heroin, as reliable.
In Fernandez, the Commonwealth introduced evidence that police officers conducted weekly field tests of the residue of plastic bags pulled from the defendant's trash with "Scott-Reagent tests" for cocaine manufactured by NIK Public Safety, Inc. Id. at 139 & n.4, 934 N.E.2d 810. "All the field tests ... returned positive results for the presence of cocaine." Id. at 139, 934 N.E.2d 810. The defendant argued that the judge erred by failing to subject this evidence to a Lanigan analysis. Id. at 147-148, 934 N.E.2d 810. The court rejected this claim because "[t]he judge did perform such an analysis, albeit a limited one" (emphasis supplied). Id. at 148, 934 N.E.2d 810. Rather, the court asked "whether the Lanigan analysis that was performed was adequate." Ibid.
The court held that "the Commonwealth's relatively weak proffer," id. at 150, 934 N.E.2d 810,10 was sufficient in *866Fernandez for two reasons. First, although the Commonwealth "alerted [the defendant] about *779the field testing and how it was performed" several months before trial, the defendant "waited until the eleventh hour"-the first day of trial-"to raise the Lanigan issue." Id. at 147-148, 149, 934 N.E.2d 810. The late notice prevented the Commonwealth from calling "a scientist from the field test manufacturer to testify as to its reliability." Id. at 150, 934 N.E.2d 810. Second, the tests were never presented to the jury as being conclusive. Ibid. "To the contrary, the 'presumptive' nature of the tests was emphasized at every turn." Ibid. The judge instructed the jury twice that a field test "may or may not yield a valid result," that no field test exists "that will completely eliminate the occurrence of an invalid test result," and that "[a] forensic laboratory test is necessary to identify the nature of an unknown substance completely." Ibid.
The Fernandez decision does not excuse the complete absence of Lanigan screening in this case. It is true that defense counsel in this case, as in Fernandez, did not file his motion in limine until the first day of trial, prior to jury selection. The Commonwealth argued, and the judge agreed, that its future reliance on the field test was evident from the grand jury minutes, in which England testified and described his use of the "Narco Test Kit." The defendant replied that England's testimony before the grand jury did not put him on notice that the Commonwealth would employ England as an expert on the scientific validity of the field test.11 In any event, the defendant's late assertion of the Lanigan issue might have justified a limited Lanigan analysis, but not its *780complete abandonment.
Finally, unlike in Fernandez, the field test evidence was used without restriction and with no instruction on its presumptive nature or the possibility of false positives. On cross-examination, England did concede that a field test is a "preliminary" test, as distinguished from a "scientific" test performed at a drug testing laboratory. However, in closing argument the Commonwealth characterized the field test as conclusive evidence that "it was heroin on that date, and it is still heroin."12 Allowing the field test to be used for this purpose without prior scrutiny under the Lanigan protocol was error.
b. Prejudice. We further conclude that the error entitles the defendant to a new trial. Because the defendant preserved this issue with a timely objection, "we review the proceedings below for prejudicial error." Commonwealth v. Cruz, 445 Mass. 589, 591, 839 N.E.2d 324 (2005). "An error is not prejudicial if it 'did not influence the jury, or had but very slight effect.' " Ibid., quoting from Commonwealth v. Flebotte, 417 Mass. 348, 353, 630 N.E.2d 265 (1994). "[I]t is not enough for the *867Commonwealth to demonstrate that its other, properly admitted evidence was 'sufficient' to convict the defendant or that the inadmissible evidence was 'consistent' with the admissible evidence." Commonwealth v. Tyree, 455 Mass. 676, 701, 919 N.E.2d 660 (2010), quoting from Commonwealth v. Dagraca, 447 Mass. 546, 554-555, 854 N.E.2d 1249 (2006).
