Abstract
Eldercide by family caregivers is an underexplored but growing concern in ageing societies. Japan, the most rapidly ageing country in the world, offers an important case for examining this phenomenon. Drawing on 204 cases reported in a major Japanese national newspaper between 2000 and 2024, this study employed a mixed-method content analysis to construct a foundational understanding of eldercide by family caregivers in Japan. Quantitative findings indicate that these cases predominantly involve female victims and male offenders, both typically aged 65 and over. Qualitative analysis uncovered recurring contextual factors, including caregiver burnout, economic hardship, elder abuse, and strained familial relationships. The findings suggest that caregiver eldercide represents a distinct form of domestic homicide shaped by demographic ageing, cultural caregiving expectations, and gaps in institutional support. This study advances homicide scholarship by conceptualizing caregiver eldercide as a unique subtype of family homicide and highlights implications for theory, research, and prevention in ageing societies worldwide.
Keywords
Introduction
Ageing is a pressing contemporary issue with global implications for both criminology and victimology. According to WHO (2024), one in six people worldwide will be aged 60 years or older by 2030. Japan, in particular, is undergoing one of the most rapid demographic shifts toward an ageing population. The 2024 White Paper on the Ageing Society, published by the Cabinet Office, reports that individuals aged 65 and over constitute approximately 30 percent of the population, and that more than half of all households include at least one person in this age group (Cabinet Office, 2024). As the proportion of elderly individuals increases, so too does their vulnerability to various forms of victimization (Bezuidenhout & Booyens, 2018). This demographic trend has prompted growing scholarly interest in understanding how and why older adults become victims of crime (Kratcoski & Maximilian, 2018; Wylie & Brank, 2025).
Among the various forms of victimization, homicide represents the most extreme and violent outcome. Despite its severity, however, homicide involving elderly victims remains a significantly underexplored area within criminological and victimological research (Addington, 2022; Storey et al., 2024). In particular, familial homicide, where elderly individuals are killed by family members, has received limited scholarly attention, especially in the context of older adults receiving care (c.f., Arsuffi, 2010). As people age, many become increasingly dependent on others for daily care and support, most often provided by family members (Ogawa & Retherford, 1997). While such arrangements are often rooted in familial responsibility and affection, the intersection of age-related decline, dependency, and familial caregiving can place elderly individuals in precarious positions where the risk of abuse, neglect, or even lethal violence is elevated (Johannesen & Logiudice, 2013; Storey, 2020). This complexity underscores the urgent need for more nuanced research into the dynamics, risk factors, and prevention strategies surrounding familial homicide in ageing societies.
The Japanese context adds unique dimensions to the problem of eldercide. First, long-term care has historically been considered a family responsibility, influenced by Confucian traditions of filial piety (Fan, 2007). Although the Long-Term Care Insurance Act (2000) institutionalized public care support, reliance on family members, particularly spouses and adult children, remains deeply entrenched. Second, demographic transformations have eroded multi-generational living arrangements. Whereas nearly 20% of older adults co-resided with their children in 1980, the figure fell to below 8% by 2020 (Gender Equality Bureau Cabinet Office, 2022). Increasingly, elderly couples must care for each other,
This research aims to examine the characteristics of eldercide committed by family caregivers in Japan. To achieve this, the study analyzed cases of eldercide reported in one Japanese national newspaper between 2000 and 2024. Employing a mixed-method content analysis, the study provides both a broad overview of eldercide by family caregivers and an in-depth exploration of the circumstances and contexts in which these homicides occur. By identifying recurring patterns, motivations, and situational factors, this study makes three contributions: (1) it provides the most comprehensive portrait to date of caregiver eldercide in Japan; (2) it situates these findings within broader criminological debates on domestic homicide, homicide-suicide, and family caregiving; and (3) it highlights how demographic ageing, cultural expectations of family care, and institutional gaps interact to create lethal risks. Due to the rapid ageing of the population, the number of older people in need of care is projected to increase continuously (Yuhara, 2018). Without effective prevention strategies, this demographic shift may lead to an increase in eldercide by family caregivers. It is therefore crucial to understand the current landscape of such cases in Japan in order to inform appropriate interventions and safeguard the well-being of older adults.
At the outset, it is important to define the scope of this study. First, although there is no universally agreed-upon definition of eldercide, particularly with regard to the victim's age (Addington, 2022; Riedel, 2013), this study defines eldercide as homicide in which the victim is aged 65 or older. This threshold reflects the official designation of elderly in Japan (Suzuki & Otani, 2024), the country from which the data for this research were drawn. Second, although there is currently no standard definition of eldercide by family caregivers in Japan (Miyamoto, 2016; Yuhara, 2018), this study adopts a broad interpretation of the term family caregiver. Specifically, it refers to a family member who holds primary responsibility for the care of the victim and who ultimately causes the victim's death. This working definition acknowledges the complexity of caregiving relationships while allowing for the identification and analysis of relevant cases within the dataset.
