Current research into the relationship between schizophrenia and crime Essay

This paper will examine the current literature regarding the twin issues of schizophrenia and crime and the possible effect that the former has on the latter in the general population. Five researches, representing contemporary findings on the theme under discussion have been selected for this review. For each research their research framework will be examined, their findings and conclusions will be presented, and a comparison made between their methodology, highlighting their limitations and strengths.

Theorists examining the issue under discussion have operated under various theoretical frameworks. They suggest that a number of factors can contribute to an increased risk of criminal activity in persons with severe mental illnesses including individual and contextual as well as social factors. Therefore three models have developed. The first suggests that the correlation between criminal activity and the presence of schizophrenia should be examined at the individual level since it is individual adjustment factors that influence criminal activity (see Nordstrom, Dahlgren, & Kullgren, 2006).

The second model suggests that neighborhood-related factors should be the focus of research since these community contexts may influence criminal activity or place persons at increased risk of involvement in criminal activity (See Logdberg, Nilsson, Levander, & Levander, 2004). The third model states that social factors such as victim relations in social and family relationships influence or promote susceptibility to criminal activity (Stompe, et al. , 2006).

There is no perfect model or framework from which to address this issue. Clearly the relationship between severe mental illness and criminal activity must be complex and thus any one model may be inadequate to understand this issue completely. The five researches reviewed in this paper each take different approaches in examining the issue. Comparison will be made as to the utility and feasibility of the models chosen in understanding the issue within the population selected by the researchers.

The first two studies to be examined attempted to verify that the contribution of schizophrenia to violent crimes at the level of the population, as propounded by previous researchers, is truly what it is estimated to be. The researches reported by Fazel and Grann (2006) and Munkner, Haastrup, Joergensen and Kramp (2003) examine schizophrenics’ involvement in criminal activities at the population level. The first study by Fazel and Grann (2006) examined the impact persons with severe mental illnesses had on the occurrence of violent crimes in the Swedish population. The researchers wanted to ascertain: the population-attributable risk: the number of violent crimes committed per 1,000 persons in the whole population that would not have occurred if the risk factor – severe mental illness – had been absent, and the population-attributable risk fraction, which is the proportion of violent crimes in the whole population that may be attributed to individuals with severe mental illness (Fazel & Grann, 2006, p. 1398). This study was a register-based, retrospective study, using data available on health records and criminal involvement. In Sweden there is a system of recording healthcare information as well as criminal activity.

The country assigns a unique 12-digit number to all residents in the country. This number is used in the health care setting to record information on patients’ health care and diagnoses. Of course there are patients who enter the system without one of these unique numbers but less than one percent of the records surveyed included patients without the identification number. In the criminal justice system this number identifies offenders, regardless of type of crime, fine charged, conviction enforced or other penalties. Some personal identification numbers may also be missing.

However during the study period this portion was extremely small representing only 105 of the total 205, 846 convictions. This database was therefore a useful tool to locate and correlate schizophrenics who appear in the justice systems for involvement in criminal activities. Fazel and Grann (2006) conducted a comprehensive analysis of the database on hospital discharges, examining information dated between January 1, 1988 and December 31, 2000. They looked for patients who had been discharged with a primary diagnosis for either schizophrenia or other psychoses.

Additionally patients had to be aged 15 years and older since this was the age of legal responsibility. Also persons with other comorbid diagnoses, including substance abusers, were included in the study. The less than one percent of patients without a personal identification number were also left out of the analysis. Next the researchers searched through the national register for all persons in the selected age category who had ever committed a violent crime within the study period. Again, those with absent personal identification numbers were left out of the study.

The database included records of all criminal activity and the nature of conviction. The database did not only contain information on those sentenced to prison or detention in a psychiatric hospital but included noncustodial sentences and those persons who were simply issued a warning or charged a fine. The researchers utilized a very broad and comprehensive definition of violent crimes, incorporating a wide array of criminal activities. Homicides or attempted homicides, aggravated assault, common assault, robbery, threats, harassment, arson or sexual assault were all included.

The identification numbers of the mentally ill patients were then matched with criminal records, thereby allowing the researchers to filter all patients who had committed an offence during that time. The researchers did not control for time at which criminal activities took place, whether before or after admitting to the hospital for schizophrenia. Additionally the researchers took into consideration every single count of violent crime within each conviction as well as multiple offences by the same individual across time.

