Spina, J.
The defendant, Jeffrey Bly, was convicted of deliberately premeditated murder. On appeal, Bly asserts that (1) the trial judge wrongly admitted certain deoxyribonucleic acid (DNA) test result evidence; (2) DNA evidence collected from a water bottle and cigarette butts used by Bly should have been suppressed; (3) Ely’s statement to the police, obtained without Miranda warnings, should have been suppressed; (4) identification testimony by one of the Commonwealth’s witnesses should have been suppressed; (5) Ely’s expert on identification should have been permitted to testify before the jury; (6) evidence of Ely’s refusal to provide a court-ordered hair sample should not have been heard by the jury; (7) the judge wrongly admitted testimony regarding prior bad acts by Bly; and (8) the medical records of one of the Commonwealth’s witnesses were not produced to Bly in a timely manner or found to be privileged in a written decision by the judge. Bly also contends that we should exercise our power under G. L. c. 278, § 33E, to reduce the verdict or order a new trial. We affirm.
[475]*4751. Background. The jury could have found the following. Just before 7 p.m. on September 25, 1995, Assistant Attorney General Paul McLaughlin, who was on special assignment as a prosecutor attached to the Suffolk County district attorney’s office, was shot and killed at the Highland Street commuter rail station (Highland station) in the West Roxbury section of Boston while on his way home from work. McLaughlin was scheduled to begin the trial of Bly the following morning on charges of carjacking.1 The carjacking charges stemmed from an incident that occurred in April, 1994, between Bly and Dana Alston. Despite being the target of significant intimidation, Alston was prepared to testify against Bly. Bly was a member of a group of young men from the Theodore Street area in the Mattapan section of Boston who engaged in drug trafficking during the time period prior to McLaughlin’s murder. About one week prior to the scheduled carjacking trial, Bly told Eric Hardy, a member of the Theodore Street group, that he was concerned about the imminent trial, stating, “I got to do something,” but that he could not “mess with Dana [Alston].” Bly then exclaimed, “I got it,” and took his friend Ricardo Gittens aside for a private conversation.
On September 22, 1995, Bly and Anthony Houston, a younger member of the Theodore Street group who regarded Bly as a father figure, went to the vicinity of Washington and State Streets in Boston. Bly waited with Houston until he saw McLaughlin leaving work, followed him to South Station, then ordered Houston to follow McLaughlin to “find out where he goes, does he meet up with anybody, where he gets off at, anything in that nature.” Bly gave Houston money for train fare, as well as a pen and a notebook. Houston boarded the train behind McLaughlin, disembarked at the Highland station, followed McLaughlin to his car, had a brief exchange of words, and wrote down McLaughlin’s registration plate number before returning to Bly to share his findings.
On the afternoon of September 25, 1995, Sandra Brown overheard a conversation between Bly and several of his friends. During this conversation, Bly stated that he had to get Mc[476]*476Laughlin and that it had to be done that night. Bly was wearing a black hooded sweatshirt, black jeans, and black sneakers that day.
Mario Thompson, one of Ely’s friends from the neighborhood, owned a 1982 Buick automobile. On the afternoon of September 25, 1995, Bly asked Thompson to follow him in his car. Thompson followed Bly and Gittens, who was driving a gold Dodge Colt automobile, to the Highland station and parked in a lot across the street. Once parked, Bly left the Dodge and got into the Buick for a short conversation. Bly told Thompson that he could leave because he was not needed.
Maureen Woodsum, the only identification witness who did not know Bly, was driving home at around 6 p.m. on the evening of McLaughlin’s murder. As she was turning toward her home she saw a man she later identified as Bly standing on a bridge close to the Highland station wearing dark pants and a dark sweatshirt with the hood pulled up over a dark cap. When she turned her car, Woodsum came within about ten feet of the man and saw his face as he started to walk toward her. Woodsum thought that the man looked out of place, but she chose not to make any report of what she saw, even after learning of McLaughlin’s murder, because she did not observe him do anything wrong and did not want to be perceived as racist. Woodsum next saw that same man in February, 1998, while watching a television news broadcast, appearing in court next to a man with whom she had gone to high school. Only after Woodsum identified the man on television as the same man from the bridge did she learn that the broadcast concerned the indictment of Bly for McLaughlin’s murder.
Michael Duffy was a commuter who parked about thirty yards behind McLaughlin’s car at the Highland station on September 25, 1995. His train arrived at the station at about 6:50 p.m. After he alighted from the train, he saw McLaughlin walking to his car. As Duffy was getting into his car, he heard a brief exchange of words and then two shots. He looked up to see an unidentified figure leaning into the driver’s side of McLaughlin’s car, and McLaughlin’s legs hanging out through the open driver’s door. Duffy observed the figure put an object into his waistband and flee into a parking lot toward the train tracks. Duffy [477]*477described the assailant as a fourteen to sixteen year old black male wearing a black hooded sweatshirt and baggy jeans, but did not specifically identify Bly as the person he saw that night. Two other witnesses in a nearby parking lot saw an unidentifiable person in a black sweatshirt fleeing the scene of the shooting.
On the night of September 25, 1995, Janet Brown was at her daughter Sandra’s apartment when she heard Bly state: “I don’t have to worry no more, I took care of my problem.” Bly elaborated that he hid in the bushes and then came out when he saw McLaughlin, shooting him two or three times. When she asked Bly why he did that, he responded, that he did not want to go back to prison and that “D.A. McLaughlin was going to give him a lot of time because he had been up against him before and . . . lost both cases.” Bly further stated, “I know he’s going to try to hang me, but I’m not going out like that.” Bly later asked her for money so that he could leave town.