In determining whether an error is prejudicial "we examine various factors, including the importance of the evidence in the prosecution's case; the relationship between the evidence and the premise of the defense; who introduced the issue at trial; the frequency of the reference; whether the erroneously admitted evidence was merely cumulative of properly admitted evidence; the availability or effect of curative instructions; and the weight or quantum of evidence of guilt." Dagraca, supra at 553, 854 N.E.2d 1249.13 Here, *781the defense was based on casting doubt on the composition of the substance before Dookhan obtained access to it, and the Commonwealth introduced the field test result specifically to undermine this defense. The erroneously admitted evidence was unique in establishing the nature of the substance and was used for its full probative force without any curative instruction. Contrast Fernandez, 458 Mass. at 150, 153, 934 N.E.2d 810 (presumptive nature of field test evidence explained to jury and described by judge as "only 'incremental' evidence"). As the other evidence regarding the composition of the substance before it reached the Hinton lab was not overwhelming, the error was prejudicial.
Aside from the field test result, the Commonwealth presented little direct or circumstantial evidence of the composition of the drugs before they were sent to the Hinton lab. In some cases, a police officer with expertise in identifying drugs or a defendant's statements and course of conduct can be used to establish that a particular substance is a drug. See, e.g., Commonwealth v. Connolly, 454 Mass. 808, 812, 830-831, 913 N.E.2d 356 (2009) (three experienced officers' testimony that substance either appeared to be or was identified by a field test as cocaine, defendant's identification of substance as "crack" cocaine during two controlled purchases, and canine sniff contributed to finding that reliance on drug certificates in violation of Melendez-Diaz v. Massachusetts, 557 U.S. 305, 129 S.Ct. 2527, 174 L.Ed.2d 314 [2009], was harmless beyond a reasonable doubt); Commonwealth v. Marte, 84 Mass. App. Ct. 136, 140, 143, 993 N.E.2d 1201 (2013) (defendant's statement that "you never have to worry about my package, I've been around a long time," coupled *868with evidence of four controlled buys and field testing following each buy, contributed to finding Melendez-Diaz error harmless beyond a reasonable doubt as to drug convictions resulting from the four controlled buys).
Here, there were no controlled purchases and no incriminating statements by the defendant. England professed no ability to recognize drugs without conducting a field test ("I'm not a user so I field test everything"). Sergeant Detective William J. Feeney, *782who was the Commonwealth's expert on distribution and packaging of drugs, opined that "possession of over [one hundred] grams of [h]eroin in finger form" was consistent with possession with the intent to distribute. When asked, "[W]hat do you recognize that tan powder to be packaged like?" he responded, "It's consistent with fingers of [h]eroin, based on its color, its texture." This unsolicited remark regarding the color and texture of the substance, from an officer not shown to be an expert in identifying drugs, carries little weight. See Commonwealth v. Nelson, 460 Mass. 564, 577, 953 N.E.2d 164 (2011) ( Melendez-Diaz error not harmless where officer "was called as an expert concerning the charge of intent to distribute and thus was never asked directly to identify the marijuana"); Commonwealth v. King, 461 Mass. 354, 358, 960 N.E.2d 894 (2012) (officer "not asked about his specific training in the identification of cocaine").
The evidence that cash and a scale were found in the defendant's closet, that the substance was packaged in "finger" form, and that defendant fled while awaiting trial does not overcome the prejudice from the Commonwealth's reliance on the field test result. See King, supra at 360, 960 N.E.2d 894 ("[E]vidence [that] leads to the conclusion that the defendant took part in what appeared to be a drug transaction ... does not go to whether the substance was, in fact, cocaine"). "This is not a case where the facts independent of the [field test result] overwhelmingly prove the nature of the substances ... recovered from the defendant's apartment." Commonwealth v. Vasquez, 456 Mass. 350, 367, 923 N.E.2d 524 (2010). The improper use of the field test result clearly prejudiced the defendant, and we cannot say with fair assurance that the error had but slight effect.