All subjects within the Swedish criminal and health databases who met all of the study criteria were included in the study. There were a total of 98, 082 persons with severe mental illnesses within a total hospital population of 218,283. Of the mentally ill patients 44. 2 percent were males. Patients had an average of five hospital admissions. There was hardly any difference between genders. The median number of days spent in the hospital was 66. Patients discharged with a diagnosis of severe mental illness made up 1. 4 percent of the general population.

When personal identification numbers were matched with criminal records 6510 persons with schizophrenia had committed at least one violent crime. This represents 6. 6 percent of the patients who had a mental illness. Comparatively, within the general population, only 1. 8 percent of persons had ever committed a violent crime within the period under study. Persons with severe mental illness committed on average 3. 2 individual counts of violent crimes each. Within the general population the corresponding figure is 2. 3. Additionally there were 1266 repeat offenders within the schizophrenic category representing 19. percent of this population. There were, however only 15. 6 percent of the general population who were repeat offenders. Among persons who were convicted ten times or more there were 21 or 15. 6 percent of those with severe mental illnesses and 118 or 0. 1 percent within the general population. The population-attributable risk factor of violent crimes for persons with severe mental illnesses is 2. 4 for every 1000 inhabitant. This means that there was one violent crime among every 1000 inhabitant every five years that could be attributable to persons with such illnesses. The population-attributable risk fraction was 5. 2 percent.

A high percentage of persons with severe mental illnesses were involved in extremely serious crimes such as homicides (18%) and arson (16%). Analysis of the data by gender found only a minimal contribution of females to criminal activities. The population-attributable risk for females with severe mental illness compared to the corresponding females in the general population was 0. 6 violent crimes per 1,000 female inhabitants between 1988 and 2000. This means that within the entire thirteen year period of the study less than one violent crime was committed per 1, 000 women that was attributable to patients with severe mental illness.

The researchers concluded that only about five percent of all violent crimes in Sweden can be attributed to mental illness. This number, they feel, is sufficient enough to demand attention at more appropriate methods of managing these illnesses in the society so that negative consequences are avoided. Additionally the suggest that factors such as gender, age and type of violent crime had an impact on the populationa attributable risks of mental illnesses. The research was very succint and examined the connection between schizophrenia and crime from a broad, population perspective.

Therefore the data is adequate in highlighting the comparative impact of this illness on crime in any population since Fazel and Grann’s (2006) used a large population for the study. Furthermore the researchers point to sources establishing the validity of the information contained in the database as well as their comprehensiveness and therefore the results from the current survey is an accurate representation of the situation persisting in Sweden between 1988 and 2000 with respect to schizophrenia and crime. The second study reported by Munkner et al. 2003) also did a register-based study. The location of this study was Denmark and the purpose was “to map how the first registered violent and non-violent crimes committed were related temporally with the first psychiatric hospital contact and the diagnosis of schizophrenia among schizophrenia patients” (p. 348). The researchers conducted a database search of the Danish Psychiatric Central Research Register (PCR), the National Crime Register (NCR) and the Civil Registration Systems (CRS). The researchers examined data from November 1, 1963 for persons who were 15 years or older.

Like Sweden Danes also have a unique civil registration number (CRN). The researchers searched the NCR for all criminal activities committed by persons over 15 since November 1963. The researchers also searched the PCR for all patients born November 1, 1963 or later who were diagnosed with schizophrenia. They used the CRS to eliminate persons who had died, disappeared or left the country. This was because the NCR deleted information on such persons. Additionally some candidates were eliminated because their first appearance for psychiatric care was before they were 15. The population size of Denmark was 5. million. Eligible subjects had to have been born November 1, 1963 or later. Only 5,184 individuals within this group were diagnosed with schizophrenia. There were twice as many men as women in the sample throughout all age categories. The researchers classified the offences committed by persons with schizophrenia as violent, including homicide, robbery and arson, or nonviolent offences such as burglary or malicious bodily harm. The researchers found that 41 percent of persons with schizophrenia committed at least one offence, and 17 percent committed at least one violent offence.

