*157¶1
Kennedy, J.— Dennis Delmarter, Ardell Shaw, and Harold Bain petition for release from personal restraint based on the on-the-job malfeasance of former Washington State Patrol Crime Laboratory forensic chemist Michael Hoover. They argue that their petitions must be granted under State v. Roche, 114 Wn. App. 424, 59 P.3d 682 (2002) and Brady v. Maryland, 373 U.S. 83, 83 S. Ct. 1194, 10 L. Ed. 2d 215 (1963). Because there is a lack of substantial evidence that Hoover was stealing heroin from the state crime lab at the time he completed the tests involved in these petitioners’ cases, and because these petitioners fail to establish any violation of Brady, we deny their personal restraint petitions.
FACTS
¶2 Michael Hoover worked as a forensic chemist at the Washington State Patrol Crime Laboratory from sometime in 1989 until December 2000. See Roche, 114 Wn. App. at 428-32. Following a police investigation prompted by coworkers’ reports of suspicious activity, Hoover admitted, on December 22, 2000, to having stolen heroin from the lab and using it to self-medicate his chronic back pain, commencing in the fall of 1998. Hoover eventually pleaded guilty to charges of evidence tampering and official misconduct. Id. at 431.
f 3 Based on incidents occurring on September 20, 1996, the State charged Dennis Delmarter with one count of possession of cocaine with intent to deliver and one count of possession of heroin with intent to deliver.1 Based on a February 7, 1997, incident, the State charged him with one count of possession of heroin. Hoover tested the substances at issue in these charges on December 6, 1996, and February 19, 1997, respectively. Following a trial on September 23-24, 1997, at which Delmarter admitted to possessing *158heroin and cocaine on September 20 and heroin on February 7, and the State presented evidence that the substances field-tested positive for cocaine and heroin, a jury found Delmarter guilty of all three charges.
¶4 The State charged Ardell Shaw with one count of possession of cocaine with intent to deliver, based on events occurring on July 24, 1997.2 Another charge was based on incidents that occurred on September 2, 1997.3 Hoover tested the evidence for these cases on November 18, 1997, and September 19, 1997, respectively, and then testified at Shaw’s trials in October and December 1999. The State presented circumstantial evidence, as well as field tests, that also indicated that the substances at issue were cocaine.
¶5 On June 9, 1998, Harold Bain pleaded guilty to one count of delivery of cocaine. His statement on plea of guilty asserts: “On March 25, 1998, in Snohomish Co. WA I delivered a small amount of cocaine to an undercover officer, knowing it was a controlled substance.” The affidavit of probable cause states that the substance Bain sold “field tested positive for cocaine” and that after his arrest, Bain “admitted to selling cocaine and methamphetamine” to the officer. Hoover tested the substance at issue in the case on May 11, 1998.
¶6 To address the questions presented in the petitions filed by Shaw and Delmarter, this court ordered a reference hearing to determine the time frame of Hoover’s misconduct and the manner in which it progressed. The trial court found “by a preponderance of the evidence that Mr. Hoover’s heroin use began at least in 1998. There is an inference that the use of heroin may have begun as early as 1995, but the exact timeframe cannot be determined be*159cause of the situation at the lab.” Court’s Findings and Conclusions after Reference Hearing, at 3. The court also found that, “Mr. Hoover began stealing heroin at the same time he began abusing heroin, which occurred at least by the fall of 1998, although there is an inference that the use may have started earlier.” Id. at 4. The court found that Hoover’s coworkers did not suspect that he was using drugs; instead their suspicions “related to time management issues.” Id. at 3-4. Finally, with regard to suspicions of “dry labbing” in September 2000, the court found, “The evidence showed the results produced by Mr. Hoover in suspected cases of dry labbing were accurate. There is an insufficient showing that Mr. Hoover in fact ever dry labbed any case.” Id. at 4.
¶7 In its oral ruling on October 17, 2003, the court stated:
The Court held a hearing in this particular matter and heard from the following witnesses; Eric Nielson from the Washington State Crime Lab, Dr. David Northrop from the Washington State Crime Lab; James Boaz from the Washington State Crime Lab, and also Mr. Michael Hoover concerning what transpired in the Crime Lab involving Mr. Hoover during this time frame. These chemists who were employed by the Washington State Crime Lab, including Mr. Hoover, were all interested in performing their services in the crime lab to the best of their ability. They had integrity and trust amongst themselves. They were trying to do the right thing on these tests, so they would stand up in the Court of law in the event that one of their experiments were required to be tested.
