¶27
Madsen, J.(concurring) — I agree with the majority that the regulations in effect at the time of Ludvigsen’s breath test apply in determining its validity. This result is compelled by a plain reading of the statute, which provides *675a breath test “shall have been performed according to methods approved by the state toxicologist.” RCW 46.61-.506(3) (emphasis added).13 The statute plainly indicates the legislature intends breath tests to be performed in accordance with methods approved at the time of the test, not later.14 At the time Ludvigsen was given his breath test, the National Institute of Standards and Testing (NIST) traceability requirement was part of the regulation defining a valid breath test. Former WAC 448-13-035, -040 (2001); Cannon v. Dep’t of Licensing, 147 Wn.2d 41, 58, 50 P.3d 627 (2002) (interpreting 2001 amendment to approved breath test procedures); City of Seattle v. Clark-Munoz, 152 Wn.2d 39, 50, 93 P.3d 141 (2004) (same). Following our decisions in Cannon and Clark-Munoz, Ludvigsen’s breath test is invalid as a matter of law due to the State’s failure to comply with the NIST traceability requirement. Thus, I concur in • the result.
¶28 I write separately to observe the result in Clark-Munoz, and here, was compelled by the unusual circumstance that the toxicologist incorporated the NIST traceability requirement into the regulation defining a valid breath test. The insertion of the NIST traceability requirement was unusual in that, previously, the regulation defining a valid test pertained only to standards that must be met during the actual administration of the test and did not include tangential machine maintenance procedures. See City of Seattle v. Allison, 148 Wn.2d 75, 83, 59 P.3d 85 (2002) (discussing foundational requirements under former regulations). Once the NIST traceability requirement became part of the regulation defining a valid test, however, the State was required to prove compliance in order to *676establish the test’s validity, following this court’s decisions in Cannon and Clark-Munoz.
¶29 The majority’s analysis troubles me because while it recognizes the State no longer must prove NIST traceability, it appears to perpetuate the notion the State must prove compliance with other machine maintenance protocols in order to establish the validity of a test result. This is inconsistent with the legislature’s clear intent to remove barriers to effective enforcement of our drunk driving laws by disallowing admissibility challenges based on technical deficiencies having no effect on the accuracy of a test result. See Laws of 2004, ch. 68, § 1.15
ANALYSIS
¶30 Before Clark-Munoz, the touchstone for this court in evaluating the admissibility and validity of a breath test result was whether the State provided adequate assurance of the reliability and accuracy of the test. Thus, in State v. Baker, 56 Wn.2d 846, 852, 355 P.2d 806 (1960), this court derived its original foundational requirements from assertions by the State’s experts that a breath test is “wholly unreliable” unless (1) the machine was properly checked and in proper working order at the time of the test, (2) the correct chemicals were used, (3) the test subject had nothing in his or her mouth at the time and had not ingested anything within 15 minutes before taking the test, and (4) *677a properly trained operator administered the test in the proper manner.
¶31 Following Baker, this court consistently held once the State presents prima facie evidence of the foundational requirements, any challenges to the reliability and accuracy of the test go to the weight of the test result, not its admissibility. Allison, 148 Wn.2d at 86 (breath test ticket satisfies foundational requirements; “arguments as to the reliability of the particular test results are questions for the jury”); State v. Wittenbarger, 124 Wn.2d 467, 476, 880 P.2d 517 (1994) (defendants may challenge reliability of breath tests through cross-examination, expert testimony, and independent tests); State v. Straka, 116 Wn.2d 859, 875, 810 P.2d 888 (1991) (deviations from procedures for evaluating and certifying the machines, and for mixing the simulator solution, may be introduced to refute the accuracy and reliability of the test results but do not bar its admissibility); State v. Brayman, 110 Wn.2d 183, 192, 751 P.2d 294 (1988) (“[t]he defendant may introduce evidence refuting the accuracy and reliability of the test reading”); State v. Peterson, 100 Wn.2d 788, 791-92, 674 P.2d 1251 (1984) (“[a]ny challenge to the reliability of the Breathalyzer reading goes to its weight rather than to its admissibility”; declining to adopt more stringent foundational requirements following adoption of the per se offense); City of Bremerton v. Osborne, 66 Wn.2d 281, 282, 401 P.2d 973 (1965) (challenges to the operator’s qualifications and sufficiency of machine maintenance procedures go to weight rather than admissibility); Baker, 56 Wn.2d at 853-54 (State’s failure to use a reference thermometer to calibrate Breathalyzer chamber and test ampoule does not bar admission of test result where there is no indication the machine thermometer malfunctioned).
