Violence in Older Persons: Part I—Occurrence in Forensic/Criminal Situations, Partner Relationships, and Sexual Offenses
This is Part I of a two-part article. Part II will examine violence in hospitals and available treatment options.
When people discuss issues of violence and the elderly, they usually think of acts perpetrated against senior citizens, not those committed by them. While it is true that the elderly as a group are responsible for the lowest rate of violent crimes reported to law enforcement, they are still capable of committing acts of considerable violence.1-4 For example, elderly males are responsible for three murders per 100,000 and elderly females are responsible for 1.5 murders per 100,000, as compared to the most violent group, males under the age of 25, who committed 10.3 murders per 100,000 in 2003.1 The national crime database does not capture all acts of aggression committed by the elderly. Often, when the elderly commit acts of violence either at home or in an institutional setting such as a hospital or nursing home, they are suffering from a medical condition such as dementia, delirium, or psychosis, and the act is not reported as a crime or, if reported, is not prosecuted.5
This two-part review on violence committed by the elderly examines various situations in which elderly individuals become violent, their motivation for the aggression, and exacerbating medical circumstances (if present). Part I discusses the forensic aspects of violence in the elderly, incidents that lead to arrest, court-ordered psychiatric evaluations, involuntary hospitalizations, partner abuse, and sexual misconduct. Part II (to be published in the next issue of Clinical Geriatrics) will address violence in hospitals and available treatment options.
Older Individuals and the Legal System
There have been few prospective studies looking at police involvement and general violence with the elderly.5 What is available is nationwide crime data, which show that there were 63,894 arrests made of persons age 65 years or older in 2006, of which 2966 were related to violent crimes2 (Table I). General predictive factors for elderly persons who committed violent crimes include male gender, being a member of a minority, being of low socioeconomic status, alcohol use at the time of the offense, suffering from psychotic symptoms such as paranoia, and a history of committing violent offenses when younger.5,6 Many of these factors commonly occur in younger offenders as well, and may be more of a marker for general lawlessness or individuals prone to having the police called about them, rather than something unique to any age group. These factors do not predict violent acts that occur following the onset of a medical condition in the elderly, such as a frontotemporal dementia, a disorder that impairs impulse control. This suggests that when doctors and the legal system look at violence committed by the elderly, two general categories need to be identified: (1) general lawlessness, which can occur at any age, and (2) violence secondary to a medical condition such as dementia.
Based on studies of elderly patients referred for forensic psychiatric evaluations or services, it has been noted that at least half of the arrested elderly have some diagnosable psychiatric disorder such as dementia, alcohol abuse or dependence, psychotic disorder, depression, and/or personality disorder.6-9 Of those sent for forensic evaluations, 30-50% were found to be incompetent to stand trial or legally insane.6,8,9 The medical conditions most commonly found in elderly forensic evaluees include hypertension, a history of head injury with loss of consciousness, heart disease, stroke, liver disease, and diabetes.5,6 Most of these conditions have the potential to exacerbate comorbid psychiatric conditions such as dementia in the elderly.
