Mortality Rates of Males Who Commit Parricide or Other Violent Offense Against a Parent

Materials and Methods

This study forms part of a research project on parricidal offenders, with a particular focus on matricidal offenders, comprising information on all males who underwent a forensic psychiatric examination in Finland from 1973 to 2004 subsequent to committing a violent offense against one parent. The established practice in Finland is that a large proportion of all homicidal offenders and others who have committed serious violent offenses undergo a thorough forensic psychiatric examination as a part of a trial procedure.³² The National Authority of Medicolegal Affairs arranges forensic psychiatric examinations at the request of the court. During the past decades, approximately 200 examinations have been conducted each year, mainly on offenders charged with homicides or other serious violent offenses.^32,33 The examination is an extensive inpatient psychiatric investigation in special hospital settings lasting a maximum of two months. It contains data collected from various sources, interviews by a multiprofessional team, a comprehensive psychiatric evaluation with diagnostic procedures, standardized psychological tests, examination of the offender's physical condition, and observation of the offender by the hospital staff. The conclusions include diagnoses as well as the assessments of criminal responsibility and the need for psychiatric hospitalization.

In Finland, the minimum age of criminal liability is 15 years. In our study, a violent offense was defined as homicide or attempted homicide, aggravated assault, common assault, robbery, illegal threatening behavior, harassment, arson, or any sexual crime. In cases in which the victim was a stepparent or foster parent, only those were included in which the step- or foster parent had joined the family when the son was under the age of eight years. The 15 subjects who had offended against both parents during the period were excluded, as our purpose was to characterize differences between the offender groups subdivided by the sex of the victim. In this study, the terms matricidal and patricidal are used for all violent offenses against a parent, as the terms matricide and patricide refer only to the cases in which the parent died. The study sample includes 99 matricidal offenders and 113 patricidal offenders. A control group for the patricidal offenders, matched for the violent act, sex, and age of the offender was gathered from the mental state examination report archives (n = 111). Table 1 presents the distribution of the offenses among the offender groups.

The mean age of the matricidal group (30 years) was significantly higher than that of the patricidal group (27 years, p = .014) and the control group (26 years, p = .002). Six matricidal offenders and six patricidal offenders committed the offense while under the age of 18 years. Forty-two percent of the matricidal offenders, 27 percent of the patricidal offenders, and 17 percent of the control subjects had a psychotic disorder (p < .001), while a personality disorder was diagnosed in 51, 64, and 79 percent of the matricidal, patricidal, and control offenders, respectively (p < .001). Fifty-nine percent of the matricidal offenders, 46 percent of the patricidal offenders, and 69 percent of the control subjects had an alcohol or drug-related disorder as the primary or secondary diagnosis (p = .003).

The forensic psychiatric examination reports of the offenders were retrospectively analyzed for several variables regarding offender and offense characteristics. The background variables used in this study were index offense, age at death, psychiatric diagnosis, and criminal responsibility. Accurate information on all the subjects’ mortality was collected in November 2007 by record linkage with the Cause of Death Register at Statistics Finland, using personal identification numbers assigned to all residents in Finland. All subjects were followed up until the end of November 2007, or earlier in the event of death. The official death register includes personal information on the deceased and the date and cause of death, classified according to the International Classification of Diseases, according to the eighth revision (ICD-8) until 1986,³⁴ ICD-9³⁵ between 1987 and 1995, and ICD-10³⁶ from 1996 onward. In this study, the deaths were classified in the following way: natural, unnatural, or unknown; accidental (yes/no); and suicide (yes/no); and method of suicide (hanging, intoxication, or other). In each of the three groups studied, there were two cases without exact information on the cause of death. The number of deaths and the causes were compared with mortality data for the general Finnish population, obtained from Statistics Finland.

The original study protocol was approved by the Ethics Committee of Oulu University Hospital on August 19, 1996. The National Authority of Medicolegal Affairs and the Ministry of Social Affairs and Health and Statistics Finland approved the study. The permission to make a register link to causes of death was obtained from the Cause of Death Register.