¶32 The state toxicologist codified the Baker foundational requirements following the adoption of the implied consent statute, which provides a breath test is valid if performed “according to methods approved by the state toxicologist.” RCW 46.61.506(3); Brayman, 110 Wn.2d at *678183-84; State v. Franco, 96 Wn.2d 816, 819-20, 639 P.2d 1320 (1982).
¶33 After the Breathalyzer was replaced with the BAC Verifier DataMaster (DataMaster), the toxicologist adopted a streamlined set of breath test regulations, revised its record keeping policies, and formulated the bulk of the machine maintenance and simulator solution protocols as internal policies and procedures rather than formal administrative rules. Defendants challenged the changes as embodying a “prosecutorial bias” designed to facilitate convictions. Wittenbarger, 124 Wn.2d at 477. This court rejected the challenges, after finding the new regulations, policies, and procedures sufficiently assured the accuracy and reliability of breath test results. Wittenbarger, 124 Wn.2d 467 (State not required to preserve detailed machine maintenance records); Straka, 116 Wn.2d 859 (State not required to preserve invalid test messages); State v. Ford, 110 Wn.2d 827, 755 P.2d 806 (1988) (affirming toxicologist’s approval of DataMaster); see also State v. Schulze, 116 Wn.2d 154, 804 P.2d 566 (1991) (toxicologist need not adopt a “ ‘cookbook’ ” of protocols and procedures for administering alcohol concentration tests but may, instead, adopt general criteria).
¶34 In evaluating the propriety of the changes, this court stated the judiciary’s “ultimate concern” is that the approved methods “result in an accurate test, competently administered, so that a defendant is assured that the test results do in fact reflect a reliable and accurate measure of his or her breath content.” Ford, 110 Wn.2d at 833.
¶35 In approving the streamlined regulations, this court observed that the DataMaster, unlike the Breathalyzer, is self-certifying. Wittenbarger, 124 Wn.2d at 483. Thus, the breath test ticket itself is a “crucial document” that evidences the machine was operating properly on a particular occasion. Id.; see also Allison, 148 Wn.2d at 83 (breath test ticket provides prima facie evidence of compliance with approved breath test procedures).
¶36 Importantly, the “relevant procedures” for determining whether a breath test was performed “ ‘ “according to *679methods approved by the state toxicologist” ’ ” are those pertaining to the actual administration of the test. Allison, 148 Wn.2d at 80 (emphasis omitted) (quoting Ford, 110 Wn.2d at 833 (quoting RCW 46.61.506(3))).16
¶37 The toxicologist adopted a separate set of regulations, internal policies, and procedures relating to machine maintenance and the preparation of the simulator solution used in the test. See former WAC 448-13-110, -170, -080, and -160 (2001). Proof of compliance with this set of regulations was not a condition precedent to admissibility. Wittenbarger, 124 Wn.2d at 489-90; State v. Mee Hui Kim, 134 Wn. App. 27, 139 P.3d 354 (2006); Smith v. Dep’t of Licensing, 88 Wn. App. 875, 944 P.2d 1117 (1997). Although “approved” by the toxicologist, these protocols and procedures represent best practices that exceed the minimum requirements that must be met to establish the admissibility and validity of a test result.17
¶38 The thermometer certification requirement became part of the regulation defining a valid test in 2001. Under former WAC 448-13-040 (1999), the operator may not proceed with the test unless the temperature of the simulator solution is “thirty-four degrees centigrade plus or minus 0.2 degrees centigrade.” As proof of compliance with the requirement, the State presented evidence the breath test operator visually confirmed the temperature was within *680this range by reading the thermometer attached to the simulator, then entering “yes” in response to a question on the DataMaster. Allison, 148 Wn.2d at 82-83. Some defendants successfully moved to suppress breath tests results on the ground the State did not sufficiently establish compliance with the regulation. They argued, and trial courts agreed, the State must prove the actual temperature was within this required range, and that such proof was impossible due to a thermometer variance of 0.1 centigrade. Id. at 78. This court reversed the suppression rulings, interpreting the criterion as requiring proof the operator visually confirmed the thermometer reading, not proof of the actual temperature.18 Id. at 86.