Furthermore, most of the criminal acts committed by male schizophrenics, especially their first violent offence, occurred before their first entrance into a psychiatric hospital (71%). These findings reveal that the schizophrenics are more likely than the general population to be involved in criminal activities. Further there is the suggestion that schizophrenia, especially in males, is not being diagnosed early enough and thus they are not receiving proper psychiatric treatment. The third study examined for this paper was also register-based.

However, unlike the previous two studies, the study reported by Nordstrom, Dahlgren and Kullgren (2006) focuses only on the male population and on only one category of violent crime – homicide. Their purpose was “to identify possible triggers and contextual circumstances and their association with offender – victim relations” (p. 194). The researchers examined court records of persons accused of homicide who were referred for a forensic psychiatric exam (FPE). The FPE was not routine but done during pre-trial only where the mental state of the offender was in question.

Thus the subjects were male homicide offenders with a diagnosis of schizophrenia. The researchers commenced by scrutinizing the FPE records of all male homicide offenders between 1992 and 2000, available through the register of the National Board of Forensic Medicine, to determine those who had a diagnosis of schizophrenia. A total of 48 males fitted the inclusion criteria. Next the researchers placed the offenders into two categories based on the information provided about the victim of the offence.

Those crimes involving a member of the family such as parents, grandparents, siblings and spouse were classed as family victim. The other group was categorized as other victims. The purpose of this classification is to allow the researchers to analyze the victim relation patterns and to understand the nature of offences committed against family members. This is in view of the previous findings that the researchers examined in which it was found that family members are at the most risk for severe assault or homicide committed by persons with schizophrenia. The offenders ranged from 19 to 58 years with a mean age of 32. years. The offences committed included double murders resulting in a total of 52 victims Twenty one of these victims were family members (8 males and 13 females), they were killed by 18 offenders, representing three of the four double murders. Forty percent of the victims were immediate family members to the offenders. In the other victims group there were 30 offenders killing 31 victims (26 males, 5 females), with only one double murder. Additionally most victims were acquainted with the offender (82. 7%) with only nine percent who were complete strangers to the offender.

Varying types of weapons were used in these crimes, the most prevalent of which was a knife (51. 9%). Other weapons included dumbbells and a frying pan. Only one case involved the use of a firearm. No weapons were used in only 7. 7 percent of the homicides. Homicides where the perpetrators knew their victims took place more frequently at either’s home (81%). Comparisons between family victims and other victims group showed there were more female victims in the first; 13 of the female victims of homicides were family members, including mothers, a sister and a grandmother.

In the family victims group the homicide most often occurred in the home. Intoxication of both victim and perpetrator was more common in the other victims group (64% of victims and 63% of offenders). One of the most important findings of the research was that most offenders had had previous psychiatric contact (79. 2%) and just over 33 percent were undergoing treatment at the time of the homicide but just under half had been prescribed antipsychotic medication and only a negligible sum (4. 2%) was taking them.

The researchers concluded that feelings of threat or loss of control on the part of the perpetrator is an ingredient in determining propensity to commit a violent crime. In this research more than 50 percent of the perpetrators were under the effects of delusions or hallucinations which prompted fear and the feeling that they were not in control of the forces around them. Additionally co-occurring intoxication as well as failure to adhere to prescribed medication may also influence involvement in violent activities.

Moreover family members are at a higher risk of been killed by persons with schizophrenia. This study is useful in understanding the interplay of schizophrenia and crime within the male population and the way in which immediate family members of such offenders are affected by their violent behavior. The survey has very little usefulness, however, to an understanding of the phenomenon in the general population given the very small sample size and the researchers’ exclusive focus on only one manifestation of criminal activity, homicide.

However the research adds to the existing knowledge of factors that trigger criminal behavior in persons diagnosed with schizophrenia. Thus far research has tended to be strictly at the population level, using only data available from databases. These approaches, though they may give a good view of the criminal trends of persons diagnosed with schizophrenia in the general population, do not tell the entire story. Databases only contain information on persons who have entered the criminal justice system and thus criminal activity that does not reach noticeable levels is not accounted for.

