George Karystianis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
George Karystianis has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Tony Butler
University of New South Wales, Sydney, New South Wales, Australia
20 shared publications - Goran Nenadic
Department of Computer Science, University of Manchester, Manchester, UK
19 shared publications - Armita Adily
School of Population Health, University of New South Wales, Sydney, Australia
11 shared publications - Peter R. Schofield
Discipline of Psychiatry and Mental Health, School of Clinical Medicine, Faculty of Medicine and Health, University of New South Wales, Sydney, Australia
10 shared publications - Wilson Lukmanjaya
School of Population Health, University of New South Wales
10 shared publications Tony Butler
School of Population Health, University of New South Wales
9 shared publications
Latest Publications
- Open Access
- 11 November 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Protocol for a Feasibility Study of Computational Text Analysis of Child Protection Intake Reports in South Australia (Preprint)
UNSTRUCTURED Background: Child maltreatment statistics collected using surveys, in-depth interviews, and manual case reviews are limited in scope due to resource constraints and labour-intensive, non-scalable approaches. The advent of computational advancements such as natural language processing, text mining, and large language models opens up the possibility of rapidly analysing vast numbers of documents from child protection client information systems to generate useful information. However, often due to strict data access protocols, few research efforts have focused on the application of computational text analysis in the child maltreatment area. Objective: This study presents a protocol for the application and evaluation of a text mining method on 10,000 child protection intake reports from South Australia’s Department for Child Protection. Methods: The protocol is split into five phases: 1) Ethics, governance, and data access approvals; 2A) Establish gold standard for the automatic extraction of information from 100 redacted reports using manual expert annotation; 2B) Contextual de-identification to preserve relationships; 3) Text mining method development, validation and evaluation based on 300 reports; 4) Text mining application on large sample; and 5) Evaluation and data analysis. Discussion: This protocol describes a feasibility study applying computational text analysis to highly sensitive child protection reports to automatically extract key information. Key considerations and challenges included governance, ethics, security, and technical requirements across multiple stakeholders involved in the study and adapting the text mining methodology to address the loss of contextual information caused by redaction. Conclusion: Whilst a high level of trust between academics and the government sector and stringent data safety protocols are necessary, text mining child protection intake reports in Australia is possible with the potential to unlock valuable insights.
- 11 November 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessProtocol for a Feasibility Study of Computational Text Analysis of Child Protection Intake Reports in South Australia (Preprint)
UNSTRUCTURED Background: Child maltreatment statistics collected using surveys, in-depth interviews, and manual case reviews are limited in scope due to resource constraints and labour-intensive, non-scalable approaches. The advent of computational advancements such as natural language processing, text mining, and large language models opens up the possibility of rapidly analysing vast numbers of documents from child protection client information systems to generate useful information. However, often due to strict data access protocols, few research efforts have focused on the application of computational text analysis in the child maltreatment area. Objective: This study presents a protocol for the application and evaluation of a text mining method on 10,000 child protection intake reports from South Australia’s Department for Child Protection. Methods: The protocol is split into five phases: 1) Ethics, governance, and data access approvals; 2A) Establish gold standard for the automatic extraction of information from 100 redacted reports using manual expert annotation; 2B) Contextual de-identification to preserve relationships; 3) Text mining method development, validation and evaluation based on 300 reports; 4) Text mining application on large sample; and 5) Evaluation and data analysis. Discussion: This protocol describes a feasibility study applying computational text analysis to highly sensitive child protection reports to automatically extract key information. Key considerations and challenges included governance, ethics, security, and technical requirements across multiple stakeholders involved in the study and adapting the text mining methodology to address the loss of contextual information caused by redaction. Conclusion: Whilst a high level of trust between academics and the government sector and stringent data safety protocols are necessary, text mining child protection intake reports in Australia is possible with the potential to unlock valuable insights.