¶39 However, before this court decided Allison, the toxicologist amended the regulation to clarify his intention. Wash. St. Reg. 01-17-009 (Sept. 2, 2001). The amended regulation provided, “the operator must verify” the temperature is within the required range, using a thermometer “certified per WAC 448-13-035.” Former WAC 448-13-040 (Supp. 2002). Former WAC 448-13-035 (Supp. 2002) required the annual certification of simulator thermometers “using a reference thermometer traceable to standards maintained by the National Institute of Standards and Testing (NIST).”
¶40 In Cannon, this court concluded former WAC 448--13-040 required the State to prove thermometer certification in order to establish the validity of the breath test, rejecting the State’s argument that “certified per WAC 448-13-035” was merely a descriptive phrase that did not add to the State’s foundational burden. Cannon, 147 Wn.2d at 59. Thus, for the first time, proof of compliance with a specific machine maintenance regulation became a foundational requirement.
¶41 In Clark-Munoz, this court shifted focus from accuracy and reliability to “fair play.” Based on the principle that *681the State must “abide by its own rules,” this court held breath test results were invalid, as a matter of law, even though the record was devoid of evidence that noncompliance with the thermometer traceability requirement affected the accuracy or reliability of any breath test result. Clark-Munoz, 152 Wn.2d at 50. Compare Wittenbarger, 124 Wn.2d at 483 (failure to comply with regulation requiring periodic certification of testing machine did not require suppression of test results made on a DataMaster, which, unlike the older Breathalyzer, is self-certifying); City of Seattle v. Rainwater, 86 Wn.2d 567, 546 P.2d 450 (1976) (failure to follow one and one-half minute waiting period at step 7 of new 11 step procedure did not invalidate a test made on an older model machine where the evidence showed the deviation had no effect on an accurate reading); compare Baker, 56 Wn.2d at 852-53 (observation of test subject for only 14 minutes invalidated test where State’s expert stated test is “wholly unreliable” absent at least a 15 minute waiting period).
¶42 In response to Clark-Munoz, the legislature amended the drunk driving statute to mandate that breath tests are admissible when the State produces prima facie evidence of compliance with a codified set of foundational requirements. Laws of 2004, ch. 68, § 4. The 2004 amendment was intended to eliminate challenges to breath test admissibility based on technical deficiencies not shown to adversely affect the accuracy of the result. City of Fircrest v. Jensen, 158 Wn.2d 384, 399, 143 P.3d 776 (2006). In Jensen, we rejected a constitutional challenge to the amending legislation, holding it does not violate the separation of powers doctrine. Id. We observed the 2004 amendment merely codified the foundational requirements for breath tests as developed by case law and represented the legislature’s intent “to return the requirements of BAC test admissibility to the way it was before our holding in Clark-Munoz.” Id. at 396-97.