Furthermore these methods are unable to describe the unique experiences of individuals in the community, both those who perpetrate criminal acts and those who are, in one way or another affected by them. The last two researches take some of these considerations in mind. The first of these, reported by Logdberg, Nilsson, Levander and Levander (2004) used a mixed methods approach. The purpose of the research was “to investigate the relationship between the prevalence of schizophrenia and measures of social disorganization, self-reported victimization and fear of crime within various neighbourhoods” (p. 3). The survey site, Malmo, is one of the largest metropolises in Sweden. It boasts a population of a quarter million and is subdivided into 134 administrative neighborhoods. The researchers used two methods to acquire data for the survey. The first was a register-based search of schizophrenic patients and their living conditions and the second a survey among citizens in the community on their self-reported fear of crime. Search of a computerized database for 2000 data on persons in Malmo diagnosed with schizophrenia produced 1344 persons.

Eighty one percent of an initial population of 7000 community residents participated in the survey, the second aspect of this research. This consisted of a 68-item questionnaire, measuring social integration and control, fear and victimization, which was mailed to the respondents. Pearson Product Moment Correlation was used to assess correlation between the different social variables within the communities of Malmo. Findings reveal that schizophrenia presence in Malmo communities ranges from zero to 1. 42 percent.

Additionally schizophrenic persons resided primarily in areas considered socially disorganized, that is, areas with a combination of poor social integration, weak informal social control and high levels of public disturbances (Logdberg et al. , 2004, p. 95). The researchers found that patients with schizophrenia in socially disorganized, inner-city communities had higher levels of agitation than those resident in rural areas suggesting that the community context is a constraining factor towards certain socially disturbing behaviors.

The primary benefit of this research is that it gives the actual perceptions of community dwellers rather than simply a global analysis of the situation. The use of the survey ensured that persons who are truly cognizant of the day to day situation in the communities are giving their perceptions and thus the data is truly representative of the existing feelings in the community. However, like the preceding research this study’s sample size is small, representing just over two percent of the actual population in Malmo. However the subjects and methodology chosen were adequate.

One important concern brought about by these researchers is that: “the social behaviour of persons with schizophrenia may increase the social disorganization, perceived fear, social disorder and crime in the neighbourhood, as well as the negative attitudes among the residents towards such persons” (Logdberg et al. , 2004, p. 96). This implies that the effects of the criminal activities of persons with schizophrenia does not affect the community only in regards to direct criminal and violent acts committed but may also impact the community spirit negatively.

The final study also attempts to understand the schizophrenia and crime issue from an individual level to try to predict individual factors that may predispose or influence persons being involved in criminal activity. This research is unique, as compared to the four previously discussed, in that it does not consider schizophrenics who are involved in criminal activity in a vacuum from other persons in the population, including those with schizophrenia who have never committed a violent act. The purpose of the study reported by Stompe et al. 2006) was “to examine the relevance of these factors [socioeconomic status and family structures] in the development of criminal behaviour in patients with schizophrenia” (p. 555) as compared to other nonoffenders with similar diagnosis or other healthy offending or nonoffending individuals. The study was conducted in Austria. Four groups were selected for cross analysis and correlation. The first group consisted on healthy individuals who had not been convicted of a crime – labelled healthy nonoffenders.

The second group included persons who were not diagnozed with schizophrenia or any other major mental illness but who had been convicted of a crime – labelled non-schizophrenic offenders. The third group was made up of persons diagnozed with schizophrenia but who had never been convicted of a crime – labelled schizophrenic nonoffenders. The final group was made up of persons diagnozed with schizophrenia and who had also been convicted of a crime – labelled schizophrenic offenders. The Structured Clinical Interview for DSM-III and the Socialization Report (SOREP) were used to gather data.

The first instrument was used to verify diagnosis of schizophrenia and the second gather comprehensive data on family characteristics including education, employment status, parental socioeconomic status, family structure, age at which parent died, birth order and extended stays in foster homes. There were 103 subjects included in each of four groups, to represent a total of 412 participants. They were recruited from a variety of settings in the community including psychiatric detention centers, psychiatric clinics and correctional institutions.