In a study by Lewis et al6 looking at individuals in South Carolina over the age of 60 who were referred for forensic psychiatric evaluations before trial between 1991 and 1998, it was noted that 67.7% had a diagnosis of alcohol dependence, 44.4% had dementia, 32.3% had antisocial personality disorder, and 25.2% had a psychotic disorder. Approximately 60% of the sample were facing charges for violent crime and more than 80% of those individuals had previously been charged with a crime. In a majority of the incidents in which the elderly were being charged with a violent crime, a weapon such as a gun (78.3%) or knife (3.3%) had been used.6 The high percentage of weapon use by the elderly was attributed to their perceived vulnerability, the belief that they needed a weapon to protect themselves, poor overall physical health, and limited strength.6,7 In a majority of the violent incidents that occurred, the elderly individual acted alone.6 The majority of the victims were known by the elderly perpetrator and were either acquaintances (particularly neighbors [44.2%]) or family members (particularly spouses [40.7%]).6 Violent incidents were noted to occur most often in the perpetrator’s home.6 Approximately one-third of the offenders were found not competent to stand trial. Those who were found to be incompetent were most frequently diagnosed with dementia.6 An unexpected finding in Lewis et al’s study was that being actively engaged in outpatient psychiatric treatment was a protective factor, which decreased the risk of an elderly individual being involved in a violent crime.6
In a study of pre-trial competency evaluations of geriatric defendants charged with felonies, Frierson and colleagues8 found that a large percentage of geriatric defendants, who were judged to be incompetent to stand trial, suffered from alcohol-related dementia. They also found significant deficits in both orientation and memory in those deemed incompetent to stand trial. Other associated factors included difficulty with abstraction, diminished concentration, inability to do calculations, and disturbed thought processes. These findings occurred most frequently in individuals diagnosed with delirium, depression, or dementia.8
A study of geriatric offenders seen post-trial at their pre-sentencing evaluations in Sweden compared the rates of psychiatric morbidity in older evaluees (60 yr and older; median age, 65.8 yr; n = 210) to younger evaluees (15-59 yr; median age, 34 yr; n = 7087) between the time period of 1988-20009 (Table II). The study found a similar gender distribution between the two groups, with less than 10% of the evaluees being female. The psychiatric diagnoses that were significantly more common in the elderly group compared to the younger group were dementia, psychosis related to affective disorders (but not schizophrenia), and symptoms related to cerebral lesions. The younger cohort was noted to have a statistically significant higher percentage of personality disorders (33.2%) than the elderly population (19.5%). There was a statistically significant difference between the age of the offender and the likelihood of being found “legally insane,” with 62.6% of the geriatric group being found legally insane versus 50.5% of the younger group. Of the factors studied, having dementia, being charged with a sexual offense, being of non-immigrant status, being charged with homicide, absence of a personality disorder, and absence of schizophrenia were the respective factors more commonly seen in the forensic evaluations of the elderly proband.9 The difference in personality disorder rates between this study and that of Lewis et al’s study may be due to methodological differences, cultural differences found in Sweden versus the United States, and litigant population sampling differences.6,9
Geriatric Prison Population
A British study by Fazel and colleagues10 of convicted geriatric prisoners (age 60 yr or older; mean age, 65.5 yr) found they had a similar prevalence of total mental illness and personality disorders as arrestees evaluated pre-trial, but found differing percentages of specific psychiatric diseases in the convicted population. In Fazel’s study, 72.9% of the sample (n = 203) had been in prison for less than four years (median, 16 mo), suggesting relatively recent crimes for the age group. Half of the sample was imprisoned for sexual crimes, 24.6% for violent crimes, and 14.3% for drug charges. Half of the study population had a psychiatric history prior to arrest, with 31.5% (n = 64) screening positive for an active Axis I psychiatric disorder and 30% (n = 60) screening positive for a personality disorder at the time of interview. Of the psychiatric problems identified, personality disorder (30%; n = 61) was the most common, followed by depression (29.5%; n = 60) and active drug use or dependence (4.9%; n = 10). Only 0.9% (n = 2) of the convicted prison population had a diagnosis of dementia. In these cases, the dementia was identified post-conviction. Although England does have different laws regarding competency to stand trial than the United States, they are similar in founding principles. These data therefore suggest that most individuals who commit dementia-driven acts are not placed in jail but are referred to psychiatric hospitals or nursing homes.10
A study of patients/prisoners over the age of 65 (72% convicted of a crime) at the Three Bridges medium-security forensic unit in London, a facility that provides care for patients with mental disorders who are severely disruptive and/or dangerous, found that only one-third of the geriatric cases referred to them committed the crime for which they were incarcerated after the age of 65.11 The average age for the commission of the index crime after the age of 65 was 67.4 years, with the average age for first psychiatric referral occurring at 68.3 years. Nearly 43% of this group were referred for sexual behaviors, 35.7% for violent behaviors, and 14.3% for destruction of property; 35.7% were referred from jail, 35.7% from the community, and 14.2% from other psychiatric hospitals.11
It was found that 78.6% of the individuals who committed a crime after the age of 65 met criteria for a psychotic disorder, affective disorder, or organic mental disorder (such as dementia, brain damage, and mental retardation).11 Similar findings occurred in 53.6% of the patients who had committed a crime before age 65. All of the victims of homicides committed by men over the age of 65 were female family members, either spouses or grandchildren, which differed from the pre-65 offenders’ victims, who included people outside of the family and individuals of both genders. The victims of sexual crimes committed by offenders over age 65 were usually solitary (ie, just one victim) and female, most often a female grandchild or a female family acquaintance age 6-16 years. Pre-65-year-old offenders, on the other hand, had a considerably wider range of victims, a greater number of victims, and were more likely to have offended against both males and females. In short, the pre-65-year-old offenders had a sexual abuse pattern similar to that seen in younger sexual offenders (ages 20-50 yr).11,12 It is likely that the over-65 age group who committed acts against family members and acquaintances for the first time were beginning to suffer from some form of medical condition that affected their judgment, such as Pick’s disease or another disease process affecting their frontal lobes or causing a loss of social and executive function.