- Open Access
- 3 November 2026
- Published by Frontiers Media SA in Frontiers in Digital Health
- Vol. 8, 2041010
Correction: De-identification of child protection intake reports using relationship-based replacement (DI-REBAR)
Correction on: Lukmanjaya W, Cox S, Perez-Concha O, Butler T, Bromfield L and Karystianis G (2026) De-identification of child protection intake reports using relationship-based replacement (DI-REBAR). Front. Digit. Health 8:1949307. doi: 10.3389/fdgth.2026.1949307 Wrong content There was a mistake in Table 1 as published. The column "Examples of redacted notification data" have identifying information examples (i.e., "Jane Doe", "John", and "John Doe Jr") bolded. The correct format is to have them highlighted in black to simulate redaction (i.e., "Jane Doe") or a redacted label as an alternative format (i.e., [REDACTED]). The original version of this article has been updated.
- 3 November 2026
- Published by Frontiers Media SA in Frontiers in Digital Health
- Vol. 8, 2041010
Open AccessCorrection: De-identification of child protection intake reports using relationship-based replacement (DI-REBAR)
Correction on: Lukmanjaya W, Cox S, Perez-Concha O, Butler T, Bromfield L and Karystianis G (2026) De-identification of child protection intake reports using relationship-based replacement (DI-REBAR). Front. Digit. Health 8:1949307. doi: 10.3389/fdgth.2026.1949307 Wrong content There was a mistake in Table 1 as published. The column "Examples of redacted notification data" have identifying information examples (i.e., "Jane Doe", "John", and "John Doe Jr") bolded. The correct format is to have them highlighted in black to simulate redaction (i.e., "Jane Doe") or a redacted label as an alternative format (i.e., [REDACTED]). The original version of this article has been updated.
- Open Access
- 4 May 2026
- Published by SAGE Publications in International Journal of Social Psychiatry
Early Mental Health Treatment and Domestic Violence Outcomes in Psychosis: A Population-Based Cohort Study
Background: Evidence for effective interventions among individuals with psychosis who perpetrate domestic violence (DV) remains limited. In this population, justice system involvement frequently occurs during periods of inadequately managed mental health symptoms. Aims: We examined whether early mental health treatment following a DV charge reduces subsequent DV reoffending among individuals with psychosis. Methods: We conducted a population-based retrospective cohort study of individuals in New South Wales, Australia, with a recorded diagnosis of psychosis who were subsequently charged with a DV offense. The primary exposure was receipt of mental health treatment within 1 month following the index DV charge. The primary outcome was time to the first subsequent DV charge. Secondary outcomes included time to first DV conviction, violent DV charge, and violent DV conviction. Time-to-event analyses were performed using competing risks regression. Results: During a median follow-up of 4 years, 3,513 of 8,458 individuals (41.5%) were charged with a subsequent DV offense. Among men, early mental health treatment was associated with reduced risk across all DV outcomes, including any DV charge (adjusted sub distribution hazard ratio [sHR] 0.86, 95% CI [0.79, 0.93]), conviction (sHR 0.81, 95% CI [0.73, 0.89]), violent charge (sHR 0.83, 95% CI [0.75, 0.93]), and violent conviction (sHR 0.78, 95% CI [0.68, 0.89]). There was no evidence of an association between early treatment and DV reoffending among women. Conclusions: These findings support integrating timely psychiatric care into DV offender management for men with psychosis, while indicating that alternative, sex-responsive strategies may be required for women with psychosis.