When one examines the floor plan of the Washington State Patrol Crime Lab, these were somewhat self-sufficient chemists. They didn’t really look over each others’ shoulders. They had a place to perform their experiments, and another place, an office, to write their notes. From just the physical layout, they didn’t really check on each other. There was no mechanism for doing that, because they didn’t expect to do that because they trusted each other. They were there to do the State’s work. So it would be difficult to find someone who would actually do the sorts of things that Mr. Hoover did, and basically, the only way *160they found out was when they videotaped him in 2000 concerning his taking of these drugs. So when I look at these particular questions, the first question being, when did Mr. Hoover begin to abuse controlled substances? Well, we know from the record that there was some medical information submitted to the Court that he ceased going to the doctor in 1995. We also know that in April there was a scratching and snorting sound coming from his cubicle area. The explanation for this was he was using an antihistamine. Nothing more was said about that. It was a one-time incident. There was also the incident involving Mr. Boaz in August of ‘98 wherein he began to lock his drawer concerning the heroin cases that Mr. Hoover was taking. And if one examines Exhibit No. 13, you can see that during certain months he did have a large differential in heroin cases that he would examine.
MS. WEBBER: Judge, “he” being?
THE COURT: Mr. Hoover. There were some months when he would do less, but there were also a significant number of months he would do more heroin cases than the others.
And also, in the fall of 1998, Mr. Hoover decided to perform the bulk heroin project, wherein he explained that there was a dog handler that came to the lab, that he’d never seen him again. He was not authorized to perform that experiment on behalf of the lab. He was told eventually not to do it. It was never approved again. That happened in 1998. So when one examines this record, I’m satisfied that by a preponderance of the evidence that when he started to do the bulk heroin project, from this Court’s perspective, that was a coverup, and that when Boaz started to lock his drawer, that was a pretty good indication that he was in fact abusing drugs and taking drugs.
Also, I would agree with Ms. Tarantino, the real time frame when he started to abuse drugs is really unknown. It could go back as far as ‘95, and it could go up through ‘97 at that one incident, and it could also be ‘98. We all know that after that he did start to abuse. I’m not helping the Court of Appeals at all, but the record is unclear because of the way the operation was set up, and the lack of supervision at the Crime Lab as to when he actually started to abuse drugs.
MS. WEBBER: Judge, can I ask you to clarify that point? You are saying at least the evidence that we have at this point is in August of ‘98 is when he began to use drugs?
*161THE COURT: I’m satisfied that that’s the best evidence that we have, that based on the circumstances of the heroin project, the bulk heroin project, the locking of the drawers, he was making a specific effort to take more cases. The only reason I can think of him doing that was he was attempting to take drugs and abuse them. However, when one examines Exhibit 13, you will see that over the previous several months back into ‘97 that he in fact had significant months where he did most of the heroin cases, more than the others. I would also say there’s an inference that can be drawn that it went back to prior to 1998, but I can’t say for certain.
And the second question is when did coworkers begin to suspect drug abuse and on what basis? Well, I don’t believe they really ever suspected that he was using drugs. But it happened, as Ms. Tarantino would say, but I do believe again when Boaz locked his drawer in August of‘98, that he had some suspicion that something was going on. I don’t believe he fully appreciated what it was all about. He thought he was maybe trying to take easy cases and so forth. I do believe that was a red flag.
Number 3 is when Hoover began to steal controlled substances from the State Crime Lab. That to me would go to the same as question number one. When he began to abuse, he began to steal the drugs. So my reasoning for one and three would be the same.
Number 4, when and how suspicions that Hoover was stealing these samples emerged among the coworkers. I believe they relate to his time management problems. The earliest probably was September of 2000, and when they observed the videotapes is when they really became suspicious that he was in fact taking and using these drugs.
Finally, number 5 is whether Hoover was repeatedly testing the same purified sample across a number of cases, the process of dry labbing, when and how suspicions arose that he was dry labbing. I suspect in the fall of 2000 is when they suspected he was dry labbing. I’m not satisfied, however, that there was any specific evidence establishing dry labbing. Nielson said his tests were satisfactory and so forth. So I’m not convinced that there has been a sufficient showing that there has been any dry labbing for purposes of this case.