Sandra Brown saw Bly during the evening of McLaughlin’s murder at Mattie Dailey’s apartment.2 At one point he was soaking his hands in bleach. She noticed that he was not wearing his black hooded sweatshirt, as he had been earlier that day. She heard Bly say that he had no more worries and that he did not have a prosecutor.
Bly told Eric Hardy the day after the murder that he killed McLaughlin because “he was going to give [Bly] life on a carjacking case” and that Bly “didn’t want to get life.” Bly recounted the entire crime to Hardy, explaining that he hid in the bushes near McLaughlin’s car waiting for him to arrive and that he had shot McLaughlin twice, and then fled to the waiting car of Gittens, losing his sweatshirt and bandana as he ran. Bly later asked Hardy to get rid of Gittens’s car, but Hardy was stopped by the police on his way to accomplish that task and the car was impounded. Bly threatened to kill Hardy if he “snitch[ed].”
After recognizing McLaughlin from television news reports about the murder as the man whom Bly asked him to follow, Anthony Houston testified that he “took it kind of hard.” He [478]*478saw Bly the following day and asked him about the murder. Bly told him, “don’t worry about it, don’t say nothing and if anybody asks, [Bly] never asked [Houston] to do anything like [follow McLaughlin], never said anything about it.” About one month after the shooting, Bly asked Houston, “You wouldn’t turn me over for [$30,000]? . . . You wouldn’t try to snitch or anything like that against me?” Houston assured Bly that he would not because he was loyal.
Mario Thompson approached Bly regarding the McLaughlin murder after recognizing televised reports of the murder scene as the area that Bly had led him to that night. Bly told him to tell anyone who asked that they were together that night. When Thompson pressed further, Bly “gave [Thompson] a little dope slap and he just said ‘do what I tell you.’ ”
Police searching the area around the Highland station found several articles of clothing. Along the railroad tracks, police officers discovered a black hooded sweatshirt, a left-hand brown glove, and a green bandana. The matching glove and a knit cap were discovered inside the sweatshirt. Investigators also recovered a bullet from the front passenger side floor of McLaughlin’s car. Additionally, police responding to the scene discovered an area directly across from McLaughlin’s car that was concealed by bushes. Investigation of this area revealed footprints, as well as a milk crate and a cigarette butt. The footprints at the scene could not be matched to any particular brand or manufacturer of footwear.
A rusty gun was found in May, 1998, in a neighborhood near the Highland station. Testing of the gun yielded inconclusive results. A fingerprint found on the gun did not belong either to Bly or any of the other known fingerprint samples in the case.
2. DNA evidence. The Commonwealth sent various items of clothing and blood samples to three separate laboratories for DNA testing. Bly forcefully attacked the findings and methodology of each. Bly also offered his own expert who testified as to the flaws in the Commonwealth’s DNA evidence.
The first DNA testing in the case was performed by CBR (Center for Blood Research) Laboratories. CBR examined various blood stains at six different genetic markers and found a match between the blood found on the garments and McLaugh[479]*479lin’s blood found at the scene. The likelihood of a random person having this type of profile was estimated to be between one in 12,000 and one in 19,000, dependent on the race used for comparison.
Having confirmed that the garments were linked to the murder, additional tests were conducted in an effort to determine a profile of the wearer. Based on their testing of “friction areas,” where the wearer of a garment potentially can leave skin cells at points where it rubs against the body, CBR found DNA material in the collar and the cuff of the right sleeve of the sweatshirt that were consistent with each other. CBR determined that the samples contained a mixture of DNA from one or more persons, and submitted its report. At that time, CBR did not have a known sample from Bly with which to compare their findings. Dr. David Bing, the former director of CBR’s laboratories, testified that after reviewing Ely’s genetic profile just prior to trial, Bly could not be excluded as a donor of the samples recovered from the sweatshirt. Further, based on a statistical analysis, Dr. Bing opined that the DNA profile found on the sweatshirt would occur randomly in one out of 800 to one out of 400 African-Americans.
On cross-examination, Dr. Bing conceded that he had been asked to go beyond their normal laboratory protocol to determine if there was any detectable DNA material whatsoever on the crime scene garments, regardless of whether the quantity was sufficient to meet the threshold for testing. He also acknowledged that a similar practice by CBR resulted in their test results being excluded in a case.
After receiving CBR’s report, the crime scene evidence was sent to the Federal Bureau of Investigation (FBI) for analysis. By that time, the Commonwealth had obtained known DNA samples from both Bly and two other possible suspects, which also were sent to the FBI.3 The FBI tested the items they [480]*480received using a method known as polymerase chain reaction (PCR).
The FBI determined that the DNA pattern for blood on the bandanna matched that of McLaughlin such that only one in 2.3 million Caucasians would have that profile. The left glove also had visible blood stains that were tested by the FBI, but those yielded inconclusive results from mixture samples with potentially multiple donors. FBI testers found a small piece of human biological material caught inside the left glove, and determined that Bly was the potential major contributor to that DNA, although that too had a mixed pattern. The likelihood of finding such a pattern at random in the African-American population was one in 480. The right glove also had very small bloodstains, visible only with the use of enhancement equipment. These stains did not yield strong results, but neither McLaughlin nor Bly could be excluded from the mixed DNA patterns discovered there with random match probabilities of one in eighty-five for African-Americans and one in fifty-five for Caucasians.