2. Limitations on defense evidence. The defendant contends the judge erred by not allowing Irwin to testify that Dookhan had a key to the evidence safe at the Hinton lab and in denying his request to introduce copies of Dookhan's indictments in evidence. Because these evidentiary issues might recur at any new trial, we comment briefly.14
From conducting the investigation of Dookhan's criminal activity, Irwin had personal knowledge that Dookhan possessed a key that opened the locked door to the evidence safe. The judge disallowed Irwin's testimony about Dookhan's possession of the key under the erroneous premise that what defendant was seeking *783to introduce through Irwin was a statement against penal interest attributed to Dookhan. Irwin's testimony should not have been excluded on this ground. However, the judge did permit the defendant to present ample evidence to suggest that Dookhan had free access to evidence stored at the Hinton lab, and it was obvious that Dookahn, as primary chemist, had access to the eleven "fingers" seized from the defendant. "A trial judge has discretion to exclude evidence that would be merely cumulative of evidence already admitted." *869Commonwealth v. Urrea, 443 Mass. 530, 544, 822 N.E.2d 1192 (2005). At any retrial, Irwin's testimony regarding the key should be reviewed in this light.
The judge did not abuse her discretion in excluding the indictments from evidence. Indictments have no probative value or evidentiary significance. See Commonwealth v. Kelley, 33 Mass. App. Ct. 934, 935, 600 N.E.2d 603 (1992) ; Commonwealth v. Johnson, 43 Mass. App. Ct. 509, 513-515, 684 N.E.2d 627 (1997). The defendant was permitted to introduce substantial evidence of Dookhan's misconduct, including the colloquy in which she pleaded guilty to the crimes charged in the excluded indictments. The indictments were not necessary to prove the time frame of Dookhan's misconduct, which the defendant was able to establish through Irwin's testimony that it dated back as far as 2008 and 2009, and by the colloquy itself, which discussed specific misconduct in June, 2011.
Conclusion. The judgment is reversed and the verdict is set aside.
So ordered.
Sergeant Detective William J. Feeney testified that a "finger" is a quantity of heroin purchased by mid-level dealers, and that the name is derived from the practice of packaging approximately ten grams of heroin inside the finger of a latex glove, tying it off in a knot, and then cutting off the finger.
On cross-examination, defense counsel established that England tested only one of the eleven packages.
At the plea colloquy, an assistant attorney general recited the following evidence of Dookhan's misconduct: improperly removing ninety drug samples from the evidence safe at Hinton lab, forging the signature of an evidence officer, specific instances of tampering with the testing of drug vials, submitting a discovery packet to a prosecutor that contained an altered test, and lying about her qualifications.
Daniel Renczkowski testified that he observed instances where Dookhan's laboratory bench practices subjected her samples to cross-contamination. Nicole Medina and Renczkowski testified that someone forged their initials on laboratory documents. Medina also testified that she saw Dookhan using a computer in a restricted area of the laboratory, and Peter Piro once observed Dookhan mishandling balances and scales.
The Supreme Judicial Court has since observed that it " 'may be impossible' for any defendant to prove that the drug analysis in his or her case was tainted by [Dookhan's] misconduct" because "even if Dookhan herself were to testify in each of the thousands of cases in which she served as primary or secondary chemist, it is unlikely that her testimony, even if truthful, could resolve the question whether she engaged in misconduct in a particular case." Bridgeman, 476 Mass. at 305, 67 N.E.3d 673, quoting from Scott, 467 Mass. at 351-352, 5 N.E.3d 530.
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993).
In Commonwealth v. Marte, 84 Mass. App. Ct. 136, 141, 993 N.E.2d 1201 (2013), we noted that since 2009, "the evidentiary standards for properly admitted field tests are heightened," and that the requisite foundation for admission of field test evidence "includes the experience of the officer(s) conducting the test, the methodology of the testing, and definitive identification of the substance, i.e., as cocaine, heroin, or another controlled substance." In Marte, however, we did not address the issue of scientific reliability presented in Commonwealth v. Fernandez, 458 Mass. 137, 149-150, 934 N.E.2d 810 (2010), stating that "that issue is not present in this appeal, there having been no Lanigan challenge raised below." Marte, supra at 143 n.6, 993 N.E.2d 1201.