Eldercide in Homicide Scholarship
Although older adults account for a relatively small proportion of homicide victims worldwide (Coelho et al., 2010; Feldmeyer & Steffensmeier, 2012), this pattern must be understood in a demographic context. According to the United Nations Office on Drugs and Crime (2023), the global homicide rate was approximately 5.8 per 100,000 population in the most recent reporting period, with the vast majority of victims being young men aged 15 to 29 years. Japan represents a distinctive case in this regard. Despite having one of the lowest homicide rates globally (∼0.2 per 100,000 population) (UNODC, 2025), homicides involving older victims account for a substantial share of recorded cases. Official statistics indicate that victims aged 65 and over constitute approximately one quarter of all homicide victims in Japan (Japanese Ministry of Justice, 2024).
Although the absolute number of eldercide cases has been increasing in parallel with the global ageing population (Karbeyaz et al., 2018), existing literature presents a mixed picture regarding the demographic characteristics of eldercide victims, particularly in terms of gender (Buschmann et al., 2016; Krienert & Walsh, 2010) and race/ethnicity (Abrams et al., 2007; Titterington & Reyes, 2010). However, a recent systematic review suggests that older homicide victims are more likely to be White females, although the authors caution against drawing strong conclusions about racial and ethnic patterns due to the challenges of comparing such diversity across global contexts (Rogers & Storey, 2019).
The relationship between victim and offender in cases of eldercide varies considerably; however, the majority of incidents involve a perpetrator known to the victim, most commonly a family member or acquaintance (Falzon & Davis, 1998; Rogers & Storey, 2019). Consequently, eldercide victims are more likely to be killed in their own homes (Abrams et al., 2007; Bows, 2019). In contrast, when the offender is a stranger, eldercide is more frequently associated with robbery or the commission of another felony (Block, 2013; Titterington & Reyes, 2010).
The routine lifestyles of older adults may help explain key risk factors associated with eldercide committed by strangers. Some research suggests that decreased mobility among older individuals contributes to their overall lower rates of homicide victimization (Nelsen & Huff-Corzine, 1998). However, other studies indicate that certain lifestyle characteristics, such as living alone, physical disability, and reduced social or physical activity, can increase the risk of eldercide, particularly in felony-related cases that occur within private settings (Fox, 1991; Roberts & Willits, 2013).
Although limited, the literature reveals a distinctive pattern of eldercide. While some older adults may be more likely to be killed by strangers in the context of a felony, others appear to be at heightened risk of being killed within their own homes by someone known to them. Within this context, caregiving relationships can emerge as a potentially significant factor deserving closer examination. As the global population continues to age and more individuals rely on family members for care, it becomes increasingly important to explore how these dynamics may contribute to lethal outcomes.
Research on Eldercide by Family Caregivers
Eldercide by family caregivers constitutes a unique and concerning subset of homicides involving older adults. Distinct from stranger-perpetrated eldercide, these incidents frequently occur within domestic settings and often involve complex interpersonal and caregiving dynamics. Although the international literature on this topic remains limited, existing research suggests that victims in these cases are disproportionately non-Hispanic White women aged 80 and above (Karch & Nunn, 2011). Several risk factors have been identified in relation to eldercide by family caregivers, including mental illness, substance abuse, and financial issues (Benbow et al., 2019). Moreover, research indicates that male family caregivers may be particularly vulnerable to perpetrating such acts, as they often face challenges in providing adequate care and are less likely to seek assistance from formal support services (Salari, 2007).
In Japan, where rapid population ageing is underway, only a limited number of studies have examined eldercide committed by family caregivers. Existing research suggests that these incidents frequently involve both the victim and the perpetrator being older adults aged 70 and above (Yuasa, 2023; Yuhara, 2016). In terms of gender and the victim–offender relationship, perpetrators are more likely to be male, and victims tend to be female, with the two typically sharing a close familial bond (Miyamoto, 2016; Yuhara, 2016). Victims often suffer from physical conditions such as dementia or are bedridden (Ohrui et al., 2005), which contributes to caregiver stress and burnout, factors identified as primary drivers of eldercide within families (Yuhara, 2018). These challenges are further compounded when caregivers themselves experience physical decline or illness (Nakao, 2017), and particularly when male caregivers attempt to manage all care responsibilities alone, often beyond their personal capacity (Hane, 2006).
The limited available evidence suggests that eldercide by family caregivers may be shaped by a constellation of interpersonal, health-related, and structural stressors, which is often rooted in the unique dynamics of caregiving within aging households. While some risk factors appear consistent across contexts, such as caregiver stress and gendered caregiving roles, the phenomenon in Japan may present distinct characteristics due to its rapidly aging society, cultural expectations around family care, and limited institutional support. Yet, despite its growing relevance, this issue remains underexplored in the Japanese context.