There were between group matches in terms of age and severity of offence where applicable. Chi-square tests, one-way ANOVA and the Duncan’s new multiple range test were used in analysis. The level of significance for correlations between variables was set at a p value of <0. 01 using univariate tests. SPSS® 12. 5 facilitated data analysis. This study produces findings similar to what was found in the previous research discussed in this paper by Logdberg et al. (2004). Persons diagnosed with schizophrenia, in addition to other offenders, were disproportionately represented in families of a lower social classs.

It can therefore be assumed that these families also reside in socially disorganized inner-city communities. The absence or death of a parent during childhood was more frequent among patients diagnosed with schizophrenia, including both offenders and nonoffenders. This parental absence was primarily in the form of an absent father figure in the home. This issue further adds fuel to the ongoing debate on the importance of the father in the home and the consequences of single parent, moreso single-mother, families.

Additionally having step siblings in the home during childhood seems to be a risk factor for offending behavior in both healthy and schizophrenic offenders. Having a parent who had being diagnosed with schizophrenia was more prevalent among the schizophrenic population, but more so among those who were criminal offenders. Lastly not having completed school or completing without adequate qualifications was a predictor of developing schizophrenia in later life. The methodological design of the current research was very strong.

By using four groups the researchers were able to predict the relationship of various environmental factors to the presence of either schizophrenia or offending behavior or both. In this way the researchers were able to exemplify, better than any of the other researches presented in this paper, the possible factors that would either predispose or place persons at an increased risk of developing schizophrenia and for those persons to become involved in violent behavior. What the researchers fail to account for is the nature of the offences for those in the offending groups.

The other researchers discussed in this paper specified the categories of offences that were committed, some specified their severity. Thus, while this research is useful because of its strong methodological framework, its findings are still not completely trustworthy if they are to be assessed relative to the other works here presented as the researchers fail to control for some variables that other researchers have. It is important to establish a comprehensive comparison of the researches presented, in order to determine further areas of contradictions as well as to discover similarities in the research methods, procedures and findings.

First there were variations as well as similarities in the statistical procedures the researchers used during analysis. Pearson Product Moment Correlation and Chi-Square tests were the most frequently used to estimate relationship between data variables and probabilities. Nordstrom et al. (2006) used the Pearson’s chi-square along with t-tests and Munkner et al (2003) used Chi-square, for instance. One important variation between the surveys was the different levels of significance used. The p-value for the research reported by Munkner et al. (2003) was <0. 5 while Stompe et al (2006) and Logdberg et al. (2004) used levels varying from <. 001 to <. 05. This variation means that the statistical significance of data produced in these researches are not equivalent and thus when analyzing and comparing the researches collectively this has to be taken into consideration. As previously discussed the sample sizes used in the researches were very varied from an extremely large sample to one that was extremely small. Evidently those researches that examined the issue from the level of the population had a much larger sample size.

The report by Fazel and Grann (2006) represented 98, 082, the one by Munkner et al. (2003) with 5,184 and by Logdberg et al. (2004) 1344 schizophrenics along with approximately 5670 community dwellers. These represent the largest populations studied among the five researches reviewed in the papers. The other sample sizes were 48 (Nordstrom et al. , 2006) and 412 (Stompe, et al. , 2006). The researchers were not often similar in the choice of study location. There was an imbalance of researches conducted in Sweden, representing three of the five studied in this paper.

Of these one was conducted at the population level (Fazel & Grann, 2006), one at the community level (Logdberg, Nilsson, Levander, & Levander, 2004) and the other among a specific sector of the country, male schizophrenic offenders, regardless of geographic location. For the two remaining researches one was conducted in Denmark (Munkner et al. , 2003) and the other in Austria (Stompe, et al. , 2006). The prevalence of research in Sweden has been explained and justified by the researchers operating in this area.

Sweden, they suggest, has one of the most comprehensive databases in the western world and researchers worldwide acclaim it to be one of the most useful sources of data on population-wide medical and criminal reports (Fazel & Grann, 2006). However Denmark is as good a location to conduct a study since their National Crime Register (NCR) is felt to be “the most thorough,comprehensive and accurate crime register in the western world” (Wolfgang as cited in Munkner et al. , 2003, p. 348).

