JUSTICE DOYLE
delivered the opinion of the court:
Defendant, Bob Allan, was indicted in the circuit court of McHenry County for one count of reckless homicide (111. Rev. Stat. 1989, ch. 38, par. 9 — 3(a)). Following a jury trial, defendant was found guilty and sentenced to a two-year term of imprisonment.
On appeal, defendant raises the following issues: (1) whether the jury was incorrectly instructed in such a manner that it was required to find that he acted recklessly if it found his blood-alcohol content to be .10% or more; (2) whether he was denied a fair trial when the prosecutor failed to complete impeachment by introducing a prior inconsistent statement that a defense witness denied making; and (3) whether the trial court abused its discretion in considering the death of the victim as an aggravating factor in sentencing.
The following facts are relevant to our disposition of the issues raised on appeal. On June 29, 1988, at about 10:50 p.m., Jamie Dass, the deceased, was driving his vehicle westbound on Illinois Route 120 in McHenry, Illinois. Defendant was driving alone eastbound on Route 120 approaching McHenry. Witnesses established that defendant’s vehicle entered Dass’ lane of travel for approximately 60 feet and travelled straight before essentially striking Dass’ vehicle head on.
Route 120 is a four-lane roadway at the point of the collision, with two lanes running east and two lanes running west. There is also a concrete median separating the two east and westbound lanes which is one to two inches high. There is a solid yellow line that runs on each side of the median, a white, dotted line separating each of the east and westbound lanes, and a white, solid line running along the outer edge of the outside east and westbound lanes. The shoulders are paved.
At the point where Route 120 enters McHenry from the west, it is a two-lane highway with the lanes separated by a double yellow line. The road widens from two lanes to four lanes at Dot Street. Prior to the road becoming four lanes, there is a sign for eastbound traffic which states “[d]o not drive on the shoulder.” There is no sign notifying eastbound motorists that the road widens from two lanes to four lanes, while there is a sign indicating to westbound traffic that the road narrows from four lanes to two lanes.
There are mercury vapor streetlights on the north side of Route 120 which illuminate all four lanes of the road. There is a Kinder Care facility on the south side of Route 120 which one of the police officers testified was lighted on the evening of the accident. Its floodlights light up its parking lot. There is a farm field to the south of this stretch of Route 120, and there are no lights to the south until Kinder Care. One of the police officers who was at the accident scene testified that he could see approximately one-half mile west of the accident scene.
Defendant testified that he was employed as a research and development engineer at the time of the accident. He had flown from Minnesota to the area for the purpose of working on a project for one of his employer’s customers. On the day of the accident, he awoke with four or less hours of sleep. He flew to O’Hare Airport and rented a car. He drove directly to the customer’s facility in Northbrook, arriving at about 9 a.m. He worked at that location until between 5 and 5:30 p.m.
During the day, defendant made plans with Ron Nordin, the customer account manager, to go to Nordin’s office located at his home near Hebron. The purpose of the meeting was to discuss work-related events for the following day.
Defendant had never been to Nordin’s residence before and received instructions on how to get there. He was able to locate Nor-din’s home, arriving at between 7 and 7:30 p.m. Defendant did not eat at Nordin’s but had lunch earlier. Defendant testified that he brought two six-packs of beer and consumed about four beers while there, the last one being just before he left. Nordin, his wife, and another employee also consumed some beer. Defendant denied consuming “hard liquor” at Nordin’s.
He departed Nordin’s at approximately 10:30 p.m. intending to find a motel close to Northbrook. Nordin gave him oral instructions on how to get back to Northbrook. Defendant believed he was not under the influence of alcohol at the time he received the instructions and that he had sufficient control of his faculties to drive a motor vehicle. As he proceeded east on Route 120, he came to an intersection with a green traffic light. There was a 45-mile-per-hour speed limit sign and a sign stating to not drive on the shoulder. Defendant was looking for signs indicating that he was on Route 120.