Although Lewis’, Frierson’s, and Fazel’s studies provide interesting data about older persons in the legal system, one needs to remember that the studies do not capture all, or even most, events of geriatric violence, nor do they reflect all geriatric individuals who are arrested for the commission of a violent crime.6,8-10 These studies only look at individuals referred for competency evaluations or who are convicted, and do not include geriatric patients taken to hospitals for evaluation.6,7 An Israeli study looking at involuntary psychiatric hospital admissions in individuals who had not been previously hospitalized noted that the age groups most likely to be involuntarily committed for the first time were those ages 18-24 and those 65 years and older.13 Although no specific comments were made concerning the diagnosis of the older patients, it would be unusual for this population to be having a first-break episode of schizophrenia, which was the most common diagnosis in the younger age group. It is likely that these elderly individuals were hospitalized for behavior related to cognitive decline or the psychosis that is typically seen in delirium, depression, or dementia.
Although many of the elderly persons who commit crimes have the potential of suffering from a medical or psychiatric disorder that can affect their judgment or cause impulsivity, they are also capable of committing violence for more typical reasons (eg, money, anger, jealousy, revenge). There are also some distinctive motives for older persons, such as murder/suicide related to failing health/terminal medical diagnoses, where elderly individuals kill their spouses as well as themselves after learning that one or the other of them is dying.5 There have also been reports of violence committed by the elderly related to non-terminal general medical problems. For example, there were two cases of individuals who killed their spouses after being prescribed sildenafil.14 In one case, the individual was upset when the sildenafil did not help correct his diabetes-induced erectile dysfunction. He was fearful that his wife was going to leave him because of his impotence, so he killed her. In the other case, the sildenafil was successful in reversing the erectile dysfunction, but the man killed his wife after she rejected his advances. In both cases, the perpetrators were beginning to show signs of dementia but were still highly functional and had long-strained marriages, which became socially unbalanced with the addition of the medication.14 Paranoia was a factor in both homicides.
A sometimes-overlooked problem is violent partner abuse in elderly couples. This can be a situation where the violence has been occurring for decades or could be a situation where a once-stable relationship becomes physically abusive due to medical changes or life changes in either partner. As previously mentioned, 40% of those found incompetent after the commission of a violent crime had attacked their spouse.6 Data from the 1993 to 2001 National Crime Victimization Survey estimated that 4.4 per 10,000 women over the age of 55 were the victims of non-lethal partner violence annually, with approximately 14,700 incidents reported yearly.15 These figures most likely represent an underestimation of the frequency of violent partner attacks due to underreporting.6 Multiple medical studies have reported higher rates of 1.5% to 5% of women over the age of 55 experiencing partner abuse annually.11,13-18 A prospective study by Mouton et al18 looking at postmenopausal women (age 50-79 yr) found that the incidence of new partner violence over a three-year time period was 3.7 per thousand.
A study of 38 women (age 55-90 yr) who were identified as victims of partner abuse noted that 50% had discussed their physical abuse with their primary care doctor.19 Many of the elderly victims were reluctant to identify the abuse to their doctor and only revealed it after being injured.19 The reasons seniors gave for staying in abusive relationships were similar to those reported by younger women (eg, financial concerns, fear of physical reprisal for leaving, concerns about being on their own and their ability to be financially self-sufficient), but also included generational beliefs about the sanctity of marriage and the lack of recognition that the abuse was inappropriate (“men are allowed to discipline their wives”).15,17,19 It was also noted that one-fifth of the women in the study had the same primary care provider as their abusive spouse, which made it difficult for some of these women to discuss the issue of violence with their doctors.19 A study by Zink and colleagues16 noted that older female victims of partner violence were diagnosed by their primary care doctors with higher rates of depression (P = 0.005) and chronic pain (P = 0.039) than the non-abused elderly.