The FBI also determined that McLaughlin was a potential donor of a bloodstain on the front of the sweatshirt with the same statistical likelihood as the blood on the bandanna. Bly could not be excluded as a donor to a mixed stain found in the collar of the sweatshirt, but the other potential suspects in the case were excluded. The FBI testing, however, also indicated that there was some other potential donor to that sample that they were unable to type. Similar results were obtained from friction areas of the knit cap. The Commonwealth did not offer the statistical prevalence of the pattern present in the friction sample from the collar of the sweatshirt or the knit cap.
[481]*481In November, 1998, the Commonwealth sent the crime scene clothing to a third laboratory, Bode Technology Group (Bode), whose representative, Dr. Kevin McElfresh, testified for the Commonwealth. Bode utilized a type of PCR-based testing known as short tandem repeat (STR) testing. Bode performed two different sets of testing. First, using DNA extracted from the collar of the sweatshirt, Bode performed STR analysis in December, 1998 (December testing). This study returned interpretable results at five of the eight genetic locations that were tested. On the basis of these results, Dr. McElfresh offered the opinion that the DNA pattern from the samples found on the sweatshirt matched Bly and appeared at random at a rate of one in 631,000 in the African-American population. The statistical opinion arising from the five genetic loci identified during the December testing constituted the extent of Dr. McElfresh’s testimony on direct examination concerning the sweatshirt sample.
Cross-examination of Dr. McElfresh elicited testimony that a second set of testing performed by Bode in January, 1999 (January testing), yielded results at all eight genetic locations. The January testing suffered from two distinct methodological weaknesses. First, there was evidence that the January testing showed extra fluorescence that potentially indicated contamination in the process used to amplify the DNA. Dr. McElfresh interpreted the results of these tests as containing “artifact,” meaning the false indication of an allele that was not present, rather than contamination that would affect the results. Dr. McElfresh also conceded that his laboratory protocol called for the reextraction and new testing of samples that show evidence of contamination.
Additionally, the January testing was subjected to a process known as “reamplification.” In order to observe accurately the pattern contained in a given sample, the STR method of testing requires amplification, a process by which enzymes are used to stimulate the DNA to replicate itself. Reamplification is a repetition of this initial process that aims to identify patterns from a very small amount of original DNA. According to Ely’s expert, reamplification is not scientifically reliable because it increases the likelihood that contamination will influence the results, and [482]*482its necessity indicates that the sample of DNA being tested is too small. Dr. McElfresh did not defend the process of reamplification as scientifically reliable — reamplification was not a part of Bode’s protocol. Rather, Dr. McElfresh testified that the reamplified results served only to confirm the December test results.
3. DNA issues. Prior to trial, Bly made a motion in limine seeking to exclude all evidence and testimony concerning the DNA testing conducted by Bode. After a hearing, the judge denied the motion, finding that flaws in the Bode testing raised by Bly went to the weight, rather than the admissibility, of the evidence.
Bly now challenges the Bode evidence on three grounds. First, Bly argues that the Bode evidence was scientifically unreliable because it was founded on reamplified DNA. Second, he contends that Bode’s testing was undermined by contamination in the control materials. Last, Bly complains that the database used by Bode to reach its statistical conclusions was not made wholly available to the defense and the portion that was made available showed it to be invalid.
(a) Reamplification and contamination. Consideration of the reliability of expert scientific evidence requires a fact-based determination that encompasses questions of credibility. See Commonwealth v. Gaynor, 443 Mass. 245, 264 (2005). This task prompts the judge to review the specific testing performed for its conformance with scientific norms, and to then decide whether the expert’s conclusions are sufficiently reliable. Id. We review the judge’s decision under an abuse of discretion standard. Id.
The dispute here arose during trial and revolves around which tests were relied on by the expert, rather than whether a particular test was unreliable. The Commonwealth does not dispute that the January testing was flawed and unreliable. Instead, the Commonwealth contends that Dr. McElfresh’s opinion was based on the December testing alone, and that it did not depend on the faulty January testing. The Commonwealth further points out that the only testimony concerning the January testing was elicited on cross-examination. Bly does [483]*483not challenge the validity of the December testing.4
5Rather, he points to repeated instances during the trial where Dr. Mc-Elfresh testified on cross-examination that he relied on all of the information in Bode’s file, essentially adopting the faulty January testing in order to reach his ultimate conclusion in the case.
The direct examination of Dr. McElfresh at trial does not refer specifically to any of the testing dates as the source of his opinion. His direct examination testimony refers only to tests on the collar of the sweatshirt that returned results at five of eight possible genetic loci and a resulting match frequency of one in 631,000. Subsequent testimony showed that the results obtained at five of the genetic loci were from the single-amplification December testing and that the January testing, which produced results at eight genetic loci, merely confirmed those results. Moreover, counsel for the Commonwealth represented to the court that the flawed January testing produced a match frequency between the sweatshirt sample and Bly of one in 1.4 billion. The absence of this higher match frequency rate from McElfresh’s trial testimony resolves any confusion produced by his cross-examination and makes clear that the opinion expressed during his direct examination was limited to the results of the December testing.6
There was no abuse of discretion in the admission of Dr. Mc-Elfresh’s testimony at trial (or denying the motion to strike that [484]*484testimony), where the basis for the expert opinion, the December testing, was reliable, as determined during the pretrial proceedings. This is not an instance where cross-examination produced a repudiation of direct testimony, such that only one version of events could be believed. See Sullivan v. Boston Elevated Ry., 224 Mass. 405, 406-407 (1916). Rather, whatever incongruity arose from cross-examination at trial went to the weight of the testimony. Id. at 406.