After he passed through the intersection and a curve, he noticed that all of the lighting was on his left. At that point, the westbound traffic appeared to be passing by him farther to the left than it had prior to his going through the intersection and curve. He became confused and thought that he was too far to his right on a two-lane road. He drove into the lane immediately to his left. He then proceeded eastbound in that lane of traffic for a very short time. He believed he was still on a two-lane highway at that time. He did not know at the time of the accident that the concrete median divided the two east and two westbound lanes.
After proceeding in that lane for a very short time, he recognized oncoming traffic that he believed was in his lane. He slammed on his brakes in an effort to avoid an accident and had a head-on collision. His only reason for going into the other lane was confusion. There was no sign indicating that the road had become four lanes.
Defendant testified that he was not feeling any effects of the four beers at that time. He did not believe his consumption of the beer in any way impaired his driving ability that evening. Defendant described the lighting conditions as very dark. The lighting was primarily to the north, and he did not notice any lights in the Kinder Care parking lot at that time.
Defendant was transported to a hospital where he received treatment from a plastic surgeon for the laceration on his face. He received injections of a pain-deadening substance in his nasal area. At that time, he was interviewed by a police officer. According to defendant, he was able to identify for the officer the day of the week, the date, what road he was on, and which direction he was travelling. He told the officer he was on a two-lane highway at the time of the accident. He also told the officer he had a few beers.
On cross-examination, defendant admitted that he had traveled that same portion of Route 120 through McHenry earlier that day when he drove to Nordin’s. It was daylight when he did so. He did not remember the roadway having four lanes or being divided by a concrete median. As he was traveling east into McHenry later that evening, he did not notice the lanes or pavement being wider, nor did he notice a concrete median. He also did not see any lines to his right on the pavement. He admitted that there was a solid yellow line on his side of the median and that he intentionally drove across the median into the other lane. He believed the median was part of the shoulder of the road.
On redirect examination, defendant testified that people had told him earlier that day that the roads in that area twisted and curved and split and that he should be careful because it was confusing. That thought was on his mind as he drove eastbound and observed the oncoming traffic and lights to his left.
Ron Nordin testified on behalf of defendant that he knew defendant as a co-worker but was not a personal or social friend of defendant. According to Nordin, defendant arrived at his house at about 6:15 or 6:30 p.m. and left sometime after 10 p.m. Beer was consumed that evening, but he could not say how many beers defendant had. Nordin did not have an opinion as to whether defendant was under the influence of alcohol when he left his home the night of the accident. Defendant did appear to have control of his faculties and did not appear to be staggering or unable to keep his balance. Nordin had no fear that defendant was unable to drive.
On cross-examination of Nordin, the following colloquy occurred:
“[Mr. Browder:] Q. And isn’t it also correct that that evening you yourself had consumed brandy?
A. No.
Q. You never consumed any brandy that evening?
A. No.
Q. Do you remember speaking with an attorney from our office, Michelle Wiejacka, in June of ’89 about this event?
A. Yes.
Q. Do you remember telling her what happened that evening?
A. Yes. Not everything, that’s two years ago or a year ago.
Q. Okay.
So if you were to have told her that you were drinking brandy on that day, you might have forgotten since then?
A. I don’t think I would have forgotten then, no.
Q. So it is your testimony that you did not consume any brandy that evening?
A. Not that I can recall, no.
Q. When you say ‘not that I can recall,’ then it is possible that you did?
A. I don’t think so, no.
Q. So it is your testimony that, no, you did not consume any brady [sic]?
A. Yes.”
On redirect examination, Nordin testified that defendant did not exhibit any outward signs of intoxication nor did he ever tell defendant that defendant had had enough or to not have any more beer. On re-redirect, he added that he did not know if the beer defendant consumed affected defendant’s ability to carry on a conversation, but that defendant was responsive to questions and spoke cogently throughout the evening.
Dr. George Gallant, the emergency room physician who treated defendant, testified that he noted the odor of alcohol on defendant’s breath in the emergency room. Officer Thomas Beyer of the McHenry police department testified that he first spoke with defendant at 1:44 a.m. on June 30, 1988, in the emergency room. Defendant told him he was driving on a two-lane highway at the time of the accident. Officer Beyer observed that defendant’s breath had a strong odor of alcohol, that defendant’s eyes were bloodshot and a little glassy, and that defendant’s speech was slurred. Based on his experience of observing hundreds of intoxicated people, Officer Beyer opined that defendant was under the influence of alcohol.