¶43 As amended in 2004, RCW 46.61.506(4) prohibits the suppression of breath test results based on technical deficiencies that do not adversely affect the accuracy or *682reliability of the test result. Breath tests now are treated like other scientific evidence: once the State satisfies the foundational requirements, test results generally are admissible. Jensen, 158 Wn.2d at 397 (citing State v. Copeland, 130 Wn.2d 244, 254, 922 P.2d 1304 (1996)). In codifying the foundational requirements, the legislature sought to achieve greater uniformity in admissibility determinations and make our drunk driving laws more effective. Laws of 2004, ch. 68, § 1. When deviations from additional testing procedures or machine maintenance protocols are so serious as to render test results unreliable, a court has discretion to exclude them in accordance with the rules of evidence. See Jensen, 158 Wn.2d at 398-99 (following enactment of Substitute House Bill 3055, courts continue to have discretion to exclude breath test results on grounds of irrelevancy and undue prejudice); Copeland, 130 Wn.2d at 270 (test results may be excluded under ER 702 as “not helpful to the jury” when laboratory error renders the results unreliable). However, ordinarily, such deviations go to the weight, not the admissibility, of the test results. Id. at 271-72; State v. Kalakosky, 121 Wn.2d 525, 541, 852 P.2d 1064 (1993); State v. Cauthron, 120 Wn.2d 879, 890, 846 P.2d 502 (1993).
CONCLUSION
¶44 Following Cannon and Clark-Munoz, the State’s failure to prove compliance with the thermometer certification protocol renders Ludvigsen’s breath test invalid as a matter of law. Thus, I concur in the result.
¶45 However, our legislature has now mandated that breath test results will be suppressed due to procedural deficiencies only if the deficiencies are shown to adversely affect the accuracy or reliability of a breath test. Once the State produces prima facie evidence of compliance with the statutory foundational requirements, breath test results are admissible, with all other challenges to the test’s accuracy or reliability going to the weight of the evidence. See ER 401. 402. 403. 702.
See also former WAC 448-13-060(5) (2001 & Supp. 2002) (“the criteria applied to determine the validity of any test and so certify it, should be those provisions of the Washington Administrative Code in effect at the time the test is administered”).
See Poston v. Clinton, 66 Wn.2d 911, 915-16, 406 P.2d 623 (1965) (holding an amended statute governing breath test procedures operates prospectively since the legislature intended a breath test to be administered in compliance with then-applicable procedures).
“The legislature finds that previous attempts to curtail the incidence of driving while intoxicated have been inadequate. The legislature further finds that property loss, injury, and death caused by drinking drivers continue at unacceptable levels. This act is intended to convey the seriousness with which the legislature views this problem. To that end the legislature seeks to ensure swift and certain consequences for those who drink and drive.
“To accomplish this goal, the legislature adopts standards governing the admissibility of tests of a person’s blood or breath. These standards will provide a degree of uniformity that is currently lacking, and will reduce the delays caused by challenges to various breath test instrument components and maintenance procedures. Such challenges, while allowed, will no longer go to admissibility of test results. Instead, such challenges are to be considered by the finder of fact in deciding what weight to place upon an admitted blood or breath test result.” Laws op 2004, ch. 68, § 1.
Under the former regulations, these were codified in former WAC 448-13-060 (2001) (validity and certification of test results), which incorporated by reference former WAC 448-13-040 (administration of breath test on the DataMaster), former WAC 448-13-050 (2001) (test defined), and former WAC 448-13-055 (2001) (interference with breath test). Former WAC 448-13-060 provided, “[a] test shall be a valid test” if certain criteria are met. It further provided, “[i]f these criteria are met, then these and no other factors are necessary to indicate the proper working order of the instrument, and so certify it, at the time of the breath test.”
See Washington State Patrol, Breath Test Section, Breath Test Program Policy and Procedure Manual, Purpose Statement: “This notebook should contain the most recent written policies and procedures followed by personnel within the Breath Testing Program. Generally, these policies and procedures exceed the requirements of the Washington Administrative Code and reflect an attempt to ensure the highest possible confidence in the Breath Testing Program.” available at http://breathtest.wsp.wa.gov/default.asp (follow link to BTP Public Records Index) (last visited Dec. 19, 2007).
The regulations applicable in Allison did not include the thermometer certification requirement at issue in Cannon and Clark-Munoz.