Current Study
This study seeks to advance this underexplored area by extending existing research on eldercide by family caregivers in Japan in several important ways. First, while most previous studies focused on cases before 2015 (e.g., Hane, 2006; Miyamoto, 2016; Yuhara, 2016), this research incorporates the most recent data available to provide an updated understanding of the issue (c.f., Yuasa, 2023). Second, given that many existing studies cover relatively short timeframes—typically less than five years (Miyamoto, 2022; Miyamoto & Mitsuhashi, 2012; Yuasa, 2023)—this study examines more than two decades of eldercide cases involving family caregivers to identify broader trends and characteristics. Third, whereas some studies do not clearly differentiate between general homicides by family members and eldercide by family caregivers (Miyamoto, 2016; Miyamoto & Mitsuhashi, 2012; Yuasa, 2023), this study focuses exclusively on the latter. By doing so, it aims to provide the most comprehensive and up-to-date profile of eldercide by family caregivers in Japan. Finally, and relatedly, although eldercide perpetrated by family caregivers has attracted increasing scholarly attention in Japan, no national-level datasets currently exist that systematically capture the prevalence of such cases. Official police and court statistics do not classify homicides by caregiving relationship, nor do they distinguish eldercide from other forms of family homicide. Consequently, the national prevalence of eldercide committed by family caregivers remains unknown. Previous Japanese studies have therefore relied primarily on newspaper-based case identification to document and analyze this phenomenon (e.g., Miyamoto, 2022; Yuasa, 2023). The present study adopts this established approach and extends prior research by examining cases over a longer time period, thereby offering the most comprehensive empirical portrait of caregiver-perpetrated eldercide in Japan to date. This knowledge is essential for developing targeted prevention strategies and offers a foundation for meaningful international comparisons.
Methodology
Research Design
This research employed a mixed-method approach. This decision was guided by the study's dual aim: to provide both a broad overview of eldercide committed by family caregivers and a detailed exploration of the circumstances and contexts in which these homicides occur. Achieving this objective necessitated the integration of both quantitative and qualitative analyses, as recommended for studies that seek to capture both patterns and meanings in complex social phenomena (Creswell, 2014).
Data Source
This study utilized newspaper reports as the primary data source to investigate the characteristics of eldercide committed by family caregivers in Japan. According to Parkin and Gruenewald (2017), newspapers are a reliable form of open-source data and often align closely with official homicide statistics. Indeed, a number of previous studies on homicide have successfully relied on newspaper articles to gather detailed information about cases, particularly when official data are limited or lack contextual detail (e.g., Colasanti et al., 2024; Salari & Sillito, 2016). Newspaper data thus offer a valuable resource for examining both the quantitative trends and qualitative nuances of eldercide cases.
Among the various newspapers available, Asahi Shimbun was selected as the data source for this study. It is one of Japan's four major national daily newspapers and is widely regarded for its extensive coverage and reliability (Miyazawa, 2008). Moreover, Asahi Shimbun has been used as a data source in previous research on eldercide committed by family caregivers in Japan, further supporting its suitability for this study (Hane, 2006; Miyamoto, 2022).
A search was conducted in the Asahi Shimbun newspaper database for the period between 2000 and 2024. The year 2000 was selected as the starting point because it marks the introduction of the Long-Term Care Insurance Act in Japan, a policy shift that significantly influenced caregiving practices. Following Miyamoto (2022) and Yuhara (2016), the keyword “nursing care” (介護) was combined with terms referring to potential criminal charges relevant to eldercide by family caregivers, including “homicide 1 ” (殺人), “injury causing death” (傷害致死), “abandonment by a person responsible for protection” (保護責任者遺棄), and “homicide-suicide” (心中).
This study employed the following eligibility criteria. As defined earlier, eldercide by family caregivers refers to a homicide committed by a family member, regardless of their age or gender, who holds primary responsibility for the care of a victim aged 65 or older. The study focused exclusively on newspaper articles that formed part of the Asahi Shimbun's nationwide coverage and that reported cases involving the arrest of a suspect in connection with eldercide by a family caregiver. Cases of attempted homicide were excluded from the analysis in order to maintain a consistent focus on completed acts of lethal violence.
The database search, conducted by the second author in May 2025, initially yielded 1,458 newspaper articles. A preliminary screening by the second author reduced this number to 377 articles. The first author then conducted a secondary screening, which led to the exclusion of 176 articles based on the following criteria: duplicates (n = 25), lack of relevance to arrests (n = 118), absence of nationwide coverage (n = 6), involvement of attempted homicide rather than completed acts (n = 20), victims aged 64 or younger (n = 4), and insufficient information to assess the case (n = 3). After applying these criteria, a total of 201 articles were retained for analysis. Three of these articles each reported two distinct cases, resulting in a final dataset of 204 eldercide cases committed by family caregivers.