Unfortunately, most studies of older people focused solely on the abused partner and provided little or no information about the abuser, such as the abuser’s medical condition or whether the medical condition contributed to the abuse.17 Many of the new-onset cases of abuse in the elderly relate in some way either to the abuser’s declining health or the declining health of the abused.17,18 New-onset combative and aggressive behavior is one of the most common causes for an elderly individual to be placed in a nursing home or to be started on a neuroleptic medication.20,21
Sexual Aggression and/or Crimes
As noted in the prison population studies and the Swedish pre-sentencing evaluation study, elderly individuals are often arrested for sexual crimes.9-11 These crimes can range from committing physically violent and forceful acts, to crimes that occur with a willing but manipulated victim, to “victimless crimes” such as engaging in or soliciting prostitution, exposing oneself in public, or urinating in public.12 When looking at elderly individuals who are arrested for sexual crimes, it is important to examine the nature of the offense, whether there is an ongoing pattern of behavior, or whether the behavior is new or unexpected. Issues involving paraphilic behavior rarely develop spontaneously in elderly individuals unless there is another comorbid medical problem such as dementia, brain tumor, or the activation of a delirium from medications or recreational substances; however, it is possible for an individual to have a life-long paraphilic interest that has not been discovered by others until the individual becomes impaired and “makes a mistake.”
Thornton,22 in a 10-year study investigating the effects of age on sexual recidivism, found that the odds of re-offending decreased by 0.02 per year after release from incarceration. The over-60 age group had the sharpest decline in rate of re-offense, with an annual rate of 4.3%. These findings are similar to other studies on sexual recidivism, which show the general risk for “older age groups” (exact age ranges vary between studies) to sexually re-offend being between 3.3% and 6.1%.23,24 It is important to realize that selection criteria for study group offenders often differ, that not all studies show a linear decrease of risk with age, and that risk of recidivism may be strongly influenced by paraphilic tendencies that may not be identified or separated out in all studies.12,24
Thornton noted that individuals who offended while younger had more general criminal characteristics, while individuals who kept offending into old age were more typically “sexual specialists”(ie, had a paraphilia).22 It needs to be remembered that this study looked at people released from prison and did not include individuals suffering from known dementia or classic frontal lobe disinhibition.
Thornton felt that sexual assaults by the over-60 age group may be underreported because families were unwilling to report a “grandfather close to death” and because sexual assaults and touching that occurred in nursing homes and hospitals were often excused as being the result of a medical disorder or condition.22 In general, caution needs to be taken when interpreting the data from recidivism studies because it is difficult to ensure that all or even most acts of recidivism are captured (self-report vs reported criminal record) and the definition of recidivism can differ from study to study (eg, any arrest vs an arrest for just a sexual incident).12,24,25
Dickey et al,26 in a study looking at the relationship between age and types of sexual crimes committed, noted that the frequency of sexual crimes decreased with age for the three types of offenders studied: rapists, sexual sadists, and pedophiles, who acted against individuals they were not related to. Rapists showed the most notable decline in re-offense after age 40, the diminished rate of assaults being attributed to declining physical strength and sexual drive. What was interesting to note was that although the frequency of pedophilic acts also decreased with age, it did so at a slower rate than the other two categories of sexual crime. Forty-four percent of the pedophiles who re-offended in this sample were in the “older adult age range” (age 40-70 yr).26 When compared to rapists and sexual sadists, pedophiles comprised 60% of all older offenders, indicating that pedophiles continue to offend into their later years at a greater rate than other sexual offenders.26
In a study that further examined the data from the aforementioned British prison study10 involving prisoners 60 years and older, the rate of psychiatric disorder was found to be approximately equivalent for sex offenders (n = 101) and other criminals (n = 102).27 The age at which the individual was convicted for the current containment offense for the sexual offenders was 62.7 years with a standard deviation of 7.3, which again indicated that most of these crimes occurred when the individual was older. What was noted as a difference between the two groups was that elderly sexual offenders more often had schizoid, obsessive-compulsive, or avoidant personality traits and had fewer antisocial traits than the elderly non-sexual offenders.28 The higher rate of schizoid, obsessive-compulsive, and avoidant traits has been replicated in population studies of younger sexual offenders and is not likely a characteristic common to just elderly offenders.12 The individuals found guilty of more violent acts such as rape and “buggery” (ie, sodomy) were more likely to have been diagnosed with a personality disorder than the less aggressive sexual offenders27 (Table III). In a later study of a subgroup of this population, no frontal lobe differences were noted between the sexual offenders and the regular prison population.28 It needs to be remembered that the population studied were convicted of crimes, and did not take into account individuals who may have committed a sexual crime but who were found not criminally responsible due to a medical condition. The aforementioned Three Bridges study11 may be a better model of elderly individuals suffering from a mental illness who committed a sexual crime.