The statistical significance of group differences in categorical variables was assessed with the Pearson chi-square test or Fisher's exact test when appropriate, and the differences in continuous variables with Student's t test; p < .05 was defined as the limit of statistical significance. We calculated standardized mortality ratios by comparing the observed number of deaths in the study subjects with the expected number of deaths as predicted by the age- and sex-specific mortality rates of the general population. The result was expressed as the SMR, together with its 95 percent confidence interval (CI). The age-specific SMRs were determined for four age groups (15–24, 25–49, 50–64, and 65–79 years). A Cox proportional hazards model was used to estimate the survival time from the date of offense to the date of death of the groups studied in the subcategories of natural deaths, suicides, and unnatural deaths other than suicide, separately. All data were entered on computer in the SPSS Program for Statistical Analysis.

Results

Mortality and Criminal Responsibility

During the follow-up, 26.3 percent of the matricidal and 14.3 percent of the patricidal offenders with full responsibility had died, compared with 21.1 percent of matricidal and 32.8 percent of patricidal offenders with diminished responsibility. The proportions of the dead subjects in the groups defined as not guilty by reason of insanity were 16.7 percent in the matricidal and 32.4 percent in the patricidal groups. Figure 1 illustrates the proportions of deceased subjects in the parricidal offender groups according to criminal responsibility. No statistically significant differences were observed between the offender groups in any category of criminal responsibility.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

The proportion of deceased subjects according to criminal responsibility among 99 matricidal, 113 patricidal, and 111 control offenders who underwent a mental state examination in Finland from 1973 to 2004.

Survival After the Date of the Crime

When the Cox proportional hazards model was used, no association of matricidal, patricidal, and control offender groups with survival time was found among the offenders who died due to natural causes or unnatural causes other than suicide. In the case of suicides, however, the matricidal offenders had a significantly shorter survival time (mean, 5.1 years; SD 2.8) after the offense than did the patricidal (mean, 8.9 years; SD 8.1) or the control (mean, 13.1 years; SD 7.9) offenders. After adjustment for age at date of offense and status of victim (died, survived) the hazard ratio (HR) of death by suicide was higher among the matricidal (HR = 4.8; 95% CI, 1.11–20.86; p = .035) but not the patricidal offenders, when compared with the control offenders (Fig. 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Survival time after the offense in cases of offender suicides among 99 matricidal, 113 patricidal, and 111 control offenders.

Discussion

This study provides systematic information on mortality among parricidal offenders. We used a nationwide sample of parricidal males who had undergone a mental state examination. The SMR of the parricidal offenders aged 25 to 49 years was higher than that of the general population.

A novel finding was that the matricidal males who died of suicide had a significantly shorter survival time after the offense than did the patricidal and control offenders who died of suicide.

It is a well-known fact that the mortality rates among offenders and psychiatric patients are elevated compared with those of the general population.^4,37–42 In our study, 20 to 30 percent of the parricidal offenders died during the follow-up of 3 to 24 years. In a study with a sample of 53 mentally ill parricide offenders discharged from high-security care, 17 percent died during follow-up, as the mean length of follow-up was 7.2 years.²⁴ Also, the proportion of deceased offenders in our study is in accordance with the mortality rates presented in studies of offenders in general, when taking into consideration the different follow-up periods. For example, in a Swedish study concerning offenders who underwent a forensic psychiatric examination, a total mortality of 17 percent was determined during a follow-up ranging from 13 to 16 years.¹ Of Swedish homicide offenders, almost half died during the follow-up of 22 to 32 years.⁶ In our study the mean age at death was low in every offender group, being slightly over 40 years, which is consistent with the data in earlier studies.^3,43