- 4 May 2026
- Published by SAGE Publications in International Journal of Social Psychiatry
Open AccessEarly Mental Health Treatment and Domestic Violence Outcomes in Psychosis: A Population-Based Cohort Study
Background: Evidence for effective interventions among individuals with psychosis who perpetrate domestic violence (DV) remains limited. In this population, justice system involvement frequently occurs during periods of inadequately managed mental health symptoms. Aims: We examined whether early mental health treatment following a DV charge reduces subsequent DV reoffending among individuals with psychosis. Methods: We conducted a population-based retrospective cohort study of individuals in New South Wales, Australia, with a recorded diagnosis of psychosis who were subsequently charged with a DV offense. The primary exposure was receipt of mental health treatment within 1 month following the index DV charge. The primary outcome was time to the first subsequent DV charge. Secondary outcomes included time to first DV conviction, violent DV charge, and violent DV conviction. Time-to-event analyses were performed using competing risks regression. Results: During a median follow-up of 4 years, 3,513 of 8,458 individuals (41.5%) were charged with a subsequent DV offense. Among men, early mental health treatment was associated with reduced risk across all DV outcomes, including any DV charge (adjusted sub distribution hazard ratio [sHR] 0.86, 95% CI [0.79, 0.93]), conviction (sHR 0.81, 95% CI [0.73, 0.89]), violent charge (sHR 0.83, 95% CI [0.75, 0.93]), and violent conviction (sHR 0.78, 95% CI [0.68, 0.89]). There was no evidence of an association between early treatment and DV reoffending among women. Conclusions: These findings support integrating timely psychiatric care into DV offender management for men with psychosis, while indicating that alternative, sex-responsive strategies may be required for women with psychosis.
- Open Access
- 20 February 2026
- Published by Frontiers Media SA in Frontiers in Psychiatry
- Vol. 16
Who you gonna call? Examining police reports involving people with psychosis to improve front line management
Background: In New South Wales (NSW), Australia, police respond to mental health related events every nine minutes, sometimes resulting in negative outcomes, including arrest, use of force and death. Methods: In this exploratory study, we manually analyzed 100 “patient” narratives (i.e., narratives with the involvement type for the suspected individual set by police as “patient”) to describe police encounters with individuals reported to have psychotic disorders in NSW from 2017 to 2021 to identify the reasons for police involvement, arrest circumstances, and outcomes. Results: Of 100 events, 59% involved males, 40% females, and one transgender patient. Most (60%) reported multiple mental illnesses, and 44% involved substance use and alcohol. Half (53%) occurred at the patient’s residence, followed by public places (17%). Police were called by the patient (25%), or a family member/partner (24%). Common reasons included welfare concerns (22%) and domestic disputes (18%). Only 11% resulted in arrest, while 71% resulted in transport to hospital/mental health unit. More than half (55%) resulted in the patient being scheduled under the Mental Health Act. Use of force was minimal, reported in 5% of events. Conclusions: Our preliminary findings challenge the stereotype of “dangerous” mentally ill individuals, with most police encounters involving welfare outcomes. Interactions with police such as those examined here present an opportunity for mental health services diversion, and cross-sectoral collaboration to address community care gaps. This study underscores the role police play in managing those experiencing mental health crises and the need for police training to effectively manage these situations.
- 20 February 2026
- Published by Frontiers Media SA in Frontiers in Psychiatry
- Vol. 16
Open AccessWho you gonna call? Examining police reports involving people with psychosis to improve front line management
Background: In New South Wales (NSW), Australia, police respond to mental health related events every nine minutes, sometimes resulting in negative outcomes, including arrest, use of force and death. Methods: In this exploratory study, we manually analyzed 100 “patient” narratives (i.e., narratives with the involvement type for the suspected individual set by police as “patient”) to describe police encounters with individuals reported to have psychotic disorders in NSW from 2017 to 2021 to identify the reasons for police involvement, arrest circumstances, and outcomes. Results: Of 100 events, 59% involved males, 40% females, and one transgender patient. Most (60%) reported multiple mental illnesses, and 44% involved substance use and alcohol. Half (53%) occurred at the patient’s residence, followed by public places (17%). Police were called by the patient (25%), or a family member/partner (24%). Common reasons included welfare concerns (22%) and domestic disputes (18%). Only 11% resulted in arrest, while 71% resulted in transport to hospital/mental health unit. More than half (55%) resulted in the patient being scheduled under the Mental Health Act. Use of force was minimal, reported in 5% of events. Conclusions: Our preliminary findings challenge the stereotype of “dangerous” mentally ill individuals, with most police encounters involving welfare outcomes. Interactions with police such as those examined here present an opportunity for mental health services diversion, and cross-sectoral collaboration to address community care gaps. This study underscores the role police play in managing those experiencing mental health crises and the need for police training to effectively manage these situations.