Sexual Crimes Committed by Clergy
Due to recent scandals, there is great concern about how frequently the clergy engage in sexual crimes. Unfortunately, there are few scientific data about this group due to the reluctance of active or former members of the clergy to participate in research studies.25,29 Studies that do exist have shown that clerics who engage in sexual crimes are similar to other sexual offenders except that they tend to be better educated, are less often married, have a lower degree of antisocial traits, and are older.29 Their sexual crimes are driven more by their long-term sexual deviances/paraphilias than some other sexual offenders whose crimes are more opportunistic and violently criminal in nature (eg, opportunistic rape but no underlying paraphilias). Clergy who engage in sexual crimes are often patient and lure their victims into sexual acts. They often try to maintain long-term sexual relations with their victims, and in so doing cause great psychological harm. It is not clear if the older age is an indication that they truly offend at older ages or that their actions are just being discovered when they are older, due to earlier cover-ups by them and their religious institutions.12,29 In a small study by Langevin and colleagues,29 the clergy were noted to have used force with victims at a greater frequency than the control sexual offender group.
Other studies suggest that clergy who engage in sexual crimes often establish relations with young children (ages 6-12 yr), build trust, seduce with special privilege or praise, and then threaten retribution if the acts are disclosed.12,29 Fear then binds the child to victim status until released or until the clergy “moves on.”
When the elderly commit violent acts, one must determine whether these actions are due to general lawlessness or are the result of a medical condition. Predictive factors for violent lawlessness are the same for the elderly as for other age groups and often involve alcohol or drug use/abuse, sociopathy, and the use of weapons. In criminal cases involving violent acts, the most common victims are neighbors and family members. It is sometimes helpful to examine an elderly individual’s current life situation to understand their motivation for violent acts, especially in cases involving terminal illness or changes in long-standing relationships, such as with spouses. In cases where violence or inappropriate sexual behavior has occurred that is out of character for the elderly individual, medical causes such as delirium, dementia, or organic brain dysfunction need to be ruled out.
The authors report no relevant financial relationships.
Dr. Ryan Hall is an Affiliate Instructor at the Department of Psychiatry and Behavioral Medicine, University of South Florida, Tampa, FL, and is a 2006 Rappeport Fellow of the American Academy of Psychiatry and the Law; Dr. Richard Hall is Courtesy Clinical Professor of Psychiatry, University of Florida, Lake Mary, and Affiliate Professor, Department of Psychiatry and Behavioral Medicine, University of South Florida, Tampa, FL; and Ms. Chapman is Research Assistant to Dr. Richard Hall.
1. Centers for Disease Control and Prevention (CDC). Homicide and suicide rates—National violent death reporting system, six states, 2003. MMWR Morb Mortal Wkly Rep 2005;54:377-380.
2. U.S. Department of Justice, Federal Bureau of Investigation. Crime in the United States 2006. Table 38. Available at: www.fbi.gov/ucr/cius2006/data/table_38.html. Accessed April 8, 2008.
3. U.S. Department of Justice, Federal Bureau of Investigation. Crime in the United States 2005. Table 38. Available at: www.fbi.gov/ucr/05cius/data/table_38.html. Accessed February 26, 2008.
4. U.S. Department of Justice, Federal Bureau of Investigation. Crime in the United States 2005. Table 46. Available at: www.fbi.gov/ucr/05cius/data/table_46.html. Accessed February 26, 2008.
5. Tueth MJ, Zuberi P. Life-threatening psychiatric emergencies in the elderly: Overview. J Geriatr Psychiatry Neurol 1999;12:60-66.
6. Lewis CF, Fields C, Rainey E. A study of geriatric forensic evaluees: Who are the violent elderly? J Am Acad Psychiatry Law 2006;34:324-332.
