OPINION
SCHUMACHER, Judge.
After a hearing, the trial court committed appellant Michael Kenneth Pirkl to the Minnesota Security Hospital as a psychopathic personality. The security hospital filed a report and a review hearing was held. The trial court then committed Pirkl to the security hospital for an indeterminate period as a psychopathic personality. The court also denied a motion for amended findings. We affirm.
[904]*904FACTS
Pirkl’s lengthy history of sexual assaults against women and other crimes began when he was a juvenile, when he was convicted of burglary in 1969, illegal entry of a bus, theft, illegal entry of a building, and attempted aggravated rape in 1971, and theft, aggravated assault, and attempted aggravated rape in 1972.
Pirkl’s adult criminal record began in 1974 with convictions for felonious sale of a controlled substance and felony theft arising from the robbery of a woman. Also in 1974 he was charged with aggravated rape and pleaded guilty to indecent liberties after he sexually assaulted a woman in her home by forcing her to have sexual intercourse by threats of bodily harm.
In 1978, Pirkl was convicted of attempted burglary. During his parole for that offense in 1980, he absconded from a halfway house and was later apprehended and returned to the Minnesota Correctional Facility at Still-water.
In 1980, while on parole, Pirkl entered a woman’s home wearing a nylon stocking over his head and carrying a knife. He found her asleep in her living room and attempted to remove her clothing. After she awoke, he threatened to stab her if she did not cooperate. He forced her to submit to sexual intercourse. Pirkl was charged with criminal sexual conduct in the first degree and pleaded guilty to criminal sexual conduct in the third degree. While serving his sentence for the 1980 conviction, Pirkl absconded, once from a conditional release for sexual offender counseling and once from a halfway house.
On September 7,1985, while still on parole, Pirkl committed another sexual assault. He forced his way into the apartment of a woman with whom he was acquainted after asking to use her telephone. He grabbed the woman by the throat, dragged her to the bedroom, and threw her on the bed. He struck her repeatedly, cutting her face. He told her to scream all she wanted because no one would hear, and told her that if she vomited, he would make her eat the vomit. Pirkl then tore off her robe and penetrated her vagina and anus with his finger. He severely wrenched the woman’s back by twisting her body violently. As he left, he told her he would kill her if she moved.
As a result of the attack, the woman suffered severe back injuries, was unable to walk without pain, and had to take narcotic painkillers. She was so traumatized that she never returned to the apartment, quit her job, and moved out of state. Because of her fear of another attack, she was afraid to be alone at night or with strangers in public.
On October 16, 1985, Pirkl committed another sexual assault. At approximately 4:30 a.m., he entered the apartment of another woman with whom he was acquainted, once again wearing a nylon stocking over his head and carrying a knife. He had earlier learned the woman would be alone in the apartment because her boyfriend would be away that night. When she awoke, she struggled with Pirkl and his knife cut her hand. He forced her upstairs where her children were asleep. The woman screamed in hopes of waking her children. When they awoke, they began to scream as well, and Pirkl fled. As a result of this incident, the woman and her children suffered emotional problems and continued fear.
Pirkl pleaded guilty to first degree criminal sexual conduct for the September 7, 1985 incident and first degree burglary for the October 16, 1985 incident. In addition, during treatment Pirkl admitted to committing 8 to 12 date rapes in which he stalked women, got them drunk, and often used physical aggression to force sexual intercourse. Pirkl also admitted to extensive substance abuse.
Pirkl was admitted to the security hospital in 1986 for a presentence investigation. While Pirkl expressed interest in treatment, because of his legal status the hospital was unable to provide treatment. He was encouraged to obtain treatment in prison. He obtained sexual offender and chemical dependency treatment while incarcerated.
In late 1993 and early 1994, Pirkl was given psychological evaluations pursuant to Department of Corrections risk assessment and release procedures. Two psychologists, Ralph Cornelia and Dr. Richard Friberg, recommended commitment, despite Pirkl’s involvement in treatment programs. These [905]*905recommendations were based on PirH’s recent dishonesty to a parole agent, historical information, and testing results that showed antisocial tendencies and a likelihood of reof-fending. One psychologist, Dr. John Austin, recognized PirH had received significant benefits from treatment. Dr. Austin believed if PirH were released to a halfway house for sexual offenders and avoided chemicals, he could receive the support he needs to avoid further sexual offenses. Dr. Austin did not believe PirH met the criteria for commitment.