*162Report of Proceedings (Oct. 17, 2003) at 4-9.4
ANALYSIS
f 8 Where a personal restraint petition presents a challenge to a decision of which the inmate “generally has had no previous or alternative avenue for obtaining state judicial review,” this court reviews the petition according to RAP 16.4. Roche, 114 Wn. App. at 440-41 (citing In re Pers. Restraint of Cashaw, 123 Wn.2d 138, 148-49, 866 P.2d 8 (1994)). Under RAP 16.4(c)(3), restraint is unlawful where material facts exist that have not been previously presented and heard, which in the interest of justice require vacation of the conviction. Roche at 441. To require vacation of a conviction under RAP 16.4(c)(3), a petitioner must establish that: (1) the results will probably change if a new trial is granted, (2) the evidence was discovered since trial, (3) the evidence could not have been discovered before trial by exercising due diligence, (4) the evidence is material, and (5) the evidence is not merely cumulative or impeaching. Roche, 114 Wn. App. at 444 (citations omitted).
Roche
¶9 In Roche, this court reversed two convictions because Hoover’s misconduct broke the chain of custody and tainted the integrity of two trials. Id. at 437, 440, 446. The State charged James Roche with felony drug possession based on incidents occurring in December 1999, after which Hoover tested the substances involved and testified at the trial. Id. at 431-32. Similarly, Roy Sweeny’s felony drug possession charge was based on events taking place in October 1999, and he was tried and convicted in part based on his stipulation to Hoover’s lab test of the substance involved. Id. at 434.
*163¶10 But Roche does not require a new trial in every case in which Hoover ever had a part. In re Personal Restraint of Brennan, 117 Wn. App. 797, 800, 72 P.3d 182 (2003) involved an arrest and Hoover lab test performed in 1996, followed by Brennan’s guilty plea in May 1998. This court found three significant differences that distinguished Roche: (1) Brennan pleaded guilty whereas Roche and Sweeney went to trial; (2) independent evidence, including a field test and Brennan’s confession, established that the substance taken from Brennan’s person was methamphetamine, whereas Roche did not admit that the substances found in his home were methamphetamines, despite a positive field test; and (3) there was no temporal connection between Hoover’s misconduct in 1998 and his test of the substance found on Brennan in 1996, whereas there was compelling evidence that Hoover was abusing heroin during the time he conducted the tests at issue in Roche. 117 Wn. App. at 803-04. Because Brennan failed to present evidence suggesting that the test result in his case was inaccurate, and his confession and the positive field test supported his guilty plea, this court concluded that the interests of justice did not require that Brennan be allowed to withdraw his guilty plea. Id. at 804.
¶11 Unlike the circumstances in Roche, Bain and Delmarter do not have material evidence casting doubt on the accuracy of the tests in their cases. Thus, Hoover’s misconduct regarding other tests in other cases at later dates could constitute only impeachment evidence in their cases. Like Brennan, Bain pleaded guilty, independent evidence of a field test and a confession supported his plea, and there was no temporal connection between Hoover’s misconduct beginning in the fall of 1998, as found by the trial court at the reference hearing, and Hoover’s test of the evidence in Bain’s case in May 1998, when there was insufficient evidence to indicate that Hoover was using drugs.5 Similarly, although Delmarter went to trial, inde*164pendent evidence — field tests and his admissions — supported the jury’s verdict, and Hoover’s participation in the case took place in 1996 and 1997, significantly before the fall of 1998. Based on Brennan, the interests of justice do not require that Bain be allowed to withdraw his guilty plea or that Delmarter be granted a new trial.
¶12 In Shaw’s case, however, although Hoover tested the substances at issue in 1997, he actually testified at Shaw’s trials in late 1999, well within the time frame of Hoover’s misconduct as found by the reference court. Additionally, Shaw challenges the reference court’s findings of fact. The reference court’s findings are sufficient if supported by substantial evidence, in that “ ‘the record contains evidence of sufficient quantity to persuade a fair-minded, rational person that the declared premise is true.’ ” In re Pers. Restraint of Gentry, 137 Wn.2d 378, 410, 972 P.2d 1250 (1999) (citations omitted).
¶13 Shaw contends that findings 5, 7, 8, and 9 are not supported by substantial evidence. Finding 5 lists the factors demonstrating that Hoover’s heroin use began in 1998. In particular, Shaw challenges the reference court’s finding in 5.b. that the scratching and sniffing sound noticed by James Boaz in April or November 1997 “was a one time event.” Court’s Findings at 3. Shaw contends that the reference court was mistaken about Boaz’s testimony because Boaz described the sounds as “sometimes” or “occasionally” coming from the workbench or desk, and made other statements that could indicate it was an ongoing behavior. However, the following exchange also occurred on cross-examination of Boaz:
Q You don’t have any specific recollection except for any scratching or snorting happening except for the one time that you asked Mr. Strongman about that; is that correct?