7. Yorston GA, Taylor PJ. Commentary: Older offenders—No place to go? J Am Acad Psychiatry Law 2006;34:333-337.
8. Frierson RL, Shea SJ, Shea ME. Competence-to-stand-trial evaluations of geriatric defendants. J Am Acad Psychiatry Law 2002;30:252-256.
9. Fazel S, Grann M. Older criminals: A descriptive study of psychiatrically examined offenders in Sweden. Int J Geriatr Psychiatry 2002;17(10):907-913.
10. Fazel S, Hope T, O’Donnell I, Jacoby R. Hidden psychiatric morbidity in elderly prisoners. Br J Psychiatry 2001;179:535-539.
11. Tomar R, Treasaden IH, Shah AK. Is there a case for a specialist forensic psychiatry service for the elderly? Int J Geriatr Psychiatry 2005;20(1):51-56.
12. Hall RC, Hall RC. A profile of pedophilia: Definition, characteristics of offenders, recidivism, treatment outcomes, and forensic issues. Mayo Clin Proc 2007:82(4):457-471. [Erratum in Mayo Clin Proc 2007;82(5):639.]
13. Bauer A, Rosca P, Grinshpoon A, et al. Trends in involuntary psychiatric hospitalization in Israel 1991-2000. Int J Law Psychiatry 2007;30:60-70. Epub 2006 Dec 4.
14. Mintz D. Unusual case report: Nonpharmacologic effects of sildenafil. Psychiatr Serv 2000;51:674-675.
15. Rennison C, Rand MR. Nonlethal intimate partner violence against women. A comparison of three age cohorts. Violence Against Women 2003;9(12):1417-1428.
16. Zink T, Fisher BS, Regan S, Pabst S. The prevalence and incidence of intimate partner violence in older women in primary care practices. J Gen Intern Med 2005;20:884-888.
17. Zink T, Jacobson CJ, Regan S, et al. Older women’s descriptions and understandings of their abusers. Violence Against Women 2006;12:851-865.
18. Mouton CP, Rodabough RJ, Rovi SL, et al. Prevalence and 3-year incidence of abuse among postmenopausal women. Am J Public Health 2004;94:605-612.
19. Zink T, Jacobson CJ Jr, Regan S, Pabst S. Hidden victims: The healthcare needs and experiences of older women in abusive relationships. J Womens Health (Larchmt) 2004;13:898-908.
20. Ballard C, Waite J. The effectiveness of atypical antipsychotics for the treatment of aggression and psychosis in Alzheimer’s disease. Cochrane Database Syst Rev 2006;(1):CD003476.
21. De Deyn PP, Katz IR, Brodaty H, et al. Management of agitation, aggression, and psychosis associated with dementia: A pooled analysis including three randomized, placebo-controlled double-blind trials in nursing home residents treated with risperidone. Clin Neurol Neurosurg 2005;107:497-508.
22. Thornton D. Age and sexual recidivism: A variable connection. Sex Abuse 2006;18(2):123-135.
23. Fazel S, Sjöstedt G, Långström N, Grann M. Risk factors for criminal recidivism in older sexual offenders. Sex Abuse 2006;18(2):159-167.
24. Doren DM. What do we know about the effect of aging on recidivism risk for sexual offenders? Sex Abuse 2006;18(2):137-157.
25. Haywood TW, Kravitz HM, Grossman LS, et al. Psychological aspects of sexual functioning among cleric and noncleric alleged sex offenders. Child Abuse Negl 1996;20(6):527-536.
26. Dickey R, Nussbaum D, Chevolleau K, Davidson H. Age as a differential characteristic of rapists, pedophiles, and sexual sadists. J Sex Marital Ther 2002;28(3):211-218.
27. Fazel S, Hope T, O’Donnell I, Jacoby R. Psychiatric, demographic and personality characteristics of elderly sex offenders. Psychol Med 2002;32(2):219-226.
28. Fazel S, O’Donnell I, Hope T, et al. Frontal lobes and older sex offenders: A preliminary investigation. Int J Geriatr Psychiatry 2007;22(1):87-89.
29. Langevin R, Curnoe S, Bain J. A study of clerics who commit sexual offenses: Are they different from other sex offenders? Child Abuse Negl 2000;24(4):535-545.