In our study, the all-cause SMRs were increased three- to fourfold, respectively, among the matricidal and patricidal offenders of the 25- to 49-year age group compared with that of the general population of the same age. This result is in accordance with that of a study of mentally disordered parricidal offenders, in which the researchers found an almost fourfold increase in mortality among male offenders compared with that of the general population.²⁴ In comparison with offenders in general, the overall mortality rate has been estimated to be 4-fold among ex-prisoners,⁴ 5-fold among habitually violent offenders,⁷ 3-fold among homicide offenders,⁶ and up to 13-fold among mentally disordered offenders,² compared with that of the general population. A recently published study of patients discharged from a medium-security care unit showed a sixfold increase in all-cause risks of death compared with that of the general population.⁴³ Thus, parricidal offenders have a moderately increased mortality risk, similar in magnitude to that presented in most of the studies in which different offender samples were used. In general, offender mortality rates tend to be proportionately higher among the younger age groups when compared with the general population.^3,9–11 One explanation for this may be that antisocial behavior, risk-taking, and impulsiveness tend to diminish with increasing age, which may account for our results, as well.

Overall, the proportion of deaths due to unnatural causes (suicides, accidents, and homicides) was high, constituting approximately 50 percent of the total deaths among the matricidal and patricidal offenders. One-third of deaths among the parricidal offenders were attributable to suicides. Kullgren et al.⁴⁴ reported that among male offenders referred for forensic psychiatric examination and followed up for four to eight years, 56 percent of the total deaths were unnatural: 37 percent from suicide and 19 percent from accidents and homicides. In a large cohort of males with a history of imprisonment, accidents accounted for 22 percent, suicides for 17 percent, and drug overdose for 18 percent of all mortalities.⁹ With respect to the general population, among the Finnish working-age males in 2006, 13.3 percent of total deaths were due to accidents and 8.9 percent to suicides.⁴⁵ In our study, 6.0 percent of all matricidal offenders (n = 99) and 7.9 percent of all patricidal offenders (n = 113) committed suicide, which exceeds the suicide rate in the general population. The percentage of suicides among the parricidal offenders in our data set is in accordance with the completed suicide rate of 7.1 percent found by Bourget et al.¹⁵ among parricide offenders.

The challenge for the health care and criminal justice systems is to recognize those at risk. It is known that prisoners and released prisoners, as well as those discharged from forensic psychiatric units, have increased risk of committing suicide.^4,5,43 The suicide risk of released prisoners seems to be at its highest during the weeks following the release.⁵ Parricidal offenders have many factors that may increase the suicide risk (e.g., high psychiatric morbidity, social disadvantages, being stigmatized by the community, and guilt over the parricidal offense). Those responsible for the treatment and follow-up of parricidal subjects should be alert for the risk of suicide, and particular attention should be paid to the period after the offender is discharged from the institution. More information is needed on the specific risk factors for suicide.

Of unnatural deaths, suicides predominated in the matricidal group, whereas accidents constituted most of the unnatural deaths in the control group. These proportional differences may reflect the differential psychopathology and lifestyles among the matricidal and control offender groups. An interesting finding was that the matricidal offenders who died of unnatural causes and suicides were also younger at death than were the corresponding control subjects. Those matricidal offenders who committed suicide died a shorter time after the offense than the corresponding patricidal or control offenders. On the basis of the shorter temporal relationship between the offense and the suicide, it may be speculated that suicide is associated more strongly with the offense among matricidal offenders than among patricidal and control offenders. It is probable that the guilt and shame caused by the offense decrease in the course of time, having less impact as a factor that contributes to the suicide.

The briefer survival time after the offense among the suicidal matricidal offenders may illustrate the different underlying dynamics of the offense and perhaps the emotional connection to the mother. Although the study also includes offenses that did not result in the death of a parent (attempted homicides and other violent offenses), the psychological break in the relationship with the mother may have resulted in the nonsurvival of the offender. According to the attachment theory, the child's tie to the mother is the most fundamental in the psychological development of the child and the tie usually continues to be significant into adult life, more strongly for some individuals than for others.⁴⁶ Green¹² found that many of the matricidal males had failed to outgrow the attachment and dependence on their mothers. The killing of a mother has emotion-laden meanings, as it is committed within a mother-child bond that is usually the first relationship of the child and remains strong throughout the course of life.