When asked how many times he had used a NarcoPouch test over the course of his career, he answered, "I mean a thousand is a lot, but I'd say-[I] use it all the time.... I field test everything."
In both Fernandez, supra at 149 n.17, 934 N.E.2d 810, and Commonwealth v. Vasquez, 456 Mass. 350, 364 n.15, 923 N.E.2d 524 (2010), the Supreme Judicial Court cited the National Research Council, Strengthening Forensic Science in the United States, A Path Forward, 134-135 (2009), for the proposition that forensic drug testing usually involves "presumptive identification" by a field test, followed by definitive identification using gas chromatography-mass spectrometry. Neither of these cases accepted the National Research Council's report as establishing the scientific reliability of field tests. See State v. Martinez, 143 Conn. App. 541, 563 & n.8, 69 A.3d 975 (2013), reversed on other grounds, 319 Conn. 712, 127 A.3d 164 (2015) (noting "that the leading treatises on scientific evidence offer no clear consensus on the reliability of field tests").
The judge declined to hold a full evidentiary hearing, but required the Commonwealth to provide "some foundation" for the field tests' reliability. Fernandez, supra at 148, 934 N.E.2d 810. The Commonwealth offered "a case from the Criminal Court of the City of New York, People v. McIntyre, 185 Misc. 2d 58, 711 N.Y.S.2d 317 (N.Y. Crim. Ct. 2000), and a regulation applicable to the Department of Correctional Services of the State of New York," both concerning "the general acceptance of the field tests ... for obtaining presumptive results." Ibid. These sources discussed the reliability of the brand and manufacturer of the field tests used in Fernandez's case, which employs a chemically treated swab that turns from pink to blue to indicate the presence of cocaine. Id. at 139, 148, 934 N.E.2d 810. Here, the Commonwealth offered no information about the NarcoPouch 924 field test for heroin, the broken vial technology, or the manufacturer's quality control process.
The defendant further contends that the Commonwealth violated its automatic discovery obligations by failing to disclose England as an expert witness. See Mass.R.Crim.P. 14(a)(1)(A)(vi), as amended, 444 Mass. 1501 (2005). Defense counsel represented that he first learned that the Commonwealth intended to offer England as expert through a motion in limine the Commonwealth had filed "several days" before trial. The prosecutor responded, "For the record, I never even filed a motion in limine." We discern no discovery violation; it is apparent that the Commonwealth did not intend to rely on England to offer expert opinion evidence on the scientific validity of the NarcoPouch 924 test.
The prosecutor was free to make this argument, which was firmly rooted in the evidence admitted at trial.
Dagraca, supra at 552-553, 854 N.E.2d 1249, applies the harmless beyond a reasonable doubt standard, derived from Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967), applicable to errors of constitutional dimension, while we apply the less stringent prejudicial error standard, derived from Kotteakos v. United States, 328 U.S. 750, 764-765, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946), used to evaluate evidentiary errors. See Flebotte, supra. The difference between the two standards is not of kind but of degree. See Brecht v. Abrahamson, 507 U.S. 619, 641-642, 113 S.Ct. 1710, 123 L.Ed.2d 353 (1993) (Stevens, J., concurring) (discussing development and application of prejudicial error standard). Compare Tyree, supra (to assess harmlessness under Chapman standard we review "the totality of the record before us, weighing the properly admitted and the improperly admitted evidence together" to determine whether "we are satisfied beyond a reasonable doubt that the tainted evidence did not have an effect on the jury and did not contribute to the jury's verdicts") with Flebotte, supra (assessment of prejudicial error involves "pondering all that happened without stripping the erroneous action from the whole" to determine whether the judgment "was not substantially swayed by the error").
We need not address the defendant's claim regarding the prosecutor's exercise of a peremptory challenge as the unique facts underlying this claim are unlikely to repeat.