Further, Bly complains that Dr. McElfresh’s trial testimony contradicts his pretrial indication that any apparent reamplification in the reporting was due to typographical error. These contradictions, to the extent that they exist, are attributable to the January testing, however.6 While Bly was permitted to raise on cross-examination the flaws that existed in the January testing in an effort to diminish the over-all weight of Dr. Mc-Elfresh’s testimony, such a contradiction does not warrant the exclusion of separate and otherwise reliable results obtained from the December tests.
(b) Bode’s database. Ely’s remaining complaint with the DNA testing is that his pretrial motion in limine seeking to exclude the statistical results of Bode’s database comparison was wrongly decided. Although there was evidence that Dr. McElfresh’s final statistical conclusions utilized two separate databases, the defendant disputes only the use of Bode’s own database on appeal.
Bly challenges the use of Bode’s database in two respects. First, he argues that the database produced unreliable match frequencies. Bly complains that his access to the underlying documentation of Bode’s database was restricted by the labora[485]*485tory and the limited sample he was permitted to review showed an unacceptable differential in probability distribution when compared to the Promega database. He further contends that the Bode database was unreliable because it has not been published in a peer review journal. Second, Bly argues that the underlying data constituted potentially exculpatory evidence and the Commonwealth’s purported withholding constitutes a failure to uphold his constitutional rights.
We address the latter contention first. The first step in assessing Ely’s claim is to determine whether the requested evidence is, in fact, exculpatory. Commonwealth v. Healy, 438 Mass. 672, 679 (2003). In this context, we have interpreted the meaning of exculpatory broadly, extending beyond alibi and proof of innocence to “all evidence which tends to ‘negate the guilt of the accused’ ... or stated affirmatively, ‘support[] the innocence of the defendant.’ ” Id. at 679, quoting Commonwealth v. St. Germain, 381 Mass. 256, 261 n.6 (1980). Moreover, Bly’s due process right encompasses all evidence that is of significant aid to his case, “whether it furnishes corroboration of the defendant’s story, calls into question a material, although not indispensable, element of the prosecution’s version of the events, or challenges the credibility of a key prosecution witness.” Commonwealth v. Ellison, 376 Mass. 1, 22 (1978).
In this case, the underlying data sought by Bly consisted of testing documents for each of the individuals that made up the Bode database, a total of over 1,600 people. Bode provided Bly with a representative sample of less than one per cent of this documentation. In order to provide Bly with the complete set of documents, Dr. McElffesh estimated that production would cost $250,000, and the Commonwealth argued that such an undertaking would take at least one to two months, followed by a similar time period for the defense analysis. The Commonwealth did not dispute the relevance of the requested documents, only the practicality of producing them. Had Bly been permitted to review the underlying data, they could have formed a valid means of impeachment of Dr. McElfresh and the Bode testing results.
Having determined that the evidence sought was potentially exculpatory, our next step is to determine whether the nondisclo[486]*486sure of such evidence constitutes a due process violation. Where a defendant’s specific discovery request goes unmet, we have outlined a standard of prejudice favorable to the defendant. Commonwealth v. Healy, supra at 679-680. Bly need only show that “a substantial basis exists for claiming prejudice from the nondisclosure.” Id. at 680, quoting Commonwealth v. Tucceri, 412 Mass. 401, 412 (1992). This threshold can be met with record support for the conclusion that the jury would have been influenced by timely disclosure of the evidence in question. Commonwealth v. Healy, supra, quoting Commonwealth v. Daye, 411 Mass. 719, 729 (1992).
The judge dismissed Ely’s discovery request by suggesting that there were alternative methods of attacking the reliability of the database, pointing particularly to the Commonwealth’s suggestion that Bly compare the over-all results of the Bode database to other, peer-reviewed databases. Bly has not at any time explained why such a solution was inadequate or inflicted prejudice, relative to his preferred manner of attacking the database. Indeed, Ely’s database argument, both through his own expert at trial and on appeal, illustrates that such an approach can be reasonably undertaken. See infra at 487-489. In light of Ely’s current attack on the Bode database and his silence as to why such method was not an acceptable substitute for the evidence sought, we conclude that he has failed to meet his burden of showing the existence of a substantial basis for claiming prejudice.
We turn now to Ely’s primary contention that use of the Bode database produced scientifically unreliable match frequencies because it was not subjected to peer review and had an unacceptable differential in comparison to the Promega database. Generally, peer review and publication are important factors for the validation of a database. Commonwealth v. Curnin, 409 Mass. 218, 223-224 n.9 (1991). We have held, however, that they are not absolute prerequisites to admissibility. Commonwealth v. Lanigan, 419 Mass. 15, 25 (1994). The Commonwealth conceded that Bode’s database has not been published in a peer review journal. Testimony at the pretrial hearing, however, indicated that the database was not published because its content did not represent any new development [487]*487worthy of publication. The very reason that the database showed signs of reliability, i.e., that its results conformed with other, previously published databases, is thus the same factor that explained its lack of publication. Moreover, Dr. McElfresh testified that both Dr. Martin Tracey of Florida International University, and Jeff Ban of the Department of Forensic Science DNA Laboratory in Richmond, Virginia, had reviewed the Bode database and found it to be reliable. Although such scrutiny does not constitute formal peer review, see Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 593-594 (1993) (Daubert), the judge appropriately considered these facts as corroborative of the database’s reliability.