A The one clear recollection I have is having asked Mark about it, and that’s why that date stands out, because it was very early when I transferred over.
Q The sniffing actually happened at his desk where he did paperwork; correct?
*165A Correct.
Q The explanation you were given, he was using an antihistamine; is that right?
A To my recollection, correct.
Q There was something about what you observed that would suggest that that planning was false or incorrect; is that right?
A There was nothing that made me question it.
Q Again, they happened at two different locations?
A One at his work area. Later on at his lab area. The other at his office area.
Report of Proceedings (Sept. 23, 2003) at 268.
¶14 Based on this testimony, that Boaz had a clear recollection of only one specific incident, it was not a mistake, nor was it unreasonable for the reference court to find that only one incident was supported by a preponderance of the evidence. Even if the reference court had found that the scratching and sniffing happened more than once, it would not thereby have been bound to find that Hoover’s drug use began in 1997, as Shaw argues, especially in light of the testimony that Hoover’s coworkers knew he had allergies and was using an antihistamine.
¶15 Shaw also challenges finding 7, contending that 1997 is a more likely date for the beginning of Hoover’s drug use. Shaw relies solely on his disagreement with the reference court’s finding in 5.b. regarding the scratching and sniffing sounds. Because that finding is supported by substantial evidence, and because finding 7 is supported by testimony that Hoover began his bulk heroin project without authorization in the fall of 1998, Shaw fails to establish error.
¶16 Shaw next contends that finding 8 should not be considered because it does not answer the Court of Appeals’ question, “When and how did suspicions that Hoover was stealing the samples emerge [ ] among coworkers?” Court’s Findings at 2. But substantial evidence supports the reference court’s findings that his coworkers did not suspect Hoover was stealing samples. Neilson, Northup, and Boaz *166described their observations of strange behavior and testified that they suspected that he was doing sloppy work and mismanaging his time.
¶17 Shaw also challenges finding 9 regarding dry-labbing. Although Boaz suspected dry-labbing because certain results of Hoover’s tests appeared too similar to come from specific samples, Boaz testified that he had no evidence to confirm his suspicions. Moreover, Neilson and Northup testified that there was no actual evidence of dry-labbing and that the similar results could be explained by the similarity of the samples. All three stated at the hearing that they had no evidence to indicate that Hoover’s test results were inaccurate. Based on this testimony, as well as evidence that retesting of some of the samples confirmed Hoover’s previous results, the challenged finding is supported by substantial evidence.
¶18 The question remains as to whether Shaw is entitled to a new trial under Roche, because Hoover testified at his trials during the time frame of Hoover’s misconduct. The State contends that Shaw cannot establish that evidence of Hoover’s misconduct would probably change the result of the trial and that the evidence is merely impeaching.
¶19 If Hoover’s misconduct had been known at the time of trial, we think it is unlikely that the State would have called him as a witness or relied on his lab test, specifically because Hoover’s credibility was “totally devastated by his malfeasance.” Roche, 114 Wn. App. at 437. Instead, the State may have relied on the field tests and circumstantial evidence in each case. Both cases included evidence that Shaw arrived at appointments made for the purposes of doing drug sales and that Shaw attempted to avoid being caught with the evidence by stashing a baggie of drugs in the car in one case and throwing a baggie into a dumpster in the other. In one case Shaw admitted to being a cocaine user and seller and witnesses in both cases identified themselves as cocaine users and the substance in question as cocaine. Thus, Shaw cannot establish that without *167Hoover’s testimony, the results of his trials probably would have been different.
¶20 Shaw must also establish that the evidence would be material and more than merely impeaching. While evidence that Hoover began stealing and using heroin from the lab approximately one year after he conducted the tests in Shaw’s cases could certainly be used to impeach Hoover and undermine his credibility, it would not be material to the accuracy of the original tests performed at a time when no substantial evidence exists that Hoover was stealing and abusing heroin. Based on the lack of evidence to indicate that the tests were inaccurate, as well as the independent circumstantial evidence and field tests, Shaw’s case is distinguishable from Roche, such that his personal restraint petition is properly denied.
Brady Violation
¶21 Each petitioner also contends that he is entitled to a vacation of his conviction based on a violation of due process under Brady v. Maryland, 373 U.S. 83, 83 S. Ct. 1194, 10 L. Ed. 2d 215 (1963). A Brady violation includes three components: (1) evidence favorable to the accused because it is exculpatory or impeaching, (2) willful or inadvertent suppression of that evidence by the State, and (3) resulting prejudice. Brennan, 117 Wn. App. at 805 (citing Strickler v. Greene, 527 U.S. 263, 281-82, 119 S. Ct. 1936, 144 L. Ed. 2d 286 (1999)). Prejudice occurs “ ‘if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.’ ” Strickler, 527 U.S. at 280 (quoting United States v. Bagley, 473 U.S. 667, 682, 105 S. Ct. 3375, 87 L. Ed. 2d 481 (1985)).