As shown in previous work with the same study sample as was used in this study, the motives for the patricidal offenses included more interpersonal conflicts and confrontation with the victim than did the matricidal offenses, which may reflect the relationship with the victim—for example, abuse of the son by the father.⁴⁷ A limitation in the current study is that specific variables regarding the quality of the parent-child relationship are not available. This question should be investigated in further studies.

Our previous work on the present data showed that the matricidal offenders had psychotic disorders more commonly than did the patricidal offenders.⁴⁷ In addition, paralleling the major mental psychopathology found in this group, homicidal matricidal offenders were more commonly found to be criminally irresponsible. According to Finnish legislation, offenders without criminal responsibility are not sentenced but are committed to a mental hospital as criminally insane.³² Thus, nearly all offenders with psychotic disorders are ordered to be admitted to a forensic hospital for intensive treatment. In our study, the degree of criminal responsibility was not associated with mortality. Similarly, mortality was not associated with psychiatric diagnosis or disposition (prison or inpatient forensic psychiatric care) among Swedish homicide offenders.⁶ On the other hand, the mortality of mentally disordered offenders discharged from forensic psychiatric care has been estimated to be remarkably high when compared with that of the general population.^2,43 The long hospitalization probably had a preventive effect on unnatural deaths in the psychotic group, and therefore the mortality of the offenders with no criminal responsibility did not exceed the death rates of offenders with nonpsychotic disorders.

The main methodological weakness of this study is the inevitable small sample size, which causes a loss of statistical power. For example, due to the small number of subjects, it was not possible to calculate cause-specific SMRs for the matricidal and patricidal groups. Because parricide is a rare phenomenon, a large sample is difficult to acquire, and statistical analysis remains restricted. Hillbrand and Cipriano⁴⁸ have recently illustrated difficulties in accessing a statistically relevant nonbiased sample of parricidal offenders. In Finland, during the years 1998 to 2006, from one to eight parents per year (altogether 26 mothers and 21 fathers) died in parricides.⁴⁹ Violent offenses other than homicides were included in our data, on the basis that studies comparing parricides and attempted parricides have not found major differences.^14,17,48 The proportion of offenses other than homicides or potentially lethal offenses was small. On the other hand, this was a study with access to a nationwide database, containing all male parricidal offenders who had been admitted to undergo a forensic psychiatric examination. The study sample lacked those offenders who committed suicide immediately after the offense. Homicide followed by immediate suicide is an infrequent phenomenon, however. In a recent Finnish study covering one year, there were 10 homicide-suicides altogether, with the homicide victim being the spouse, or children, or both in every case.⁵⁰ It may be assumed that the data set includes almost all offenders with psychotic disorders and most offenders with severe personality disorders, as some offenders with less severe mental disorders may not have been thoroughly examined and were not included in the study.⁵¹ The data set probably covers almost all homicides, as the more severe offenses lead more commonly to the mental state examination, but for less severe offenses, the coverage may be of a lower level.

The advantage of our study is that all deceased offenders could be traced reliably via the national death register system. The availability of high-quality mortality data for the general population ensures the validity of comparisons.⁵² In comparisons with the general population, one has to bear in mind that parricidal offenders differ from the general population in many variables that we were not able to control in this study.

Conclusions

The present study illustrates the vulnerability of parricidal offenders in terms of mortality. Mortality rates of parricidal offenders were found to be moderately increased compared with that of the general population in the age group of 25 to 49 years. The results suggest that parricidal offenders frequently die of unnatural causes and suicides, as do violent offenders in general. Suicides of matricidal offenders appear to be temporally more closely associated with the offense than the suicides of patricidal and control offenders. Several risk factors (e.g., mental morbidity and social disadvantages) contribute to the increased risk of death among parricidal offenders. The risk of suicide as an ultimate consequence of a long-term process should be recognized and taken into consideration during treatment. Appropriate interventions to prevent suicide should be provided within the health care system and prisons. On the basis of this study, matricidal offenders in particular seem to be at risk during the years after the offense. In the future, it is important to investigate the factors that influence the recovery and psychological survival of parricidal offenders.