At about 2:35 a.m., Officer Beyer asked defendant to submit to a blood test, and defendant said he would. Carl Fletcher, a medical technologist, testified that he drew the blood pursuant to the officer’s request. He had previously drawn blood from defendant at between 11:45 p.m. and midnight for medical purposes. The results of the analysis from the first blood sample indicated a blood serum alcohol level of .183%.
Mr. Joseph Principe, a forensic scientist with the Northern Illinois Police Crime Laboratory, testified that he performed two analyses on the blood sample taken from defendant pursuant to Officer Beyer’s request. The results were .108 and .107% blood alcohol. The third digit is customarily dropped resulting in an alcohol content reading of .10% for the entire sample.
Mr. Michael Caplis, director of an analytical research laboratory, testified that the difference between a blood serum alcohol content and a whole blood-alcohol content is that the blood serum level would be falsely elevated. The serum ratio to whole blood is 1.18 to 1. Applying that ratio, a .183% serum blood alcohol would compute to approximately a .15% whole blood alcohol. Mr. Caplis opined that a blood alcohol of .10% would indicate that a person weighing 170 pounds and being normal and healthy would have the equivalent of four 12-ounce beers present in his system. That same person with a .15% blood alcohol would have the equivalent of between five and six 12-ounce beers in his system. Caplis further opined that, based on the blood alcohol of .10% at 2:35 a.m. and the blood alcohol of .15% at 11:40 p.m., defendant’s blood alcohol would have been more than .10% at 10:55 p.m. On cross-examination, Caplis conceded that if you applied a 1.3 to 1 ratio, a .183% blood serum alcohol would compute to .128% whole blood alcohol. On redirect examination, Caplis stated that even if he used the lower figure of .128%, he would still opine that the blood-alcohol level would be over .10% at 10:55 p.m.
The jury received the following instructions relevant to this appeal:
“If you find that at the time the defendant drove a vehicle that the amount of alcohol concentration in the defendant’s blood or breath was .10 per cent or more, you may infer that the defendant was under the influence of alcohol. You never are required to make this inference. It is for the jury to determine whether the inference should be drawn. You should consider all of the evidence in determining whether the defendant was under the influence of alcohol.” (People’s Instruction No. 11.)
“Being under the influence of alcohol at the time of the alleged violation shall be considered prima facie evidence of a reckless act.” (People’s Instruction No. 12.)
“A person is under the influence of alcohol when, as a result of drinking any amount of alcohol, his mental or physical faculties are so impaired as to reduce his ability to think and act with ordinary care.” (People’s Instruction No. 13.)
“Prima facie evidence is evidence which is sufficient to authorize a finding on a matter in issue unless contradicted or explained.” (Court’s Instruction No. 1.)
“A person acts recklessly when he consciously disregards and [sic] unjustifiable risk that circumstances exist or that a result will follow, and such disregard constitutes a gross deviation from the standard of care which a reasonable person would exercise in the situation.” (People’s Instruction No. 8.)
Defendant was found guilty of reckless homicide. At the sentencing hearing, the trial judge stated:
“In reviewing the statutes, the aspects of aggravation and mitigation, there is no question that Mr. Allan, in terms of his past history, essentially has none outside of traffic offenses.
Looking at the statute also, as it talks about aggravation, it addresses itself to the aspect of causing bodily harm, and there is no question there was great bodily harm caused here, i.e., the death of a young man.
And the State advances the position of the sentencing must be a deterent [sic].
I, too, went to a wake many years ago for a young man who was killed in a diving accident where he broke his neck and he died. I knew the family well, and tried to understand the ways of life. It was very difficult.”
The trial court sentenced defendant to two years’ imprisonment.