Bly also contends that the Bode database was unreliable because it showed an unacceptable differential in the level of probability distribution when compared to a published database.* 7 Although not offered at the motion in limine, Dr. William Shields, Ely’s expert, produced an exhibit at trial that showed the statistical differences between the Bode and Promega databases at a single genetic location, THOl.8 Using statistical analysis of the allele distributions in the two databases, he testified that a “chi square test pattern” determined the probability of the two databases being truly different at this location to be greater than ninety-nine per cent.
In stark contrast, Dr. Tracey testified during the pretrial proceedings that the allele and genotype frequencies in the Bode database were not statistically significantly different from other databases, including Promega. Dr. Tracey testified that he compared the frequencies in at least nine different databases, including Bode, Promega, and four specialized databases compiled by the sheriff’s office in Broward County, Florida. Dr. Tracey also testified that he examined the allele frequencies at [488]*488all five of the genetic locations that returned results in the December testing. The extensive analysis performed by Dr. Tracey indicated that the frequency of particular alleles in the Bode database was consistent within expected statistical deviations, based on comparison with these other databases, across all genetic locations that he studied.9 Based on these results, Dr. Tracey opined that the Bode database would be accepted generally and its results reliable.
In denying the defendant’s motion in limine, the judge credited Dr. Tracey’s opinion that the Bode database was reliable. He also relied on testimony that the database was built in accordance with Bode laboratory protocols, and found that the laboratory generally adheres to the principles set forth in the 1996 National Research Council report that we previously have found to be authoritative. See Commonwealth v. Rosier, 425 Mass. 807, 812 (1997). The judge also cited pretrial testimony that the Bode database was both in linkage equihbrium and in Hardy-Weinberg equihbrium.10 Last, given the avenue of attack sought by Ely’s discovery request, it is relevant that each of the database samples had been typed more than once, and most often more than four times.11 In concert with the lack of mistaken genotyping found in the small representative sample, [489]*489this testimony supports the conclusion that the underlying work of the Bode database was sound.
In these circumstances, the judge did not abuse his discretion in finding the Bode database to be scientifically reliable. Because Ely’s proffered evidence at trial concerning the disparity in match frequencies between the Bode database and others was not presented in conjunction with his pretrial motion in limine, see Commonwealth v. Sparks, 433 Mass. 654, 659 (2001), he waived its consideration on the question of admissibility.12 Its omission from the pretrial hearing did not create a substantial likelihood of a miscarriage of justice, see Commonwealth v. Wright, 411 Mass. 678, 682 (1992), because the judge had been made aware of Dr. Tracey’s opinion that the differences between the Bode database and more established DNA compilations were within acceptable statistical variances. Any difference brought out by Dr. Shields during trial was properly a matter of weight to be decided by the jury. Commonwealth v. Vao Sok, 425 Mass. 787, 806 n.27 (1997). See Daubert, supra at 596 (“Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence”). In accordance with the judge’s instruction, the jury were free to reject or accept Dr. Shields’s opinion of Bode’s database comparison just as they could evaluate the weight of all testimony.
4. Suppression issues, (a) Physical evidence. Bly argues that the method used by the Commonwealth in obtaining his known DNA sample constituted a nonconsensual seizure and thus violated his rights under the Fourth and Fourteenth Amendments to the United States Constitution, under art. 14 of the Massachusetts Declaration of Rights and under the search warrant requirements of G. L. c. 276, § 1. The judge denied Bly’s [490]*490motion to suppress on this issue, finding that the cigarette butts and water bottle seized by the police constituted trash that was abandoned by Bly.
The thrust of Ely’s argument is that the police, suspecting they could not gain consent, deceived Bly into providing the same material that would have resulted from a consensual search and seizure.13 Additionally, Bly argues that he did not abandon voluntarily the items collected by the police, but rather was required by institutional rules to leave those items behind when he left.
To succeed on appeal, Bly must bear the threshold burden of showing that a warrantless search or seizure occurred. Commonwealth v. D’Onofrio, 396 Mass. 711, 714-715 (1986). This question is analyzed under the familiar two-part query whether Bly had a subjective expectation of privacy in the items seized, and if so, whether that expectation was reasonable objectively. See Commonwealth v. Montanez, 410 Mass. 290, 301 (1991), citing California v. Ciraolo, 476 U.S. 207, 211 (1986).
Bly makes much of the judge’s finding that the institutional rules prohibited him from leaving the administration building where he was interviewed with property that he did not possess when he entered. He argues that no legal abandonment can occur in circumstances where he was required by the Commonwealth to leave the items behind. Based on the factual findings of the judge, however, there is no evidence that Bly showed any expectation of privacy in the items in the first place. After being interviewed for two hours, smoking the cigarettes and drinking from the water bottle, Bly requested use of a telephone to consult with his mother. The detectives took him to another room in the same building in order to make use of a secure telephone line. Bly did not attempt to take the cigarette butts and water bottle with him when he left the interview room to travel within the building, and he did not request to go back, and collect them after using the telephone, even when prompted. The detectives then waited one-half hour before collecting the items in order to allow an opportunity for Bly to return or [491]*491protest. Our conclusion that Bly had no subjective expectation of privacy is compelled not by a finding that he legally abandoned them as much as it is by his wholesale failure to manifest any expectation of privacy in the items whatsoever.