A petition for commitment as a psychopathic personality was filed. At the hearing, Dr. James M. Alsdurf, the second court-appointed examiner, recommended that PirH should not be committed as a psychopathic personality because he did not evidence such a lack of power to control his sexual impulses so as to meet the standards necessary for commitment as a psychopathic personality. It was Dr. Alsdurf s opinion that the precipitant to PirH’s assaults was not an impulsive urge or an inability to control his sexual impulses, but that the offenses instead involved a fair amount of planning and deliberateness. Dr. Alsdurf also cited the times when PirH chose not to respond to his sexual impulses and noted the intensity of some of his impulses had changed. He acknowledged, however, that if PirH were intoxicated or using drugs, he would pose a substantial significant risk of sexual reoffending.
Dr. Richard Ascano, the first court-appointed examiner, testified that Pirkl met the standards for commitment as a psychopathic personality and that until PirH’s rejection and abandonment issues are addressed, he will be at significant risk for reoffending. Dr. Ascano testified PirH is capable of controlling his behavior in a courtroom. If there is a situation in which PirH feels rejected or abandoned, however, his anger will be reHn-dled and in his moment of rage, he will be in a state of impulsivity and emotional instability with a profound probability of acting out. He testified the test results and PirH’s recent behavior, including lying to his parole officer and failing a drug test, support this conclusion.
Dr. Ascano disagreed that psychopathic personalities are less likely to act out as they age; instead, he believes that such individuals become wiser in terms of not being caught. Dr. Ascano believed that PirH would benefit from treatment focused on the psychological trauma, which is the dynamic etiology of his offenses, rather than sexual therapy. It was his opinion that the security hospital is the most secure facility that could minimize the risk of escape.
Dr. Friberg described Pirkl as a predatory offender. He selects known victims, and by various ruses is admitted to their residence and then attempts to sexually molest them. PirH fits the criteria for both a power rapist and an anger rapist. He noted that after the end of his two longest periods of treatment, PirH has shown “striMng lapses” in his judgment and ability to control himself. Shortly after completing sex offender and chemical dependency programs at Oak Park Heights, he was reported for drug use. More recently, he lied to his prospective parole officer. Dr. Friberg also found test results and behavioral observations reveal a very severe and clear-cut personality disorder, which is resistant to treatment. Dr. Friberg believed Dr. Alsdurf and others who did not recommend commitment vastly overestimated PirH on the basis of his interview behavior. Instead, the psychological test data and historical data are much more important.
Dr. Friberg supported PirH’s commitment as a psychopathic personality, based on the unusual number of indices showing PirH is at extremely high risk to reoffend, particularly if he were to relapse in his chemical dependency patterns. The security hospital, which has a four-phase highly individualized program for the most difficult and severe sex offenders, is the most appropriate placement.
Dr. Paul Reitman also recommended commitment as a psychopathic personality and found no less restrictive environment than the security hospital. Based on test results, he testified PirH has a great deal of psychopathy, lack of impulse control and remorse, and is unable to delay gratification. He noted Pirkl was an early offender, and is a very impulsive, dangerous person who is chemically dependent. He explained that test results [906]*906showed Pirkl’s treatment programs did not change him behaviorally.
The trial court committed Pirkl to the security hospital after determining he met the criteria for commitment as a psychopathic personality. The court’s findings extensively discussed Pirkl’s history, as well as the experts’ testimony and reports. The court concluded that Pirkl showed an utter lack of power to control his sexual impulses and as a result was likely to reoffend.
A review hearing was held after 60 days. Dr. Reitman, Dr. Alsdurf and Dean DeVries, a psychologist from the security hospital, testified that Pirkl’s history of drug abuse is a big risk factor in terms of reoffending. Mitigating facts included Pirkl’s completion of chemical dependency and sex offender treatment and his current age. DeVries was asked whether Pirkl has changed since his initial commitment, which would lead him to conclude he no longer fits the definition of a psychopathic personality as found by the court initially. DeVries stated Pirkl had not changed since his admission to the security hospital 60 days earlier. He recommended treatment at the security hospital.