- Open Access
- 29 September 2025
- Published by JMIR Publications Inc. in JMIR Aging
- Vol. 8, e75993
Older Perpetrators of Domestic Violence: Mixed-Effects Logistic Regression Analysis of Police Records
Background: Domestic violence (DV) among older adults is an understudied area, often overlapping with abuse of older people, intimate partner violence, and behavioral and psychological symptoms of dementia. Objective: This study aimed to examine the characteristics of older persons of interest—individuals suspected or charged with a DV offence—and survivors involved in police-attended DV events in New South Wales, Australia, and assess associations with physical and nonphysical abuse. Methods: Police records of 10,708 DV events involving 8247 adults aged ≥55 years who were identified as persons of interest from 2005 to 2016 were analyzed using text mining. A 3-level mixed-effects logistic regression model was used to identify predictors of physical and nonphysical abuse. Results: Physical abuse formed a greater proportion of all abuse committed by female persons of interest aged >65 years compared to female persons of interest aged between 55 and 64 years and male persons of interest; however, after stratified analysis, female persons of interest had similarly elevated odds of physical abuse perpetration to male persons of interest. Other factors associated with increased odds of perpetrating physical abuse were persons of interest with dementia and alcohol-related events. Dementia increased the odds of combined physical and nonphysical abuse. Substance use disorders increased the odds of events with combined physical and nonphysical abuse. Conclusions: The findings of this study suggest that DV, including physical violence, is an important issue in later life. Alcohol is a situational factor, and dementia is associated with perpetration and exposure to violence. The study highlights the need for clinicians to evaluate the risk of violence and exposure to violence in patients with dementia and for policy interventions targeting alcohol and substance use in older adults.
- 29 September 2025
- Published by JMIR Publications Inc. in JMIR Aging
- Vol. 8, e75993
Open AccessOlder Perpetrators of Domestic Violence: Mixed-Effects Logistic Regression Analysis of Police Records
Background: Domestic violence (DV) among older adults is an understudied area, often overlapping with abuse of older people, intimate partner violence, and behavioral and psychological symptoms of dementia. Objective: This study aimed to examine the characteristics of older persons of interest—individuals suspected or charged with a DV offence—and survivors involved in police-attended DV events in New South Wales, Australia, and assess associations with physical and nonphysical abuse. Methods: Police records of 10,708 DV events involving 8247 adults aged ≥55 years who were identified as persons of interest from 2005 to 2016 were analyzed using text mining. A 3-level mixed-effects logistic regression model was used to identify predictors of physical and nonphysical abuse. Results: Physical abuse formed a greater proportion of all abuse committed by female persons of interest aged >65 years compared to female persons of interest aged between 55 and 64 years and male persons of interest; however, after stratified analysis, female persons of interest had similarly elevated odds of physical abuse perpetration to male persons of interest. Other factors associated with increased odds of perpetrating physical abuse were persons of interest with dementia and alcohol-related events. Dementia increased the odds of combined physical and nonphysical abuse. Substance use disorders increased the odds of events with combined physical and nonphysical abuse. Conclusions: The findings of this study suggest that DV, including physical violence, is an important issue in later life. Alcohol is a situational factor, and dementia is associated with perpetration and exposure to violence. The study highlights the need for clinicians to evaluate the risk of violence and exposure to violence in patients with dementia and for policy interventions targeting alcohol and substance use in older adults.