Analytical Strategies
This research employed a content analysis. For the quantitative component, a descriptive analysis was conducted to identify the characteristics of eldercide committed by family caregivers in Japan. As Addington (2022) notes, given the limited body of research on homicides involving older adults, descriptive studies represent a crucial first step in informing future scholarship. Indeed, numerous studies on homicide have relied solely on descriptive statistics and have successfully uncovered valuable patterns and trends (e.g., Mahesar et al., 2024; Oya et al., 2024). In addition to descriptive analysis, Chi-squared tests were conducted, where appropriate and feasible, to examine statistically significant relationships between key variables.
To conduct the quantitative analysis, the identified newspaper articles were coded by the first author. The following variables were extracted: year of the incident, victim's age, victim's gender, offender's age, offender's gender, victim–offender relationship, motive, method of killing, time of the incident, presence of dementia in the victim, presence of other diseases in the victim, whether the victim was bedridden, whether the victim had been formally certified as requiring long-term nursing care, 2 presence of disease in the offender, evidence of prior abuse by the offender, whether the case involved a murder-suicide, and the word count of the article. All quantitative coding was conducted by the first author. As the study employed a single-coder design, formal inter-rater reliability statistics (e.g., Cohen's kappa) could not be calculated (c.f., Solinas-Saunders, 2024). To enhance analytic rigor and replicability, several safeguards were implemented. A detailed codebook was developed prior to full coding, specifying variable definitions, inclusion and exclusion criteria, and decision rules for each coded category. Coding proceeded iteratively, with initial coding of approximately 20% of cases used to refine category definitions and resolve ambiguities. Ambiguous cases were documented and resolved through consistent decision rules applied across the dataset. For key variables subject to interpretive ambiguity (e.g., caregiver burnout), coding decisions were based on explicit textual indicators within the newspaper reports. For example, caregiver burnout was coded only when articles explicitly referenced prolonged caregiving strain, exhaustion, or an inability to continue care, rather than being inferred from the victim's health status alone. When reports contained conflicting or insufficient information, cases were coded as “unknown” rather than assigned to a substantive category. These rules were applied consistently across all cases. Throughout the coding process, the first author maintained reflexive memos documenting analytic decisions and uncertainties, which were periodically reviewed by the co-authors to ensure consistency over time.
The qualitative component of this research took the form of a case study. Drawing on the patterns and characteristics identified through the quantitative analysis, this phase aimed to provide an in-depth exploration of the circumstances and contextual factors surrounding cases of eldercide committed by family caregivers. By examining selected cases in greater detail, the study sought to uncover the complex social, relational, and situational dynamics that contribute to these homicides. The original newspaper articles were in Japanese, and the first author translated them into English. The translations were then checked for accuracy through a member-checking process.
Preliminary analysis revealed that most newspaper articles provided only brief accounts of the incidents, and that the Japanese word count varied substantially across articles (maximum = 2,356; minimum = 140). To select cases for in-depth qualitative analysis, this study followed the procedure outlined by Wong and Lee (2021). The average word count was calculated (Mean = 424.31), which corresponds to approximately 212 words in English. The articles were then dichotomized into two categories: “high word count” (424 words and above) and “low word count” (under 424 words). Cases were selected from the high word count category (n = 48, 23.5%), as these articles typically contained richer contextual information necessary for qualitative exploration.
Specifically, from the subset of high word count articles (n = 48), six cases were selected for in-depth qualitative analysis using purposive sampling. Case selection was guided by two analytic objectives: (1) to illustrate configurations of caregiver eldercide that closely reflected dominant patterns identified in the quantitative analysis (e.g., spousal caregiving, caregiver burnout, old-old couples), and (2) to capture analytically distinct pathways that deviated from these dominant patterns (e.g., prior abuse, economic strain, or care refusal). The selected cases were chosen because they provided particularly clear, detailed, and information-rich accounts of these analytically relevant configurations, rather than to represent the full range or frequency of cases.
Ethical Considerations
As this research utilized publicly available data from newspaper sources, formal ethics approval was not required. However, given Japan's relatively low homicide rate (Ellis & Hamai, 2017), there is a risk that individuals involved in cases of eldercide by family caregivers could be identified. To mitigate this risk and uphold ethical research standards, all potentially identifiable information has been excluded from this paper.
Results
Characteristics of Eldercide by Family Caregivers in Japan
Table 1 presents the characteristics of eldercide by family caregivers in Japan, identified through an analysis of cases reported in the Asahi Shimbun database between 2000 and 2024. 3 The following section provides a descriptive summary of these characteristics.
Profile of Eldercide by Caregivers in Japan (N = 204).
Note: An “Unknown/Unreported” category is included where applicable to explicitly indicate cases in which relevant information was not available.
Age
To distinguish between early-stage and late-stage elders (Cabinet Office, 2024), the age variable for victims was dichotomized at 75 years. Similarly, to differentiate eldercide committed by older versus younger offenders, the offender's age was dichotomized at 65 years. Based on this categorization, the majority of victims (69.1%) were aged 75 and above, while over half of the offenders (58.8%) were aged 65 and older. A cross-tabulation of victim and offender age groups revealed a statistically significant association between the elderliness of the victim and that of the offender (χ2 (1, N = 202) = 8.112, p = .005). While statistically significant, this association should be interpreted with caution given the descriptive nature of the data and the absence of controls for potential confounding factors.