Dr. Reitman, addressing lack of control, explained Pirkl can exercise control in a very structured environment. When he perceives a victim, however, all of the biological urges and impulses make him incapable of deterring. His predatory behavior can occur when he has an uncontrollable impulse. The fact he lied to his parole officer shows he is not amenable to change. Dr. Reitman concluded Pirkl has an utter lack of control over his sexual impulses.
Dr. Alsdurf again testified Pirkl did not evidence an utter lack of control as it relates to sexual matters. He cited the very deliberate and planned out quality to most of Pirkl’s sexual offending, such as his date rapes in which he would deliberately get the women drunk to rape them. It was his opinion that if released to society, Pirkl would need to obtain a job, aftercare, and chemical dependency support and, if he remained chemically free and took advantage of various opportunities to reintegrate into society, he would be at low risk to reoffend.
The trial court determined that Pirkl met the factors for commitment as a psychopathic personality, specifically concluding that Pirkl evidenced an utter lack of power to control his sexual impulses and is likely to commit such acts in the future, and found the security hospital was the least restrictive appropriate treatment option for Pirkl. The trial court denied a motion for amended findings of fact.
ISSUES
1. Was the trial court clearly erroneous in determining Pirkl exhibited an utter lack of power to control his sexual impulses and was likely to attack the objects of his uncontrolled and uncontrollable desire?
2. Was the trial court clearly erroneous in determining commitment to the Minnesota Security Hospital is the least restrictive alternative?
ANALYSIS
1. A sexual psychopathic personality1 is defined as
the existence in any person of such conditions of emotional instability, or impulsiveness of behavior, or lack of customary standards of good judgment, or failure to appreciate the consequences of personal acts, or a combination of any of these conditions, which render the person irresponsible for personal conduct with respect to sexual matters, if the person has evidenced, by a habitual course of misconduct in sexual matters, an utter lack of power to control the person’s sexual impulses and, as a result, is dangerous to other persons.
Minn.Stat. § 253B.02, subd. 18a (1994); see State ex rel. Pearson v. Probate Court, 205 Minn. 545, 555, 287 N.W. 297, 302 (1939), aff'd, 309 U.S. 270, 60 S.Ct. 523, 84 L.Ed. 744 (1940). “The psychopathic personality statute identifies a volitional dysfunction which [907]*907grossly impairs judgment and behavior with respect to the sex drive.” In re Blodgett, 510 N.W.2d 910, 915 (Minn.), cert. denied, — U.S. -, 115 S.Ct. 146, 130 L.Ed.2d 86 (1994).
The trial court must find the person is a psychopathic personality by clear and convincing evidence. Minn.Stat. §§ 253B.185, subd. 1, 253B.18, subd. 1 (1994); In re Linehan, 518 N.W.2d 609, 613 (Minn. 1994). When there is conflicting evidence as to the existence of a psychopathic personality, the question is one of fact to be determined by the trial court upon all the evidence. In re Martenies, 350 N.W.2d 470, 472 (Minn.App.1984), pet. for rev. denied (Minn. Sept. 12, 1984). The reviewing court will not defer to the trial court if it has erred as a matter of law. In re Stilinovich, 479 N.W.2d 731, 734 (Minn.App.1992).
A. Utter Lack of Control
Pirkl argues that the trial court findings in the initial commitment as to utter lack of control were conclusory and unsupported by the evidence. He contends that the trial court relied upon only the existence of a habitual course of misconduct in sexual matters to support the commitment.