Defendant’s first contention is that the jury was instructed in such a manner that it was required to find, absent evidence to the contrary, that he acted recklessly if it found his blood-alcohol content to be .10% or more, and that such instruction denied him his right to be proved guilty beyond a reasonable doubt of the offense of reckless homicide. He further maintains that a finding of .10% blood alcohol cannot properly be the basis of a finding of recklessness because a person cannot consciously disregard his blood-alcohol content without knowing what that content is at the time he drives an automobile. It is apparent from a close reading of defendant’s appellate briefs that he has focused on the issue of whether it was proper for the jury to base a finding of recklessness on its preliminary finding, alone, that his blood-alcohol content was .10% or more. This issue may be addressed by examining the instructions in this case to determine what the jury was actually instructed to do concerning its finding of .10% blood alcohol. We need not address the content of the jury instructions except as they relate to the finding of blood-alcohol content and what the jury was allowed to further find based on the .10% blood-alcohol finding.
While section 9 — 3(c)(1) states that “a person shall be considered to be under the influence of alcohol” if the alcohol concentration in his or her blood is .10% or more (Ill. Rev. Stat. 1989, ch. 38, par. 9— 3(c)(1)), the jury in this case was instructed that it “may” infer defendant to be under the influence if it found defendant’s blood-alcohol concentration to be .10% or more but that it was not required to make such inference. The jury was also instructed that it should consider “all of the evidence” in determining whether defendant was under the influence of alcohol. The jury was not instructed that, if it found defendant’s blood alcohol to be .10%, such a finding alone should or could lead to a finding of recklessness.
The purpose of jury instructions is to guide the jury in deliberating and to assist it in reaching a proper verdict through the application of legal principles as applied to the evidence and the law. (People v. Hester (1989), 131 Ill. 2d 91, 98.) Instructions should be considered in their entirety rather than in isolation. Hester, 131 Ill. 2d at 98.
Although an accused may be convicted of a criminal offense only by proof beyond a reasonable doubt, the State may rely on presumptions or inferences. (Hester, 131 Ill. 2d at 98.) Inferences and presumptions play a vital role in the expeditious resolution of factual issues, with the value of the presumption or inference resting on the strength of the connection between the elemental or ultimate fact presumed or inferred and the basic or evidentiary fact. (Hester, 131 Ill. 2d at 98.) The use of a presumption or inference must not invade the province of the fact finder to determine ultimate or elemental facts beyond a reasonable doubt. Hester, 131 Ill. 2d at 98-99.
There are essentially two types of presumptions: mandatory and permissive. (Hester, 131 Ill. 2d at 99.) A mandatory presumption requires the trier of fact to find the elemental or ultimate fact upon proof of the basic fact, at least unless the defendant has come forward with some evidence to rebut the presumed connection between the two facts. Hester, 131 Ill. 2d at 99.
On the other hand, a permissive presumption allows the fact finder to either accept or reject the suggested presumption. (Hester, 131 Ill. 2d at 99.) It places no burden on the defendant and affects the application of the beyond a reasonable doubt test only if, under the facts of the case, there is no rational way the trier of fact could make the connection permitted by the inference. (Hester, 131 Ill. 2d at 99-100.) Hence, the validity of a permissive presumption is tested by whether there is a rational connection between the facts proved and the facts presumed, and the ultimate fact must be more likely than not to follow from the basic fact. (Hester, 131 Ill. 2d at 100.) If there is no corroborating evidence to support the inference, however, the connection between the proved fact and the presumed elemental fact must still be proved beyond a reasonable doubt. Hester, 131 Ill. 2d at 100.
To determine whether a presumption contained in a jury instruction is mandatory or permissive, a court must look at the words of the instruction. (Hester, 131 Ill. 2d at 100.) Whether a defendant has been accorded his constitutional rights depends on how a reasonable juror could have interpreted the instruction. Hester, 131 Ill. 2d at 100.
In the present case, the jury was instructed that it “may infer” defendant was under the influence if it found that his blood-alcohol concentration was at least .10%. It was further instructed that it was not required to make such an inference and that it should consider all of the evidence in determining whether defendant was under the influence of alcohol. Because the presumption was not binding and a reasonable juror could only interpret the instruction as allowing him or her to either accept or reject the presumption, it was permissive. See Hester, 131 Ill. 2d at 100-01.
