A state trooper arrested Mark Anson for OWI and took him to the Onawa police station for implied consent procedures. Anson telephoned his lawyer, who came to the police station. Por a period of about twenty-six minutes, Anson and the lawyer conferred in a closed room. They could not be observed by police officers. Sometime during this period Anson claims to have taken a drink from the attorney’s cherry cola can while the attorney was spitting his chewing tobacco into the wastebasket.
Anson and the attorney were left in the closed room for about twenty-six minutes. Then an officer knocked on the door and asked whether Anson would consent to chemical testing. Anson consented. An intoxilyzer test was performed. The officers did not ask Anson whether he had eaten, had drunk any liquid, or had smoked for at least fifteen minutes before the test. As noted above, the officers also had not observed Anson continuously for at least fifteen minutes before the test.
On the basis of the intoxilyzer test result, the Department of Transportation revoked Anson’s driver’s license. Upon judicial review of an agency action pursuant to Iowa Code chapter 17A, the district court upheld the revocation. The district court noted Anson was unsure whether his alleged drink of cherry cola occurred more or less than fifteen minutes prior to the intoxi-lyzer test. The district court held Anson had failed to prove the drink of cherry cola had affected the intoxilyzer result.
Anson has appealed. Anson contends the revocation should be overturned because the officers failed to comply with DOT regulations stating procedures for in-toxilyzer tests. Specifically, he complains the officers failed to observe him for at least fifteen minutes before the test and also failed to ascertain whether he had eaten, drunk or smoked during that period. He argues the officer’s failure to follow these procedures, when combined with his alleged consumption of liquid during the critical period, renders the intoxilyzer test result unreliable and requires the license revocation be overturned. He also argues the district court erred by imposing on him the burden to prove his drink of cherry cola had affected the test result. Anson suggests the officers’ failure to follow procedures is sufficient by itself to invalidate the license revocation. He asserts proof of an actual distortion in the test results should not be required.
Iowa Code section 17A.20 provides for appellate court review of a district court judgment in a judicial review of a final agency action. This court’s scope of review is for correction of errors of law. Busing v. Iowa Dept. of Transportation, Motor Vehicle Division, 455 N.W.2d 921, 922 (Iowa 1990).
In a civil administrative driver’s license revocation proceeding, the burden of proof is on the licensee to show compliance with all lawful requirements for retention of the license. Iowa Code § 17A. 18(3); McCrea v. Iowa Dept. of Transportation, 336 N.W.2d 427, 429 (Iowa 1983). A reviewing court may not interfere with the agency’s factual findings unless the licensee has met his burden as a matter of law. Heidemann v. Sweitzer, 375 N.W.2d 665, 670 (Iowa 1985).
I. Observation Period.
Anson argues his license revocation should be rescinded because peace officers did not continuously observe him for the fifteen-minute period immediately preceding breath testing procedures. State Trooper Chris Rich requested a sample of Anson’s breath for chemical testing at 1:06 a.m. and waited for approximately twenty minutes until appellant’s attorney arrived for a consultation with his client. Trooper Rich entrusted his prisoner to the attorney from 1:25 a.m. to 1:51 a.m. while they conferred alone in a conference room where Anson could not be viewed from the outside by police personnel. Anson was given the breath test a few minutes after the attorney consultation had been concluded.
In Kooi v. Commissioner of Public Safety, 363 N.W.2d 487, 489 (Minn.App.1985), the court explained the purpose of the observation period:
*697The purpose of the observation period is to preclude the possibility of mouth alcohol affecting the test result — mouth alcohol which is the result of burping or vomiting.
It is not always necessary to show a fifteen-minute period of continuous observation. State v. Hershey, 348 N.W.2d 1, 3 (Iowa 1984). Any reasonable likelihood an arrestee smoked, ate, or drank can be negated by circumstantial as well as direct evidence. Id.
In this case, the officer’s observation of Anson was interrupted for approximately twenty-five minutes before testing while Anson conferred in private with his attorney. The attorney brought a can of soda pop into the conference room where he allowed, even if inadvertently, Anson to have a drink of it.
While Anson may have had a drink of soda pop, there is no evidence that occurred within the fifteen-minute period prior to testing. Anson could not recall when in the period from 1:25 a.m. until 1:51 a.m. he drank the pop. He conceded it could have been more than fifteen minutes before the test was administered. The time the substance was ingested is pure speculation and falls far short of proof.
II. Burden of Proof.
The purpose of procedural standards is to insure the accuracy of the test for use in later judicial proceedings. State v. Schlemme, 301 N.W.2d 721, 723 (Iowa 1981). Even if an arrestee is not continually observed, he must present evidence the test results were affected because he smoked, ate, or drank prior to administration of the chemical test. Morgan v. Iowa Dept. of Transportation, 428 N.W.2d 675, 679 (Iowa App.1988). To successfully challenge intoxilyzer test results, a driver must do more than merely speculate something might have occurred to invalidate those results. Hounsell v. Commissioner of Public Safety, 401 N.W.2d 94, 96 (Minn.App.1987).
In this case, Anson presented no testimony on the amount of soda pop he drank, nor precisely the time when that occurred. More important, however, he presented no expert testimony or any other evidence to show a drink of soda pop prior to taking a breath test would have affected the test results. Anson argues he should not be required to present expert testimony because it “would require a great deal of expense and time to introduce scientific testimony with experts.” The purpose of procedural standards is not to provide drunken drivers with a technical defense, but to ensure the accuracy of the test for use in judicial proceedings against them. Schlemme, 301 N.W.2d at 723. If Anson believes a drink of soda pop somehow affected the accuracy of the test results, it is neither unreasonable nor unfair to expect him, who has the burden of proof, to present some evidence to support that claim.
Other courts have rejected similar claims. See, e.g., Melin v. Commissioner of Public Safety, 384 N.W.2d 474 (Minn.App.1986) (even if driver can prove he had been smoking when his vehicle was stopped, he must demonstrate the substance actually affected his test results); Pasek v. Commissioner of Public Safety, 383 N.W.2d 1 (Minn.App.1986) (driver did not successfully show chewing tobacco and snuff affected test results); Hager v. Commissioner of Public Safety, 382 N.W.2d 907 (Minn.App.1986) (driver failed to show chewing gum removed five minutes before test administered affected test results or officer should have waited longer before administering test).
The district court in this case found An-son failed to prove his drink of soda pop could have affected the test results. The court did not err in its finding nor in its interpretation of the law. We affirm the ruling of the district court.
AFFIRMED.