The supreme court has set out factors that are relevant to the predatory sex impulse and the lack of power to control it. Blodgett, 510 N.W.2d at 915. If a person has the ability to control the sexual impulse, the standard for commitment is not met. See Linehan, 518 N.W.2d at 613; In re Bieganowski, 520 N.W.2d 525, 529-30 (Minn.App. 1994) (lack of control shown), pet. for rev. denied (Minn. Oct. 27, 1994); In re Schweninger, 520 N.W.2d 446, 450 (Minn.App.1994) (lack of control not shown), pet. for rev. denied (Minn. Oct. 27, 1994). Refusal of treatment and lack of a relapse prevention plan can show an utter lack of control. See Bieganowski, 520 N.W.2d at 530 (sexual predator’s failure to remove himself from situations that provide opportunity for similar offenses and failure to avoid precursors that trigger impulsive behavior, such as consumption of alcohol, demonstrate lack of control; court also cited expert testimony and lack of treatment); cf. Linehan, 518 N.W.2d at 613 (no showing of utter lack of control where treated sex offender apparently had alcohol abuse under control and expert testimony inconsistent on lack of control).
The trial court found Pirkl evidenced an utter lack of power to control his sexual impulses. The court discussed in detail the six factors identified by the supreme court as relevant to a determination of whether the person exhibited a lack of control. See Blodgett, 510 N.W.2d at 915.
(1) Nature and Frequency of Sexual Assaults. The trial court cited the fact that since age 16, Pirkl has been convicted of six sex-related offenses, and has self-reported approximately 8 to 12 other offenses, mostly “date rapes,” that have never been prosecuted. The two assaults in 1985 occurred after Pirkl absconded from the sex offender treatment in a halfway house.
(2) Degree of Violence Involved. The trial court found Pirkl demonstrated a propensity to use weapons, threats of physical violence and physical intimidation, as well as actual physical force. In one case, the victim sustained a stab wound. In another, Pirkl beat the victim, cut her face, and wrenched her back. The report of Dr. Alsdurf indicated that according to Pirkl, at least half of his date rape victims were subjected to physical aggression, and Pirkl was “very forceful with” at least two of them.
(3) Relationship Between Offender and Victims. The trial court found it appeared Pirkl was marginally acquainted with the victims and his methods indicated a stalking pattern. He chose his victim in advance and calculated a method or ruse by which to gain access to her. He pretended to need to use one victim’s telephone, and he attacked another when he knew her boyfriend was out of town. Similarly, the date rapes suggest that Pirkl gains the victim’s confidence and attacks when she is vulnerable. According to Dr. Alsdurfs report, Pirkl would get his victims drunk to inhibit their resistance, and on at least three occasions his victims were passed out when he raped them.
(4) Offender’s Attitude and Mood. The trial court found the assaults by Pirkl were based on rage rather than sexual gratifica[908]*908tion, fostered by a feeling of rejection and inadequacy, and fueled by a significant abuse of chemicals and alcohol. Dr. Friberg testified Pirkl meets the profile of a power rapist and an anger rapist. Dr. Alsdurf reported Pirkl’s stalking pattern is “associated with an internal experience of power and a release of rage through sexual domination.” More than one evaluation suggested Pirkl rationalized his assaults by blaming the victim.
(5) Offender’s Medical and Family History. The trial court noted Pirkl comes from a dysfunctional family, exhibiting significant levels of physical and emotional abuse. From a medical standpoint, Pirkl demonstrated a long history of alcohol and drug abuse and a history of failed treatments. Virtually all sexual assaults occurred while Pirkl was under the influence of chemicals. While Pirkl has received chemical dependency treatment while in prison, his prognosis was deemed fair and the last documented use of chemicals in prison was in mid-1990. Dr. Alsdurf testified that the McAndrews Addiction Scale score suggested Pirkl is a clear risk for further abuse of drugs and alcohol. Every therapist who examined Pirkl testified that if he uses chemicals again, chances of committing additional sexual assaults are significantly increased. Dr. Alsdurf admitted that the McAndrews score demonstrates impulsive behavior, which Pirkl lacks the ability to control.
(6) Results of Testing and Evaluation. The court noted that all five psychologists who examined Pirkl or his records agreed that Pirkl is emotionally unstable, exhibits impulsive behavior, and lacks customary standards of good judgment, so as to render him irresponsible for his conduct with regard to sexual matters and dangerous to others. The majority believed Pirkl met the definition of a psychopathic personality. All appeared to agree Pirkl has an antisocial personality disorder and a deep sense of inadequacy, a primary precipitating factor in his violent behavior.
Unlike Linehan, 518 N.W.2d at 613, the trial court here did not use the Pearson language in a conclusory manner. Instead, the court extensively discussed the very factors that the supreme court held should be considered in Blodgett.