Additionally, there was independent evidence of defendant’s being under the influence, such as Officer Beyer’s testimony that he had a strong odor of alcohol on his breath, bloodshot and glassy eyes, and slurred speech. Dr. Gallant also testified that defendant had an odor of alcohol on his breath, and Officer Beyer opined that defendant was under the influence of alcohol. Thus, the presumption was not the sole source of the finding that defendant was under the influence of alcohol. (See Hester, 131 Ill. 2d at 101.) Defendant was not denied due process by the instruction allowing the jury to infer his being under the influence based on a finding that defendant had a blood-alcohol content of .10%. See Hester, 131 Ill. 2d at 102.
If the jurors reached a conclusion that defendant was under the influence of alcohol, which conclusion may or may not have been connected to a finding of blood-alcohol content, no further reason existed for their consideration of blood-alcohol content in proceeding to a determination of recklessness. In light of the instructions, a finding of blood-alcohol content of .10%, if any, was relevant only to the initial question of whether defendant was under the influence of alcohol. Thus, contrary to defendant’s assertions, the jury was not permitted to make any determination of recklessness based upon defendant’s conscious disregard of his blood-alcohol content; rather, the instructions provided only that a finding of recklessness could be predicated upon a preliminary determination that defendant was under the influence of alcohol when he drove his vehicle. Accordingly, we hold the jury was not instructed regarding defendant’s blood-alcohol content so as to have denied defendant his right to be proved guilty beyond a reasonable doubt of acting recklessly.
Defendant’s second appellate contention is that he was denied a fair trial when the prosecutor asked a defense witness whether he had consumed brandy the evening of the accident but never introduced any contrary evidence in response to the witness’ denial. The State essentially concedes that it was improper for the prosecutor to have failed to introduce evidence contrary to the witness’ denial but maintains that such error was harmless.
Notwithstanding the State’s concession of error, we are compelled to emphasize the impropriety of a prosecutor asking a witness questions for purposes of impeachment unless the prosecutor is prepared to offer proof of the impeaching information. (See People v. Olinger (1986), 112 Ill. 2d 324, 341.) A witness’ denial of a prior inconsistent statement makes it incumbent upon the cross-examiner to offer evidence of the prior inconsistent statement. People v. Moore (1973), 54 Ill. 2d 33, 37-38.
Even though the failure to follow through with proof of a prior inconsistent statement is error, such error is not necessarily reversible. (People v. Wilson (1985), 138 Ill. App. 3d 513, 525; People v. Tate (1984), 122 Ill. App. 3d 660, 666.) Where there is other positive evidence of defendant’s guilt, failure to complete the impeachment of a witness with evidence of the denied prior inconsistent statement may be harmless error. Wilson, 138 Ill. App. 3d at 525; Tate, 122 Ill. App. 3d at 666; People v. Reed (1981), 92 Ill. App. 3d 1115, 1119-20.
Under the facts of this case, the prosecutor’s failure to complete his impeachment of the defense witness, Ron Nordin, regarding whether he told an assistant prosecutor that he had consumed brandy on the evening of the accident does not rise to the level of prejudicial error. First of all, there was adequate evidence to support a finding of guilt in this case. There was evidence that defendant was under the influence at the time he drove his automobile, that he drove his automobile after having had only four hours sleep the evening before, and that he crossed a solid yellow line and a raised concrete median to enter into the deceased’s lane of travel. This evidence was sufficient for the jury to have found defendant guilty of reckless homicide, notwithstanding any error in the impeachment of Nordin.