Sex
A notable pattern emerged in relation to the sex of victims and offenders. The majority of victims were female (69.6%), whereas the majority of offenders were male (57.8%). Cross-tabulation of victim–offender sex composition revealed a statistically significant difference. The most common configuration was female victim and male offender, accounting for 50.1 percent of all cases (χ2 (1, N = 173) = 59.001, p = .001).
Victim–Offender Relationship
The victim–offender relationship varied across cases; however, the most common relationship was that of married or de facto couples, accounting for over half of the incidents (53.9%). The second most frequent relationship involved offenders killing their biological parents, comprising 36.8% of cases.
Motives
The most commonly reported motive in the newspaper articles was caregiver burnout (70.6%). Given that this research specifically focuses on eldercide by family caregivers, caregiver burnout understandably emerged as a prominent factor. Indeed, except for unknown motives (18.6%), the second most common motive, heated arguments, accounted for only 7.4 percent.
Methods of Killing
The methods of killing showed some variation. More than half of the eldercide victims killed by family caregivers died as a result of strangulation (54.9%). This was followed by bludgeoning and stabbing, each accounting for 10.3% of cases. Other methods included suffocation or drowning (8.3%), burning (6.4%), neglect (3.9%), and poisoning (2.5%).
Time for Eldercide
In many cases (43.1%), the time of the eldercide by family caregivers was unknown. Among those with known times, incidents occurred at various points throughout the day: morning (17.6%), afternoon (10.8%), night (13.7%), and midnight (14.7%).
Vulnerabilities
Vulnerabilities of victims and offenders were reported in some cases. Here, vulnerabilities were defined as physical conditions that contribute to an increased burden of nursing care or that may exacerbate the caregiving environment. However, it is important to note that newspaper articles often provide limited detail, and such vulnerabilities may be underreported due to space constraints or selective reporting. Based on the available information, most cases did not indicate the presence of these vulnerabilities. Specifically, the majority of victims were reported as not having dementia (71.6%), not having any other disease (66.7%), not being bedridden (76.0%), and not being certified for long-term care needs (94.1%). Similarly, the vast majority of offenders were reported to have no disease (93.1%).
Elder Abuse
Elder abuse against the victim was not reported in the vast majority of cases (96.6%).
Murder-Suicide
Murder-suicide cases in eldercide by family caregivers are relatively complex. While more than half of the cases (61.3%) did not involve murder-suicide, the remainder did. Specifically, 21.1 percent of cases resulted in completed murder-suicides, whereas 17.6 percent involved attempted but unsuccessful murder-suicides.
Case Study
While the preceding section primarily presented descriptive statistics, several key characteristics of eldercide by family caregivers in Japan emerged. These cases predominantly involved female victims and male offenders aged 65 and over, with caregiver burnout frequently identified as a contributing factor. Strangulation was the most commonly reported method of killing. Building on this quantitative overview, the current section offers a more nuanced exploration of the circumstances and contexts surrounding these homicides. Specifically, it examines both high-risk pathway and analytically distinct cases to elucidate the distinctions between them (Yin, 2009). Cases selected for detailed analysis were those with a “high word count” in newspaper reports (424 words and above) to ensure richer contextual information for in-depth examination.
The selection of qualitative cases based on article word count also warrants further reflection. While this approach facilitated access to more detailed narratives, it likely introduced a bias toward incidents framed as particularly complex, dramatic, or medically salient, which are more likely to receive extended media coverage. Less visible caregiving dynamics or cases lacking overt health-related markers may therefore be underrepresented in the qualitative analysis. This pattern reflects broader processes of newsworthiness and editorial selection rather than the prevalence or typicality of such cases. Accordingly, the qualitative findings should be interpreted as illuminating specific pathways to eldercide that attract detailed media attention, rather than as representative of all caregiving-related homicides.
The six cases presented below were selected from the 48 high word count articles not to indicate prevalence, but to exemplify analytically salient patterns and contrasts identified in the broader dataset. It is important to further clarify the analytic purpose of the qualitative case studies presented below. These cases are not intended to be statistically representative of the broader quantitative sample. Rather, they are used to illustrate specific high-risk pathways to eldercide that emerge under conditions of intensive caregiving strain. Although the quantitative analysis indicates that dementia, serious disease, and formal long-term care needs are not reported in the majority of cases, such health-related vulnerabilities may be underreported in newspaper coverage. The selected cases therefore serve to illuminate how caregiving stress can escalate to lethal outcomes under particular circumstances, rather than to suggest that dementia-driven caregiving stress typifies most cases of eldercide by family caregivers.