There was no error in the denial of Ely’s motion to suppress physical evidence as no search or seizure occurred.
(b) Statement to police. Bly also challenges the introduction of a statement he made to the police on the day after McLaughlin’s murder without the benefit of Miranda warnings. After holding an evidentiary hearing, the judge denied Ely’s motion to suppress this statement, ruling that it was not the result of a custodial interrogation. We review the judge’s findings of fact and conclusions of law with substantial deference, but apply constitutional principles to those findings independently. Commonwealth v. Conkey, 430 Mass. 139, 144 (1999), S.C., 443 Mass. 60 (2004).
Plainclothes officers on patrol in the vicinity of Theodore and Morton Streets in the Mattapan section of Boston saw Bly, who was known to them, on the afternoon of September 26, 1995, in a group of young men. Although Bly was not yet the focus of the investigation, the officers had received information that homicide detectives were interested in speaking with him, because he was scheduled to go to trial against McLaughlin. The patrolling officers broadcast on their police radios that they had located Bly, then engaged the group in conversation. A short time later, two detectives arrived and asked Bly if he would accompany them to the police station. Bly agreed and rode with them to the station, where he was interviewed by two homicide detectives for approximately forty-five minutes in a detective’s office with the door open. The judge found that the interview was at all times cordial and conducted in a conversational manner. When asked if the interview could be recorded by audiotape, Bly responded by saying, “Do it. I got nothing to be afraid of, nothing to worry about.”14 Bly also acknowledged on tape that he had come to the interview willingly and voluntarily. The judge found that Bly did not request an at[492]*492tomey or ask to leave at any point during the interview. When the interview was concluded, Bly was permitted to listen to the recording and given an opportunity to make corrections before the two homicide detectives gave him a ride back to Mattapan.
It is fundamental that a statement made by a criminal defendant in response to questions while in police custody must be preceded by Miranda warnings in order to be admissible against him. Commonwealth v. Haas, 373 Mass. 545, 552 (1977), S.C., 398 Mass. 806 (1986). The dispute here focuses on whether Bly was in custody at the time the statement was made, a question on which Bly bears the burden. Commonwealth v. Larkin, 429 Mass. 426, 432 (1999). Our analysis focuses on whether a reasonable person in Ely’s position would believe the circumstances to be coercive. Id. In considering whether a defendant is in police custody, we take account of the following factors: “(1) the place of the interrogation; (2) whether the officers have conveyed to the person being questioned any belief or opinion that that person is a suspect; (3) the nature of the interrogation, including whether the interview was aggressive or, instead, informal and influenced in its contours by the person being interviewed; and (4) whether, at the time the incriminating statement was made, the person was free to end the interview . . . .” Commonwealth v. Groome, 435 Mass. 201, 211-212 (2001).
Based on the judge’s well-supported findings, it is clear that Bly was not in custody at the time of the interview. Although he was in a police station, he had gone there voluntarily. Commonwealth v. Corriveau, 396 Mass. 319, 327 (1985). The interview was not conducted in an interrogation room, but rather in an opened door office. The investigation into McLaughlin’s murder was in its nascent stages, and Bly was interviewed only because he was on the victim’s trial schedule, and not because the police had any evidence specifically linking him to the crime. At no point during the interview was Bly told that he was a suspect. Bly stated that his attorney informed him that the police would be interested in speaking to all persons McLaughlin was prosecuting and the officers acquiesced in that belief. The objective circumstances could not lead Bly to believe that this was anything other than an interview for the purposes of information gathering.
[493]*493With respect to the final two factors, the circumstances of Ely’s interview bear a strong resemblance to the pre-Miranda questioning in Commonwealth v. Sparks, 433 Mass. 654, 656 (2001), which was noncustodial. The judge found, and a review of the audiotape confirms, that the tenor of Ely’s interview was cordial and conversational, not aggressive or coercive. Although the police did not affirmatively inform Bly of his right to leave, neither did they act in such a manner that would lead to any other conclusion, as evidenced by their inquiry of his voluntary presence there. At no point did Bly express a desire to leave, and at the end of the interview, the police transported him back to where he was first encountered.
The judge’s finding that Ely’s statement was made in a noncustodial setting is supported by the evidence and correctly interprets constitutional principles. There was no error.
5. Identification issues. The judge denied without a hearing a pretrial motion by Bly to suppress the eyewitness testimony of Maureen Woodsum.15 Bly claims Woodsum’s testimony played an important role in the Commonwealth’s case, because it was the only evidence heard by the jury from a witness unknown to Bly who identified him as being in West Roxbury near the time of the crime. Bly complains that the judge abused his discretion in allowing Woodsum to testify because her testimony was inherently unreliable.16 He also argues that the judge committed error in prohibiting his expert on cross-racial identification from testifying before the jury after being heard on voir dire and in [494]*494rejecting his request to provide jury instructions on the issue of cross-racial identification.