All but one of the researches studied in this paper examined the relationship between violent behavior and schizophrenia as the only category of mental disorder. Fazel and Grann (2006) included other psychoses. The other psychoses included those with: “affective psychoses, paranoid states, other nonorganic psychoses, persistent and induced delusional disorders, acute and transient psychotic disorders, manic episode, bipolar affective disorder with psychotic symptoms, and depressive disorders with psychotic symptoms” (Fazel & Grann, 2006, p. 398). Though Fazel and Grann suggest that these other psychoses made up a relatively small proportion of the population studied, their inclusion of this category still has significant implications for the validity and generalizability of their findings. Their findings are useful in so far as they give an understanding of the contribution of mental illnesses to violent crime but is unable to specify the exact contribution of schizophrenia to these crimes.

In the same way that they estimate the possible number of persons in the sample who had other psychoses besides schizophrenia is the same way that the relative contribution of schizophrenia to the activities they presented can only be estimated. There was a similar disparity in the definition or categories of crimes that were included in the discussion on criminal or violent activity. Fazel and Grann (2006) and Munkner et al. (2003) used very comprehensive criteria to define criminal activities.

They included various categories of homicide and even attempted homicide as well as robbery, assault, theft, harassment and other serious and not so severe crimes. Nordstrom et al. (2006), on the other hand restricted their study to the effect of schizophrenia on homicides alone. The other two researches dealt with the issue of crime and violent activity very superficially, choosing rather to focus on general perceptions about criminal activity rather than trying to specify the types of criminal activities that schizophrenics were involved in. What is also commendable in the researches is that most, Nordstrom et al. (2006), Stompe et al. (2006) and Logdberg et al. (2004), used the same criteria for diagnosis – the Diagnostic and Statistical Manual of Mental Disorders (DSM). For the research by Fazel and Grann (2006) the diagnosis criteria used in the Swedish database was unspecified and for the Munkner et al. (2003) survey, the International Statistical Classification of Diseases and Related Health Problems (ICD) criteria were used. One major flaw with the studies that depended heavily on information obtained through a database is that these data are never complete and persons never entering these databases are not accounted for in discussions.

Fazel and Grann (2006) for example, had to exclude persons from analysis who did not have any personal identification number but whose names appeared in either the criminal or hospital records. While the estimated effect of this excluded data does not appear to be large given the still large sample size produced and given that less than 2 percent of the data was affected in this way there is still the concern that some segments of the population are not appropriately accounted for in register-based searches and as such, wherever possible, alternative or supplementary data gathering techniques should be employed.

The size of the Fazel and Grann (2006) sample is sufficiently large to justify the use of this methodology. The alternative would have been to design instruments to survey the population and this, in itself, has limitations. There is the chance that the response rate would be below the expectations of the researchers, there is always the likelihood of bias or inaccuracy in the information provided by persons in the population who are attempting either to be outright deceptive or unable to accurately complete whatever survey instrument.

In the questionnaire aspect of their survey Logdberg et al. (2004) received an 81 percent response rate, which is commendable. However this is not always guaranteed given that completion of questionnaires is done on a voluntary basis. Fazel and Grann (2006) admit that their choice of a register-based study possibly underestimates the true picture of violent behavior among schizophrenics. This same argument can therefore be levelled against the other research in this paper that used a database search as its sole methodology. The research reported by Logdberg et al. 2004) is unique in that it incorporates the dual strategies of a register-based and a questionnaire method. However the same limitations that affect both types of research methods also apply to that research.

One other thing that population based studies are unable to predict is the causal relationship between the several variables and participation in criminal activities by schizophrenic persons. Register-based studies can only observe and are unable to truly predict a relationship between factors. Studies done at the individual level, such as the one by Stompe et al. 2006), are able to isolate various factors and determine their associations in order to predict causal variables. Perhaps the most useful survey, with respect to methodology, is the one reported by Logdberg et al. (2004) as it employs both a register-based and a questionnaire type survey. Moreover the database information did not form the primary focus of analysis, rather the data contained in the questionnaires. The primary framework that has been used in the assessment of schizophrenia on crime and violent activities is usually community-focused.

Those studies utilizing data contained in population-wide databases espouse the view that the issue should be understood, analyzed and thereby addressed from a population perspective. These researches do not focus on identifying specific etiologies whether at the individual or community level but address it by examining the population prevalence of schizophrenia and estimating the possible attribution of violent crimes to this diagnosis. It is in this vain that the effects of schizophrenia on crime can be viewed as a public health issue with the populations studied.