High-Risk Pathway Cases
This section presents illustrative cases that highlight the tragic consequences arising from the intersection of caregiver exhaustion, social isolation, and systemic shortcomings in eldercare within Japan's aging society. Through detailed examination of three high-risk pathway incidents of eldercide committed by spousal caregivers, these cases reveal recurring themes such as overwhelming caregiving burdens, lack of formal support utilization, and the invisibility of caregiving distress within existing welfare and monitoring frameworks. Each case underscores how the cumulative pressures faced by elderly caregivers, often isolated and physically or emotionally strained, can culminate in fatal outcomes.
Case 247
4
starkly illustrates the intersection of caregiver exhaustion, social isolation, and systemic gaps in eldercare within Japan's rapidly aging society: In 2005, a 67-year-old man was arrested for killing his 74-year-old wife, who had been suffering from Alzheimer's disease and a stroke. The man, exhausted from being her sole caregiver, confessed to police that he had thought about killing her before and finally did so using a necktie. The couple had no children and had not sought help from authorities or relatives. Neighbors described the man as devoted and never complaining, though his wife's condition had worsened over the past year. After the incident, he attempted suicide by trying to jump from their apartment building but failed, later calling his brother-in-law and reporting himself to the police. Authorities believe he was mentally overwhelmed and desperate.
Case 804 involves an elderly former police officer who killed his bedridden wife, a former nurse with advanced dementia, after years of intensive caregiving under deteriorating physical and emotional conditions. Despite receiving some home care support since 2007, the husband, himself frail and suffering from poor vision, hearing loss, and limited mobility, continued to shoulder most of the caregiving burden alone. The wife had reached the highest level of care dependency (Category 5), requiring constant assistance with all daily functions. While neighbors recalled the couple's once close and affectionate relationship, the husband had begun to express signs of severe caregiver fatigue. On the night of the incident in 2013, he strangled and cut his wife's neck, later found with self-inflicted wounds, suggesting a failed suicide attempt.
Case 966 also illustrates the tragic consequences of prolonged caregiving stress, where a devoted family caregiver, overwhelmed by the demands of caring for a spouse with advanced dementia, ultimately resorted to fatal violence. In 2015, a 71-year-old man was arrested for the murder of his 69-year-old wife, who suffered from advanced dementia and had been bedridden since a brain illness in 2004. Over more than a decade, he provided devoted care for her, managing all daily needs despite the availability of formal support services such as home bathing care and day programs. However, he did not seek assistance from local welfare agencies. The wife's dementia progression led to increased verbal aggression toward him, which the husband cited as a contributing factor to the fatal act of strangulation. Neighbors described the husband as diligent but overwhelmed by the caregiving burden.
Analytically Distinct Cases
This section examines three analytically distinct cases of eldercide arising within familial caregiving contexts, each illustrating distinct but interrelated factors that contribute to fatal outcomes. The cases presented in this section are described as analytically distinct because they illustrate pathways to eldercide that diverge from the dominant patterns identified in the quantitative analysis. While cases with similar configurations do occur elsewhere in the dataset, they appear less frequently and involve different combinations of caregiving stressors than those identified as “typical.” These cases highlight critical issues such as the escalation of abuse amid caregiver stress and substance dependence, the impact of economic hardship on caregiving dynamics, and the consequences of family conflict and care refusal. Collectively, they reveal how social isolation, inadequate support engagement, financial strain, and relational tensions can intersect to overwhelm caregivers, sometimes culminating in violence and death.
Case 344 is the tragic consequence of elder abuse within an isolated caregiver relationship, compounded by substance use and limited social support. In 2006, a 77-year-old woman with Parkinson's disease died from injuries and exhaustion caused by prolonged physical abuse inflicted by her 45-year-old son, who was unemployed and had a history of heavy drinking. Although initially a devoted caregiver who personally transported his mother to medical appointments and managed household expenses frugally, the son's frustration with his mother's declining health and impaired mobility escalated into verbal and physical violence starting around July. The abuse included episodes of being locked out of the home, intimidation, and escalating to severe injuries such as a fractured jaw and brain contusion, which eventually left the mother bedridden. Despite the existence of elder abuse prevention laws enabling municipal intervention, the family's social isolation, lack of involvement with care services, and minimal neighborhood interaction prevented early detection by authorities.
Economic hardship can also be a significant factor contributing to eldercide by family caregivers, as demonstrated in Case 748. A 77-year-old man was arrested for murdering his disabled younger brother. The victim required extensive care due to physical disabilities and was classified as needing level 3 care, relying on a pension and additional financial support from the suspect for daily necessities and medical expenses. In 2010, after withdrawing the brother from a care facility, the suspect reportedly pushed him into a river, leading to his death. The suspect cited economic burden as a motive, highlighting the strain that caregiving costs can impose on family members with limited income. Welfare officials noted that financial assistance and social support mechanisms were available but not accessed, suggesting missed opportunities for intervention.