(a) Woodsum identification. Ely’s contention that Woodsum’s testimony was unreliable is based on the fact that (1) her identification of Bly occurred through a television broadcast of his indictment in this case; (2) Woodsum’s identification did not occur until two and one-half years after the murder; (3) Wood-sum’s original sighting occurred while she was driving, and while she was dealing with the recent loss of her father, problems at work, and a medical condition; and (4) the cross-racial element of the identification. Bly relies primarily on Commonwealth v. Jones, 423 Mass. 99, 108 (1996), for the proposition that some eyewitness identifications occur under such highly suggestive circumstances that exclusion is warranted on grounds of fairness.17
The threshold of unreliability described in Commonwealth v. Jones, supra at 109, is when “a witness is involved in a highly suggestive confrontation with a defendant and that witness’s in-court identification of the defendant is not shown to have a basis independent of that confrontation.” In the present case, Woodsum testified under vigorous cross-examination that her identification of Bly in court stemmed from her seeing him on the bridge, rather than from her observing the television broadcast. Moreover, the suggestiveness of the subsequent [495]*495confrontation between Bly and Woodsum was far less than in the Jones case. Although no reproduction of the broadcast is contained in the record, Woodsum testified that she remembered the man she saw on television as the man from the bridge over the Highland station prior to learning that he was under indictment for the murder of Paul McLaughlin. In the Jones case itself, we distinguished the circumstances there from “a casual confrontation in neutral surroundings, such as those that occur through the media.” Commonwealth v. Jones, supra at 109-110, citing, e.g., Commonwealth v. Colon-Cruz, 408 Mass. 533, 542 (1990). See Commonwealth v. Dyer, 389 Mass. 677, 684-685 (1983). Although Bly’s appeal raises several valid factors for the jury’s consideration of the reliability of Woodsum’s identification, as did his cross-examination and closing argument, we cannot conclude that the judge abused his discretion in allowing Woodsum’s testimony. The circumstances before us do not equate to the degree of unreliability described in the Jones case.
(b) Exclusion of expert testimony. Bly argues that the judge abused his discretion in disallowing the testimony of Ely’s expert witness on eyewitness identification. After allowing a voir dire examination, the judge ruled that the jury were capable of evaluating the reliability of Woodsum’s opinion without the aid of expert testimony.
Expert testimony is admissible on the issue of eyewitness identification, within the discretion of the trial judge. Commonwealth v. Hyatt, 419 Mass. 815, 818 (1995). We took the opportunity presented in Commonwealth v. Santoli, 424 Mass. 837 (1997), to offer guidance to judges in exercising that discretion. In discerning the commonality of cases where judges were found to overstep their bounds in excluding such expert testimony, we noted that “there was little or no evidence to corroborate the eyewitness identification.” Id. at 842. As in the Santoli case, we defer here to the judge’s discretion because there was substantial corroboration of Woodsum’s eyewitness identification by witnesses who assisted Bly in preparing for the killing, and who heard his admissions thereafter. Additionally, the physical evidence collected at the scene connected Bly to the murder, as confirmed by three separate sets of laboratory [496]*496testing, albeit at varying degrees of match frequency. Bly availed himself of his opportunity to attack all of this evidence. Yet the conclusion that Woodsum’s testimony was substantially corroborated survives these attacks. The judge acted within his discretion in deciding that the nature of Woodsum’s identification, including the circumstances of her original sighting, the time delay, the television broadcast, and the cross-racial aspect were such that the jury’s evaluation could be accomplished through its common understanding without need of expert testimony.
(c) Jury instructions. Bly also argues that the judge abused his discretion in failing to charge the jury on the issue of cross-racial identification. We have said the decision to give such an instruction is within a judge’s discretion when warranted by the evidence. Commonwealth v. Hyatt, supra at 819. However, Bly was not entitled to such an instruction. At the close of the charge, Bly made a request that the jury receive an instruction that stated, in part, “You may consider in the context of my instructions on identification that persons of one race have difficulty in identifying persons of another race.” The judge denied this request, relying on the identification instruction as given, which raised, inter alla, the issues of the reliability of the initial observation, good faith mistake, the circumstances of the identification, the time between the original observation and the identification, and the lack of a multiple-person lineup.
While we acknowledge the significant body of scientific literature on the problems inherent in cross-racial identification, see Commonwealth v. Zimmerman, 441 Mass. 146, 153-156 (2004) (Cordy, J., concurring), we have never held that those problems require a jury instruction when cross-racial identification testimony is offered, and we decline to do so here. The instruction requested by Bly is expressed in broad language that ostensibly replaces the common understanding of the juror. The judge did not abuse his discretion in declining to include it in his instructions to the jury.
6. Refusal evidence. Bly contends that his constitutional rights were violated when the judge allowed the jury to hear testimony concerning his refusal to provide a court ordered hair sample. Following a hearing, the judge allowed the Commonwealth’s [497]*497pretrial motion that Bly be ordered to provide blood and hair samples for the purpose of comparison to items retrieved from the crime scene. The Commonwealth elicited testimony at trial showing that, although Bly had complied with the court order in connection with the blood sample, he had refused to refrain from cutting his hair in order for an adequate hair sample to be taken.