The court also explained its reasons for rejecting the opinions by the two experts who did not believe that Pirkl met the psychopathic personality standards. The court explained that these experts based part of their opinions on statements by Pirkl, particularly citing one expert who described Pirkl as sincere, honest and dedicated. The court, however, cited Pirkl’s dishonesty towards corrections personnel.
The court also discredited these experts’ reliance upon Pirkl’s statements to them that indicated Pirkl had benefitted from treatment. The court found Pirkl made virtually identical statements in prior treatment programs before he sexually assaulted two women. The court also found Pirkl’s self-serving statements suggesting empathy were contrary to test results showing he had little ability to develop genuine guilt.
The court’s evaluation of expert testimony is of particular significance and is not clearly erroneous. See In re Joelson, 385 N.W.2d 810, 811 (Minn.1986). The court’s conclusion that Pirkl exhibited an utter lack of control is not clearly erroneous.
Pirkl also challenges the trial court’s decision after the final hearing. Pirkl contends that the trial court’s findings regarding DeVries’ opinion on utter lack of control were clearly erroneous.2 He contends that contrary to the court’s findings, DeVries did not testify Pirkl met the standards for commitment and had an utter lack of power to control his sexual impulses.
The trial court, in its order denying Pirkl’s motion to amend its findings, explicitly found DeVries’ testimony supported the determination that Pirkl met the Pearson standard. The court found that it appeared DeVries tried to change his testimony, which the court declined to allow. The court further [909]*909found any error it may have made in interpreting DeVries’ testimony was harmless and determined more than sufficient evidence existed apart from DeVries’ testimony to support the conclusion Pirkl met the Pearson standard, including the testimony of Dr. Reitman, the previous testimony of other experts, and the substantial findings made at the initial hearing. In fact, the trial court adopted the findings from the initial hearing, as well as making additional findings.
The trial court carefully analyzed and weighed the testimony, in light of controlling case law, to decide that Pirkl exhibited an utter lack of control. The trial court’s decision is supported by clear and convincing evidence.
B. Future Dangerous Behavior
The next issue raised is whether there was clear and convincing evidence Pirkl is likely to engage in future dangerous behavior. See Linekan, 518 N.W.2d at 614. This case presents a similar situation to Linehan because there is a gap of time between the petition for commitment in 1994 and Pirkl’s last act of sexual misconduct in 1985. See id. After finding utter uncontroll-ability of sexual impulses, the trial court, in predicting serious danger to the public, should consider certain factors if such evidence is presented:
(a) the person’s relevant demographic characteristics (e.g., age, education, etc.);
(b) the person’s history of violent behavior (paying particular attention to recency, severity, and frequency of violent acts); (c) the base rate statistics for violent behavior among individuals of this person’s background (e.g., data showing the rate at which rapists recidivate, the correlation between age and criminal sexual activity, etc.); (d) the sources of stress in the environment (cognitive and affective factors which indicate that the person may be predisposed to cope with stress in a violent or nonviolent manner); (e) the similarity of the present or future context to those contexts in which the person has used violence in the past; and (f) the person’s record with respect to sex therapy programs.
Id.
The trial court in the review hearing addressed all these factors. The trial court found Pirkl’s age is a mitigating factor because he was 39 years old, and the incidence of recidivism among rapists declines after the age of 40.
The court found Pirkl habitually used force, threats, violence and weapons during his sexual assaults, and inflicted severe physical and emotional injuries upon his victims. According to the review report filed by the security hospital, if Pirkl reoffends, his history would suggest a sexual assault against an adult woman, in which it would be likely he would use violent aggression and inflict physical injuries if necessary to successfully commit the offense. The court rejected Pirkl’s arguments as to the lack of recency of his sexual assaults, noting that he has been in a prison the last nine years and has not had the opportunity to commit sexual assaults. See In re Bobo, 376 N.W.2d 429, 432 (Minn. App.1985) (good behavior in artificial environment not determinative of dangerousness, where experts testify patient remains mentally ill and dangerous). While Pirkl argues that his record does not show gratuitous infliction of physical injury or sadistic behavior, the facts show and the trial court found otherwise.