Case 1433 presents the complex dynamics of eldercide within family caregiving contexts, where care-related conflicts can escalate to fatal violence. In 2024, a 94-year-old woman with care needs died from multiple traumatic injuries inflicted by her eldest son, a 58-year-old practicing lawyer. The victim required level 2 care due to weakened mobility, and the son reportedly assaulted her using headbutts and punches after repeated disputes over his attempts to encourage her to accept formal caregiving services. According to the police, the son admitted to the violence amid communication breakdowns.
Discussion
Despite Japan's rapidly ageing population, the characteristics of eldercide committed by family caregivers remain relatively underexplored. This study aimed to address this gap by analyzing cases reported in a major Japanese national newspaper between 2000 and 2024. Employing a mixed-method content analysis, the research provided both a quantitative overview of eldercide by family caregivers and a qualitative exploration of the circumstances and contexts in which these homicides occur. In doing so, it offers the most comprehensive and up-to-date profile of eldercide by family caregivers in Japan to date. The findings reveal a set of recurring conditions and patterns that are critical to understanding the dynamics of these cases, thereby contributing valuable insights for future research and the development of targeted prevention strategies.
Although this study conceptualizes caregiver eldercide as a distinct subtype of family homicide defined by the presence of primary caregiving responsibility, the findings indicate that this category is internally heterogeneous. The quantitative and qualitative analyses identify multiple pathways through which caregiving contexts can culminate in lethal violence, including caregiver burnout and despair, escalation of prior abuse, economic strain, and conflict surrounding care refusal. Rather than treating these patterns as separate forms of homicide, this study conceptualizes caregiver eldercide as a single analytic category encompassing diverse but related pathways shaped by caregiving dynamics. This framework recognizes variation within caregiver eldercide while maintaining analytic coherence around the central role of caregiving responsibility. Yet, the theoretical implications discussed below should be interpreted in light of the study's exploratory design and reliance on newspaper-based data, which allow for the identification of patterned associations and plausible pathways but do not permit definitive causal claims.
According to the quantitative findings, one of the key conditions underlying eldercide by family caregivers in Japan is the age dynamics between victims and offenders. Notably, half of the offenders were aged 65 and above. Consistent with existing studies (Yuasa, 2023; Yuhara, 2016), this finding suggests that eldercide by family caregivers is particularly likely to occur among “old-old” couples. The prominence of old-old couples highlights an emergent form of domestic homicide specific to ageing societies. While intimate partner homicide has been extensively studied, the Japanese cases illustrate how age-related decline and caregiving strain create distinct pathways to lethal violence. Unlike younger intimate partner homicides, these killings are less likely to stem from jealousy or control and more likely from exhaustion, dependency, and despair (Salari & Sillito, 2016). This finding suggests the need to conceptualize “caregiving homicide” as a category within family homicide scholarship. In the Japanese context, this pattern may reflect broader demographic and societal changes in family structures. Historically, older adults in Japan commonly lived with their children or grandchildren and received care primarily from daughters or daughters-in-law. However, with ongoing urbanization and economic development, this traditional arrangement has declined significantly (Gender Equality Bureau Cabinet Office, 2022). Consequently, more elderly couples are left to care for one another, despite the increased likelihood of reduced mobility, chronic illness, and caregiving fatigue associated with advanced age. These shifts in familial caregiving arrangements may be a critical factor contributing to the emergence of eldercide among “old-old” households.
Another key condition identified in this study is the gender dynamic within old-old caregiving couples. Consistent with previous findings (Miyamoto, 2016; Yuhara, 2016), the quantitative results of this research indicate that eldercide by family caregivers is more likely to occur when the victim is female and the caregiver is male. This gendered pattern is largely reflective of demographic realities in Japan, particularly differences in life expectancy and healthy life expectancy. According to the Gender Equality Bureau Cabinet Office (2022), women in Japan have a longer life expectancy (87.71 years) than men (81.56 years). As a result, men are more likely to become caregivers to their aging spouses in later life, often at a time when they themselves are experiencing age-related health issues. Male caregivers in such contexts often face significant burdens (Alvarez-Roldan & Bravo-González, 2025) and may lack the physical, emotional, or psychological capacity to provide sustained care. Compounding this challenge is the documented reluctance among male caregivers to seek external assistance or utilize formal support services (Cahill, 2000). This hesitation may stem from traditional gender norms, a desire to fulfill familial roles independently, or a lack of awareness about available resources (Hirayama, 2017). When coupled with the physical and emotional toll of caregiving in later life, such resistance to seeking help may serve as a critical risk factor for eldercide by family caregivers. Understanding these gendered caregiving dynamics is essential for developing targeted interventions aimed at reducing caregiver burden and preventing lethal outcomes.
Third, the findings contribute to refining existing theories of homicide-suicide. Roughly 40% of cases involved completed or attempted suicide by the offender. Unlike younger homicide-suicide cases, which often involve control or intimate partner violence (Colasanti et al., 2024), these appear motivated by despair, fear of institutionalization, or a wish to end suffering. Rather than redefining homicide-suicide typologies, the findings suggest that caregiving contexts in later life may shape distinct motivational logics that warrant further theoretical and empirical attention. This calls for theoretical frameworks that capture how life-course dynamics, particularly ageing and caregiving, can transform the meaning of homicide-suicide.