- Open Access
- 24 July 2025
- Published by JMIR Publications Inc. in JMIR Pediatrics and Parenting
- Vol. 8, e73579
Leveraging AI to Investigate Child Maltreatment Text Narratives: Promising Benefits and Addressable Risks
The trove of information contained in child maltreatment narratives represents an opportunity to strengthen the evidence base for policy reform in this area, yet it remains underutilized by researchers and policy makers. Current research into child maltreatment often involves the use of qualitative methodologies or structured survey data that are either too broad or not representative, thereby limiting the development of effective policy responses and intervention strategies. Artificial intelligence (AI) approaches such as large language models (AI models that understand and generate language) can analyze large volumes of child maltreatment narratives by extracting population-level insights on factors of interest such as mental health and treatment needs. However, when applying such methods, it is useful to have a framework on which to base approaches to the data. We propose a seven step framework: (1) data governance; (2) researcher vetting; (3) data deidentification; (4) data access; (5) feasibility testing of baseline methods; (6) large-scale implementation of black box algorithms; and (7) domain expert result validation for such exercises to ensure careful execution and limit the risk of privacy and security breaches, bias, and unreliable conclusions.
- 24 July 2025
- Published by JMIR Publications Inc. in JMIR Pediatrics and Parenting
- Vol. 8, e73579
Open AccessLeveraging AI to Investigate Child Maltreatment Text Narratives: Promising Benefits and Addressable Risks
The trove of information contained in child maltreatment narratives represents an opportunity to strengthen the evidence base for policy reform in this area, yet it remains underutilized by researchers and policy makers. Current research into child maltreatment often involves the use of qualitative methodologies or structured survey data that are either too broad or not representative, thereby limiting the development of effective policy responses and intervention strategies. Artificial intelligence (AI) approaches such as large language models (AI models that understand and generate language) can analyze large volumes of child maltreatment narratives by extracting population-level insights on factors of interest such as mental health and treatment needs. However, when applying such methods, it is useful to have a framework on which to base approaches to the data. We propose a seven step framework: (1) data governance; (2) researcher vetting; (3) data deidentification; (4) data access; (5) feasibility testing of baseline methods; (6) large-scale implementation of black box algorithms; and (7) domain expert result validation for such exercises to ensure careful execution and limit the risk of privacy and security breaches, bias, and unreliable conclusions.
- Open Access
- 14 April 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Older Perpetrators of Domestic Violence: Mixed-Effects Logistic Regression Analysis of Police Records (Preprint)
BACKGROUND Domestic violence (DV) among older adults is an understudied area, often overlapping with abuse of older people, intimate partner violence, and behavioral and psychological symptoms of dementia. OBJECTIVE This study aimed to examine the characteristics of older persons of interest—individuals suspected or charged with a DV offence—and survivors involved in police-attended DV events in New South Wales, Australia, and assess associations with physical and nonphysical abuse. METHODS Police records of 10,708 DV events involving 8247 adults aged ≥55 years who were identified as persons of interest from 2005 to 2016 were analyzed using text mining. A 3-level mixed-effects logistic regression model was used to identify predictors of physical and nonphysical abuse. RESULTS Physical abuse formed a greater proportion of all abuse committed by female persons of interest aged >65 years compared to female persons of interest aged between 55 and 64 years and male persons of interest; however, after stratified analysis, female persons of interest had similarly elevated odds of physical abuse perpetration to male persons of interest. Other factors associated with increased odds of perpetrating physical abuse were persons of interest with dementia and alcohol-related events. Dementia increased the odds of combined physical and nonphysical abuse. Substance use disorders increased the odds of events with combined physical and nonphysical abuse. CONCLUSIONS The findings of this study suggest that DV, including physical violence, is an important issue in later life. Alcohol is a situational factor, and dementia is associated with perpetration and exposure to violence. The study highlights the need for clinicians to evaluate the risk of violence and exposure to violence in patients with dementia and for policy interventions targeting alcohol and substance use in older adults. CLINICALTRIAL
- 14 April 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessOlder Perpetrators of Domestic Violence: Mixed-Effects Logistic Regression Analysis of Police Records (Preprint)