Defendant’s reliance on People v. Rivera (1986), 145 Ill. App. 3d 609, and People v. Orr (1977), 45 Ill. App. 3d 660, is misplaced. In Rivera, the court held that the failure to prove the inconsistent statement referred to in the cross-examination of a key defense witness whose credibility was the foundation of the defense case was reversible error. (Rivera, 145 Ill. App. 3d at 621-22.) Additionally, in Rivera, the prosecutor emphasized the purported impeachment during closing argument. (Rivera, 145 Ill. App. 3d at 621.) In this case, however, Nordin did not constitute a key defense witness. His testimony as to whether defendant was under the influence of alcohol was equivocal at best. For example, he testified that defendant had consumed some amount of beer at his home prior to the accident. Furthermore, while Nordin testified that defendant appeared to have control of his faculties and was not staggering, and that he did not fear defendant was unable to drive, he also testified that he had no opinion as to whether defendant was under the influence of alcohol. Nordin’s testimony, unlike the witness in Rivera, cannot be characterized as the lynchpin of defendant’s case. Additionally, in this case, unlike Rivera, the prosecutor made no mention in argument of the purported impeachment of Nordin or the prior inconsistent statement.
People v. Orr is likewise distinguishable from this case. In Orr, the alleged inconsistent statement was the only potential evidence of motive offered by the prosecution, and thus the court held the failure to produce the statement to be reversible error. (Orr, 45 Ill. App. 3d at 666-67.) As discussed earlier, there was sufficient evidence in this case to support the jury’s verdict.
We turn then to the final contention of defendant that the trial court abused its discretion by considering the death of the victim as an aggravating factor in sentencing defendant. The State responds that the sentence was proper as the trial court placed insignificant weight on the improper aggravating factor. The State further argues that defendant’s sentence was proper in light of his record of prior driving convictions.
All penalties shall be determined both according to the seriousness of the offense and with the objective of restoring the offender to useful citizenship; therefore, a reasoned judgment as to the proper penalty to be imposed must be based upon the particular circumstances of each case. (People v. Saldivar (1986), 113 Ill. 2d 256, 268.) Such a judgment depends on, among other things, the nature and extent of each element of the offense as committed by the defendant. (Saldivar, 113 Ill. 2d at 268-69.) Furthermore, sound public policy demands that a defendant’s sentence vary in accordance with the particular circumstances of the offense committed. It is permissible, therefore, for the trial court to consider the nature and extent of the force employed and the physical manner in which the victim’s death was caused. (Saldivar, 113 Ill. 2d at 271-72.) The commission of any offense, regardless of whether the definition of the offense itself involves harm, can have varying degrees of harm or threatened harm. (Saldivar, 113 Ill. 2d at 269.) Because this varying quantum of harm may constitute an aggravating factor, the severity of a sentence depends upon the degree of harm, even in cases where serious bodily harm is arguably implicit in the offense. Saldivar, 113 Ill. 2d at 269.
Nevertheless, even though a court may properly consider the degree of harm inflicted by a defendant in fashioning a sentence for an offense wherein serious bodily harm is implied, it may not focus on the end result of the harm, the death of the victim, if such death is implicit in the offense. (People v. Martin (1988), 119 Ill. 2d 453, 459-60; Saldivar, 113 Ill. 2d at 271-72; People v. Mott (1990), 204 Ill. App. 3d 573, 577; see also People v. Conover (1981), 84 Ill. 2d 400, 404.) In referring to the statutory aggravating factor of bodily harm, the trial court stated that “there is no question there was great bodily harm caused here, i.e., the death of a young man.” It is evident from the statement that the court focused upon the death of the victim in aggravation. Such a fact cannot be considered as an aggravating factor as death is implicit in the offense of reckless homicide.
The court further commented, however, that “the State advances the position of the sentencing must be a deterent [¡sic].” The court went on to briefly discuss its difficulty in understanding the unexpected death of a young man. This passing reference to the need to deter is difficult to interpret as reflecting the court’s reliance on deterrence as a significant factor in sentencing defendant. Even were we to interpret it as such, it still appears as though the court placed more emphasis on the death of the victim in aggravation. Based on the trial court’s improper consideration of the death of the victim as an aggravating factor, we conclude that a new sentencing hearing is required.
For the foregoing reasons, we affirm the judgment of the circuit court of McHenry County finding defendant guilty of reckless homicide and reverse the sentence and remand for a new sentencing hearing.
Affirmed in part; reversed and remanded in part.
NICKELS and UNVERZAGT, JJ., concur.