The refusal evidence did not violate Ely’s privilege against self-incrimination, see South Dakota v. Neville, 459 U.S. 553, 564 (1983) (evidence of refusal to submit to blood-alcohol test does not run afoul of Fifth Amendment to United States Constitution), or the more protective art. 12 of the Massachusetts Declaration of Rights.18 Commonwealth v. Delaney, 442 Mass. 604, 608-610 (2004). In that case, decided after Ely’s trial concluded, we held that when a criminal defendant refuses to produce evidence that is the subject of a warrant or court order, admission of evidence concerning that refusal does not violate his rights under art. 12 because he does not face the choice that was integral to our decision in Opinion of the Justices, 412 Mass. 1201, 1211 (1992). Commonwealth v. Delaney, supra. The choice either to produce incriminating evidence or be punished with an inference of guilt in refusal is absent when a defendant’s decision to cooperate is foreclosed by order of a judge. Id.
7. Prior bad acts. Bly also argues that the judge erred in allowing the Commonwealth to elicit testimony concerning the details of the Dana Alston carjacking, during which Bly brandished a gun, and Ely’s subsequent effort to prevent Alston from testifying, which included a botched assassination attempt by Ely’s associate, Errol Moses. According to Bly, even if otherwise admissible, the testimony should have been excluded because of its overwhelmingly prejudicial nature. We disagree.
Evidence of prior bad acts is inadmissible for the purpose of showing bad character or a propensity to act in a manner similar to the crime at issue. Commonwealth v. Butler, 445 Mass. 568, [498]*498574 (2005), quoting Commonwealth v. Barrett, 418 Mass. 788, 793 (1994). Bly does not rely on this fundamental mle because the Alston testimony was used by the Commonwealth for the permissible purpose of showing motive. Commonwealth v. Crouse, 447 Mass. 558, 567 (2006). Instead, Bly argues that the evidence at issue is so exceedingly prejudicial that it should not have been admitted for any purpose. It is left to the judge’s sound discretion to determine whether the probative value of such evidence outweighs the risk of an unduly prejudicial effect on a defendant in the eyes of the jury. Commonwealth v. Brousseau, 421 Mass. 647, 650 (1996), quoting Commonwealth v. Robertson, 408 Mass. 747, 750 (1990).
The evidence concerning the carjacking incident was highly probative of Bly’s state of mind and motive to kill Paul McLaughlin. It gave the jury a sense of the strength of the case against Bly, and thus, how strongly he felt about his need to avoid prosecution.19 Likewise, evidence concerning Bly’s involvement in the intimidation of Alston demonstrated the degree to which Bly wished to avert trial. The Alston testimony, in concert with evidence that McLaughlin had prosecuted Bly on two prior occasions in the previous eight months, gave credence to the notion that Bly viewed him as a nemesis. Last, omission of certain details of the Alston case would have rendered the Commonwealth’s case unintelligible. Cf. Commonwealth v. Brown, 389 Mass. 382, 385 (1983). Without Alston’s testimony concerning the carjacking and subsequent attempted murder by Moses, the jury’s understanding of why Bly would have resorted to such drastic lengths to avoid trial would have been significantly impaired. There was no error.
8. Pretrial discovery. Bly claims that he was prejudiced by the failure of the judge to order the pretrial production of Janet Brown’s medical records. Bly moved before trial for access to these records, on the basis of Brown’s statements to the police and the grand jury concerning treatment at several hospitals for a nervous condition and related medications. The judge denied [499]*499the motion, after hearing argument from the Commonwealth that Brown had functioned as a dean’s list student at a community college, an Internal Revenue Service employee, and a doctor’s assistant while on this medication.
When Janet Brown’s trial testimony showed considerable confusion as to the chronology and existence of certain events, the judge suspended her testimony and ordered in camera production of her psychiatric treatment records. On May 19, 1999, during the course of trial, the judge reviewed the medical records received to that point and made them available to counsel for inspection. He conducted a similar review at the opening of court on Friday, May 21, 1999. After dismissing the jury early that same day, the judge provided counsel with an opportunity to spend the rest of the afternoon reviewing Brown’s records, in advance of her resumed cross-examination the following Monday. Defense counsel’s statement to the judge after reviewing the records indicated that Brown only took three medications relevant to her mental health. Counsel then resumed cross-examination without objection or other complaint. During cross-examination, Brown’s medications were brought to light, and the jury heard that she experienced hallucinations in the past for which she had been hospitalized, as well as testimony that she had been previously diagnosed with paranoid schizophrenia.
Bly complains that the judge failed to make a written ruling as to whether the records at issue were privileged during the pretrial proceedings, thus running afoul of the process laid out in Commonwealth v. Bishop, 416 Mass. 169, 181-183 (1993), and its progeny.20 Even assuming that Bly’s claim is correct, the judge’s failure in this regard was harmless, in light of the subsequent production and opportunity to review all of the records at issue, and the ensuing cross-examination of the witness. See Commonwealth v. Dexter, 50 Mass. App. Ct. 30, 33-34 (2000) (no prejudice resulted from denial of Bishop motion where, inter alla, defendant had opportunity to cross-examine victim on her use of medication). Bly’s argument that prejudice [500]*500resulted from the time frame he was given to review the records is unavailing in light of this cross-examination, especially when the judge heard no objection to proceeding at that time or request from Bly to make use of an expert on these issues.
9. G. L. c. 278, § 33E. We have reviewed the briefs on appeal, the entire record below, including the transcripts of both the pretrial proceedings and the postconviction hearings, and decline to reduce the conviction or to order a new trial.
Judgment affirmed.