Finally, although these trends were not statistically evident in the quantitative analysis, the qualitative findings revealed several important contributing factors that may underlie eldercide by family caregivers. These factors include prior elder abuse, economic hardship, and strained familial relationships. While each of these conditions may exacerbate caregiving stress, it is important to emphasize that they are unlikely to operate in isolation. Rather, they appear to interact with, and be intensified by, the prolonged demands of long-term elder care. In this context, caregiver burden may reach a critical threshold, particularly when compounded by unresolved familial conflict or financial instability. These findings suggest that eldercide by family caregivers should be understood not merely as a consequence of individual pathology, but as a complex outcome of intersecting interpersonal, structural, and caregiving-related pressures (Suzuki et al., 2026). Taken together, these pathways underscore that caregiver eldercide is best understood not as a uniform phenomenon, but as a heterogeneous category united by caregiving responsibility rather than by a single motive or mechanism.
Implications
This study highlights the urgent need for more nuanced research into eldercide by family caregivers, particularly in ageing societies like Japan. Longitudinal studies, for instance, could provide deeper insights into how caregiving burden evolves over time and under what conditions it may escalate into violence. Likewise, comparative research across different caregiving arrangements (e.g., spouse vs. child caregivers, or co-residential vs. non-co-residential settings) could help disentangle the role of relational dynamics, caregiving intensity, and social isolation.
From a prevention standpoint, the study's findings point to the need for targeted, gender-sensitive interventions tailored to the realities of old-old caregiving households (Miyamoto & Ishiyama, 2024). Given the high incidence of male caregivers committing eldercide against female spouses, there is a clear imperative to design outreach and support programs that specifically engage older men. Such programs should address not only the practical challenges of caregiving, such as bathing or feeding, but also the psychological strain and social stigma that may deter men from seeking help.
In addition, prevention strategies must account for structural vulnerabilities, such as economic hardship and access to formal care services. Policymakers should prioritize the expansion of respite care options, caregiver allowances, and early intervention programs aimed at identifying households at risk. Routine screening for caregiver stress and elder abuse by health and social care professionals could serve as a vital point of early detection and support. Importantly, interventions should be holistic and multifaceted with a recognition that eldercide often results from a convergence of caregiving fatigue, interpersonal conflict, and social disconnection. As such, the development of multidisciplinary risk assessment tools that consider both individual and contextual factors could enhance the capacity of professionals to identify at-risk families and intervene proactively.
Limitations
This research is not without limitations. For example, it relies solely on one newspaper source. Although Asahi Shimbun is a reputable national outlet and commonly used in research on eldercide in Japan, cross-referencing multiple newspapers or open-source outlets would likely improve case completeness and reduce missing information, as different media sources often emphasize different aspects of the same incident. The exclusive use of one newspaper may therefore have contributed to underreporting of variables such as victim health, caregiving arrangements, or prior abuse. The use of newspaper data nevertheless reflects a strategic methodological choice, as official police statistics, court records, and government reports in Japan do not systematically identify caregiver eldercide or disaggregate homicides by caregiving relationship. Future research could strengthen empirical coverage by triangulating multiple media sources. Relatedly, because data were coded from the content of newspaper articles, there is a possibility that some variables may not fully reflect the reality of the cases due to space constraints or editorial omissions. To address this limitation, future research should consider triangulating data through qualitative methods, such as interviews with investigators, legal professionals, or individuals directly involved in the cases, to obtain a more comprehensive understanding of the dynamics underlying eldercide by family caregivers. A further limitation concerns the use of article word count, following Wong and Lee (2021), as a proxy for informational richness when selecting cases for qualitative illustration. Article length may not necessarily correspond to the depth or quality of detail provided, as journalistic style and reporting practices may enable some shorter articles to convey substantial contextual information, while longer reports may still omit important details. Consequently, this approach may have resulted in the exclusion of some cases that contained rich qualitative information despite having a lower word count. Future research could address this limitation by combining word count with additional qualitative screening criteria, or by drawing on official case files or interview data to capture a broader range of information-rich cases.
Conclusion
Given the global trend of population aging, eldercide by family caregivers is likely to increase and persist as a pressing social issue. As with any complex phenomenon, the essential first step toward effective intervention is to establish a clear and comprehensive understanding of the problem. Drawing on cases reported in a major Japanese national newspaper between 2000 and 2024, this study employed a mixed-method content analysis to construct a foundational overview of eldercide by family caregivers in Japan. The findings offer a concise yet powerful portrait of the key characteristics and contextual conditions underpinning these tragedies. With this groundwork established, the next critical step is to conduct more nuanced, interdisciplinary research to deepen our understanding of the multifaceted nature of eldercide. In this regard, the present study serves as an important starting point for future inquiry and policy development.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.