When researchers are able to demonstrate that the problem is widespread across the population then policies devised have to target it at that level. Nevertheless, in order to impact and improve the national prevalence of schizophrenic crimes, it is important to determine the possible environmental factors, at the level of the family and the community that may adversely put some persons at risk for either developing schizophrenia or being involved in criminal activities.

Once these are understood then public policies can target these areas specifically. Logdberg et al. (2004), Nordstrom (2006) and Stompe et al. (2006) all examine the issue from a community level. Logdberg et al. (2004) revealed that environmental factors such as living in an inner city which concurrently experiences high levels of social decay, is a potential risk factor for involvement in criminal activity. Nordstom et al (2006) discusses characteristics within the family and social relationships that may predict involvement in criminal activity.

They found that most victims of homicides committed by persons with schizophrenia were persons in their social network, more so family members and thus these persons are at an increased risk of becoming a victim of homicide. Furthermore they highlight that the family is probably the most important social unit for schizophrenics since their condition often isolates them from many other social contacts. Stompe et al. (2006), though examining individual home characteristics, also identify the social context as an important influence on violent behavior.

They, like Logdberg et al (2004) identify socioeconomic factors within the family, even before the age of accountability, as contributing to the developing of these problems. The findings from this review of current literature on the issue of the impact of schizophrenia on criminal activity has various implications for practice, and are important issues to be considered by those who interact with these persons in the community on a daily basis. One of the most important groups of persons that this research speaks to is those in the mental health care services.

The researches presented discuss some issues that speak volumes to the flaws that exist in the system with respect to the treatment and proper management of schizophrenia. In the study reported by Munkner et al. (2003) ten percent of the males who had been convicted for their first crime had had previous contact with the psychiatry health system. However these men had not yet been diagnosed with schizophrenia. Furthermore the finding that most criminal activities are committed before the offender is initially diagnosed is also worrying.

What this suggests to those involved in treating is that there is not an effective system to ensure proper and early diagnosis of schizophrenia before it becomes uncontrollable. Also there are weaknesses in how the disease is managed since persons diagnosed are still being involved in criminal activities to the extent found by the five sets of researchers being discussed in the current paper. Another issue of note, also relevant to therapists is the rate of compliance of patients to their drug treatment. Nordstrom et al. 2006) pointed out that less than five percent of persons given medications were compliant. This forces the question of how to make any impact on schizophrenia if patients are refusing to take their medication or that therapists are not insisting or enforcing compliance enough for it to be a habit among these patients. Similarly less than half of the persons in that study had been prescribed any form of medication. This is in light of the fact that most of them were apparently significantly impaired by the condition.

The importance of this group of people in impact the outcomes of schizophrenic patients cannot be overemphasized. As Nordstrom et al. (2006) argue, they have a great role to play in helping to prevent violent crimes perpetrated by persons for whom they care or who need treatment but have not yet been noticed within the system. It is indeed a tremendous duty that personnel in the mental health services have to fulfill and it may not be feasible without additional help from the rest of the population.

Criminal justice personnel need also to take warning with these results. One of the issues that came about within the three studies that conducted a register-based survey was the possibility that not all persons with schizophrenia who had committed a violent crime had been identified in the system. In fact there is the suggestion that some of these persons, when convicted, were passed into general detention facilities, removing them from possible treatment.

The research by Fazel and Gran (2006) may not be prone to this weakness since, as they reveal, within the Swedish system persons are convicted regardless of their mental state, though this is taken into consideration for sentencing. Nevertheless it is not unlikely, and actually extremely possible, that persons with schizophrenia went unnoticed into detention facilities, having been convicted and sentenced for a violent crime without their condition been known. The researchers discussed in this paper all contain pertinent information on the issue of the relationship between schizophrenia and crime.

The researches reveal that schizophrenia does in fact have an influence on criminal activities within the general population and that, regardless of the size of the impact, it is a cause for concern. Register-based, questionnaire and other survey type instruments were used in data analysis. Though each of them has weaknesses they still contributed to a broadening of the understanding of current research frameworks and strategies. The composite findings of these researchers are of particular relevant to those in the mental health and criminal justice professions.