BACKGROUND Domestic violence (DV) among older adults is an understudied area, often overlapping with abuse of older people, intimate partner violence, and behavioral and psychological symptoms of dementia. OBJECTIVE This study aimed to examine the characteristics of older persons of interest—individuals suspected or charged with a DV offence—and survivors involved in police-attended DV events in New South Wales, Australia, and assess associations with physical and nonphysical abuse. METHODS Police records of 10,708 DV events involving 8247 adults aged ≥55 years who were identified as persons of interest from 2005 to 2016 were analyzed using text mining. A 3-level mixed-effects logistic regression model was used to identify predictors of physical and nonphysical abuse. RESULTS Physical abuse formed a greater proportion of all abuse committed by female persons of interest aged >65 years compared to female persons of interest aged between 55 and 64 years and male persons of interest; however, after stratified analysis, female persons of interest had similarly elevated odds of physical abuse perpetration to male persons of interest. Other factors associated with increased odds of perpetrating physical abuse were persons of interest with dementia and alcohol-related events. Dementia increased the odds of combined physical and nonphysical abuse. Substance use disorders increased the odds of events with combined physical and nonphysical abuse. CONCLUSIONS The findings of this study suggest that DV, including physical violence, is an important issue in later life. Alcohol is a situational factor, and dementia is associated with perpetration and exposure to violence. The study highlights the need for clinicians to evaluate the risk of violence and exposure to violence in patients with dementia and for policy interventions targeting alcohol and substance use in older adults. CLINICALTRIAL
- Open Access
- 6 March 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Leveraging AI to Investigate Child Maltreatment Text Narratives: Promising Benefits and Addressable Risks (Preprint)
UNSTRUCTURED The use of child maltreatment text narratives represents an opportunity to strengthen the evidence base for child policy reform, yet they remain underutilized. Current research into child maltreatment often involves the use of structured data, potentially overlooking key details within the narratives that can be crucial to assist in developing effective interventions. Artificial Intelligence approaches such as large language models can assist in the large-scale analysis of child maltreatment narratives extracting useful population-level insights on indicators of mental health, and treatment needs. However, when applying such methods, a framework of five key steps can assist in their careful implementation to reduce risks of privacy and bias.
- 6 March 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessLeveraging AI to Investigate Child Maltreatment Text Narratives: Promising Benefits and Addressable Risks (Preprint)
UNSTRUCTURED The use of child maltreatment text narratives represents an opportunity to strengthen the evidence base for child policy reform, yet they remain underutilized. Current research into child maltreatment often involves the use of structured data, potentially overlooking key details within the narratives that can be crucial to assist in developing effective interventions. Artificial Intelligence approaches such as large language models can assist in the large-scale analysis of child maltreatment narratives extracting useful population-level insights on indicators of mental health, and treatment needs. However, when applying such methods, a framework of five key steps can assist in their careful implementation to reduce risks of privacy and bias.
- Open Access
- 22 November 2024
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 8 (1) , e60878
Population Characteristics in Justice Health Research Based on PubMed Abstracts From 1963 to 2023: Text Mining Study
Background: The field of epidemiological criminology (or justice health research) has emerged in the past decade, studying the intersection between the public health and justice systems. To ensure research efforts are focused and equitable, it is important to reflect on the outputs in this area and address knowledge gaps. Objective: This study aimed to examine the characteristics of populations researched in a large sample of published outputs and identify research gaps and biases. Methods: A rule-based, text mining method was applied to 34,481 PubMed abstracts published from 1963 to 2023 to identify 4 population characteristics (sex, age, offender type, and nationality). Results: We evaluated our method in a random sample of 100 PubMed abstracts. Microprecision was 94.3%, with microrecall at 85.9% and micro–F1-score at 89.9% across the 4 characteristics. Half (n=17,039, 49.4%) of the 34,481 abstracts did not have any characteristic mentions and only 1.3% (n=443) reported sex, age, offender type, and nationality. From the 5170 (14.9%) abstracts that reported age, 3581 (69.3%) mentioned young people (younger than 18 years) and 3037 (58.7%) mentioned adults. Since 1990, studies reporting female-only populations increased, and in 2023, these accounted for almost half (105/216, 48.6%) of the research outputs, as opposed to 33.3% (72/216) for male-only populations. Nordic countries (Sweden, Norway, Finland, and Denmark) had the highest number of abstracts proportional to their incarcerated populations. Offenders with mental illness were the most common group of interest (840/4814, 17.4%), with an increase from 1990 onward. Conclusions: Research reporting on female populations increased, surpassing that involving male individuals, despite female individuals representing 5% of the incarcerated population; this suggests that male prisoners are underresearched. Although calls have been made for the justice health area to focus more on young people, our results showed that among the abstracts reporting age, most mentioned a population aged <18 years, reflecting a rise of youth involvement in the youth justice system. Those convicted of sex offenses and crimes relating to children were not as researched as the existing literature suggests, with a focus instead on populations with mental illness, whose rates rose steadily in the last 30 years. After adjusting for the size of the incarcerated population, Nordic countries have conducted proportionately the most research. Our findings highlight that despite the presence of several research reporting guidelines, justice health abstracts still do not adequately describe the investigated populations. Our study offers new insights in the field of justice health with implications for promoting diversity in the selection of research participants.