¶22 Bain’s argument under Brady is foreclosed by United States v. Ruiz, 536 U.S. 622, 122 S. Ct. 2450, 153 L. Ed. 2d 586 (2002) (Constitution does not require prosecutors to make certain impeachment evidence available to defendants before defendants plead guilty). Brennan, 117 Wn. App. at 806. When Bain pleaded guilty, he gave up his right to impeach the State’s witnesses. Id.
*168¶23 Delmarter cannot establish a Brady violation because of the lack of evidence to indicate that Hoover’s misconduct began before the fall of 1998, long after he tested the substances in Delmarter’s cases in late 1996 and early 1997, and after Delmarter’s trial was complete in September 1997. Evidence of Hoover’s activities long after the conviction would not have changed the result of the trial.
¶24 Shaw argues that the State violated Brady because Hoover failed to disclose the fact that he was using drugs at the time he testified in Shaw’s trials.6 Assuming without holding that this is so, Shaw nevertheless fails to demonstrate a reasonable probability that the results of his trials would have been different. If Hoover’s misconduct had been known and revealed to the defense, we think it is unlikely that the State would have called him as a witness, despite the lack of evidence that he was using drugs at the time of the tests or that the tests were in any way inaccurate, because his credibility was simply devastated. Thus, the State likely would have been left with the field tests and circumstantial evidence. Although the defense would have been able to argue the weight the jury should give to the evidence, we are not persuaded that the result likely would have been different, even if Hoover had not testified. The July 24 case included testimony by two witnesses who were helping a third person arrange to buy cocaine. These witnesses testified that Shaw arrived with the people they had paged in order to buy the drugs and approached them. When the police interrupted, Shaw threw a baggie into a nearby dumpster. One witness, an admitted cocaine user, testified that he recognized the substance Shaw attempted to discard as cocaine. Shaw also testified at trial that he had used and sold cocaine before, but denied that the substance found in the dumpster belonged to him.
*169¶25 The September 2 incident involved an undercover officer’s attempt to arrange a drug deal, posing as a buyer named “Christy.” Shaw arrived at the scene in a car, but did not actually engage in a deal. When officers approached the car, they observed Shaw make furtive movements, appearing to reach into the back seat. After the arrest, they found a plastic baggie containing off-white rock-like chunks tucked into the crack between the rear seat-back and the seat. Another passenger in the car testified that he went with Shaw to sell cocaine to “Christy” and that the baggie contained cocaine belonging to Shaw.
¶26 Given this evidence, Shaw has failed to demonstrate a reasonable probability that the outcome of the trials would have been different, such that his Brady argument fails.
¶27 The personal restraint petition in each of these consolidated cases is dismissed.
Cox, C.J., and Becker, J., concur.
Reconsideration denied February 2, 2005.
Review denied at 154 Wn.2d 1024 (2005).
The facts of the first incident are described in an unpublished opinion affirming the trial court’s decision refusing to suppress evidence due to a warrantless search of Delmarter’s car. State v. Delmarter, noted at 93 Wn. App. 1062, 1999 WL 18415.
Shaw’s conviction on this charge was affirmed on direct appeal, where Shaw challenged the sufficiency of the evidence and admission of ER 404(b) evidence. State v. Shaw, noted at 106 Wn. App. 1060, 2001 WL 746457.
Shaw’s conviction on this charge was affirmed on direct appeal, where Shaw challenged the sufficiency of the evidence and argued that his right to due process was violated by the State’s destruction of evidence. State v. Shaw, noted at 105 Wn. App. 1060, 2001 WL 410679.
See State v. Bryant, 78 Wn. App. 805, 812-13, 901 P.2d 1046 (1995) (appellate court may consider trial court’s oral decision to the extent it is not inconsistent with trial court’s written findings and conclusions).
Because Bain does not challenge the reference court’s findings of fact, they are verities for his appeal.
Shaw also claims that because there was a strong inference that Hoover was using and stealing heroin in 1997 when he tested the materials at issue in Shaw’s cases, Hoover was required to disclose these activities. But the reference court found that the evidence supported only a finding that Hoover began his misconduct in the fall of 1998.