Consultant cardio-logist Dr Neil Grubb prepared a report based on medical information given about William Payne, 53, - that he had blacked out on four occasions - and it was "reasonable" he wasn't told to stop driving.
But after being told at the inquiry that Mr Payne told another doctor during a medical test for his HGV licence that he had not suffered blackouts Dr Grubb said that was "wrong".
Mr Payne, of Bishop-briggs, was driving his 4x4 on December 17, 2010, in North Hanover Street, Glasgow, when he collided with Mhairi Convy, 18, and Laura Stewart, 20. He also injured Mark Hopwood, 39, who survived.
Yesterday at Glasgow Sheriff Court, under cross-examination by Dorothy Bain QC, rep-resenting the families, Dr Grubb said that he would have ordered a tilt table test, which diagnoses fainting, as early as April 2008.
The inquiry heard that when Mr Payne had the test after the incident his heart paused for 10 seconds.
Dr Grubb said that it was "entirely possible" that if the test was carried out in April 2008 that it would have had the same result.
Miss Bain asked the witness: "If it had the same result you would have told him to stop driving?" Dr Grubb replied: "Yes."
She said: "What happ- ened to Miss Stewart and Miss Convy could have been avoided."
The consultant said: "Yes, if he had stopped driving, yes."
The doctor was told Mr Payne had suffered more blackouts than he was aware of before compiling his report.
Miss Bain told him that under the question to renew his HGV licence in July 2010 "Is there a history of blackouts or impaired consciousness in the last five years" that Mr Payne said no.
She asked: "That's wrong isn't it?" Dr Grubb said: "That's definitely wrong."
The QC asked: "If he told the truth in July 2010 would he be driving?" Dr Grubb said: "He certainly wouldn't be driving an HGV and I believe he probably wouldn't be driving his own car."
Miss Bain said: "If he told the truth the girls wouldn't have been killed?" The consultant said: "That's correct."
The inquiry, before sheriff Andrew Normand, continues.