- 22 November 2024
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 8 (1) , e60878
Open AccessPopulation Characteristics in Justice Health Research Based on PubMed Abstracts From 1963 to 2023: Text Mining Study
Background: The field of epidemiological criminology (or justice health research) has emerged in the past decade, studying the intersection between the public health and justice systems. To ensure research efforts are focused and equitable, it is important to reflect on the outputs in this area and address knowledge gaps. Objective: This study aimed to examine the characteristics of populations researched in a large sample of published outputs and identify research gaps and biases. Methods: A rule-based, text mining method was applied to 34,481 PubMed abstracts published from 1963 to 2023 to identify 4 population characteristics (sex, age, offender type, and nationality). Results: We evaluated our method in a random sample of 100 PubMed abstracts. Microprecision was 94.3%, with microrecall at 85.9% and micro–F1-score at 89.9% across the 4 characteristics. Half (n=17,039, 49.4%) of the 34,481 abstracts did not have any characteristic mentions and only 1.3% (n=443) reported sex, age, offender type, and nationality. From the 5170 (14.9%) abstracts that reported age, 3581 (69.3%) mentioned young people (younger than 18 years) and 3037 (58.7%) mentioned adults. Since 1990, studies reporting female-only populations increased, and in 2023, these accounted for almost half (105/216, 48.6%) of the research outputs, as opposed to 33.3% (72/216) for male-only populations. Nordic countries (Sweden, Norway, Finland, and Denmark) had the highest number of abstracts proportional to their incarcerated populations. Offenders with mental illness were the most common group of interest (840/4814, 17.4%), with an increase from 1990 onward. Conclusions: Research reporting on female populations increased, surpassing that involving male individuals, despite female individuals representing 5% of the incarcerated population; this suggests that male prisoners are underresearched. Although calls have been made for the justice health area to focus more on young people, our results showed that among the abstracts reporting age, most mentioned a population aged <18 years, reflecting a rise of youth involvement in the youth justice system. Those convicted of sex offenses and crimes relating to children were not as researched as the existing literature suggests, with a focus instead on populations with mental illness, whose rates rose steadily in the last 30 years. After adjusting for the size of the incarcerated population, Nordic countries have conducted proportionately the most research. Our findings highlight that despite the presence of several research reporting guidelines, justice health abstracts still do not adequately describe the investigated populations. Our study offers new insights in the field of justice health with implications for promoting diversity in the selection of research participants.
- Open Access
- 5 October 2024
- Published by Springer Nature in Crime Science
- Vol. 13 (1) , 1-11
Characteristics of adult male victims in intimate heterosexual relationships from domestic violence police narratives
- 5 October 2024
- Published by Springer Nature in Crime Science
- Vol. 13 (1) , 1-11
Open AccessCharacteristics of adult male victims in intimate heterosexual relationships from domestic violence police narratives