The trial court did not consider any base rate statistics for violent behavior among individuals of Pirkl’s background. Pirkl does not challenge this.
The trial court found Pirkl’s environment outside the prison would be highly stressful, making it likely Pirkl would return to drug and alcohol abuse, increasing the chances he would commit another sexual assault. Further, the court cited Pirkl’s score on the MeAndrews Addiction Scale, indicating he is at high risk of returning to chemical abuse. The court found it likely Pirkl would return to his pattern of releasing hostility through sexual domination if returned to the outside world.
A mitigating factor that the court addressed was Pirkl’s successful completion of [910]*910chemical and sex offender treatment programs and his behavior as a model prisoner over the past nine years. See Linehan, 518 N.W.2d at 613 (after completing chemical dependency treatment, appellant’s alcohol abuse, which was precursor of his assaults, was under control, supporting finding he does not exhibit utter lack of power to control sexual impulses).
The trial court, however, concluded that Pirkl has not been changed by his treatment programs and remains likely to reoffend if given the opportunity. It noted that the experts agreed that further treatment would not likely reduce the risk he would reoffend in society, and two of these experts testified he should not be released. DeVries testified that the skills Pirkl has learned for coping would be severely impaired if he returned to abusing drugs and alcohol, which the court concluded was highly likely. Testimony from the initial hearing indicated Pirkl’s sincerity and successfulness in treatment cannot be trusted because he has shown signs of faking his recovery.
Pirkl cites the varying opinions by experts. He contends a careful review of the psychological testimony and reports show there is not clear and convincing evidence he will reoffend. He also relies upon the Linehan factors, citing the length of time since his last offense, successful treatment, ability to interact, his intelligence, and age. It was for the trial court, however, to weigh these factors. The trial court’s findings are supported by the record and are not clearly erroneous. The court’s reasoning and application of the facts to the Linehan factors show there is clear and convincing evidence Pirkl is likely to engage in future dangerous behavior.
2. Pirkl claims that the security hospital is not the least restrictive environment. If a court finds after a review hearing that a patient continues to be a psychopathic personality, the court shall order commitment for an indeterminate period. See Minn.Stat. § 253B.18, subd. 3 (1994). The court must find that there is no appropriate less restrictive alternative available. Minn.R.Civ.Commitment 12.06; see Bieganowski, 520 N.W.2d at 531. The trial court determined that pursuant to expert testimony, the least restrictive appropriate treatment plan for Pirkl would be the Minnesota Security Hospital. The court noted that Pirkl has fled three times in the past when given less restrictive options. Noting that he is likely to abscond and commit future sexual assaults, the court concluded the only appropriate treatment was that which restricts his liberty and separates him from the public.
Pirkl challenges this determination based upon the testimony by the experts that treatment at the security hospital is not likely to significantly reduce his risk of reoffending. He cites testimony by Drs. Alsurf and Asca-no that Pirkl needs treatment focused on his psychological trauma and anger management, and did not specifically recommend sex offender treatment. He argues he can obtain the psychological treatment in the structured parole setting. He contends that the placement is simply for the purpose of warehousing him.
Respondent argues that there is no suggestion of a facility less restrictive than the security hospital, noting the treatment Pirkl has received so far has not worked because his last two offenses occurred while on parole. One of the objectives of commitment as a psychopathic personality is to offer treatment. See Enebak v. Noot, 353 N.W.2d 544, 548-49 (Minn.1984). While treatment of persons committed as psychopathic personalities may not always be successful, the state has the power to keep trying to treat such persons. Blodgett, 510 N.W.2d at 916. The state’s interest in safety is also legitimate. Id.
The trial court found Pirkl is presently a danger to women in the community and continues to need treatment. If not confined, Pirkl would be without supervision so that he could resume his abuse of alcohol and drugs. The trial court had clear and convincing evidence to conclude that the least restrictive appropriate alternative placement for Pirkl is at the security hospital.
DECISION
The trial court’s decision to commit Pirkl to the Minnesota Security Hospital for an indeterminate period as a psychopathic per[911]*911sonality was supported by clear and convincing evidence.
Affirmed.