Introduction

While intimate partner homicide (IPH) is a global issue affecting thousands of women every year (UNODC, 2019), as the most extreme form of intimate partner violence, there are still gaps in the knowledge about this serious crime. While research in Sweden has explored the rates, trends and characteristics of IPH in general, no study to date has focused on the most extreme form of intimate partner violence (IPV) involving adolescents and young adults. This constitutes a critical lack of understanding regarding the experiences and victimization of violence among young people in Sweden that this study aims to remedy. Not only does this lack of knowledge mean that a fundamental understanding of the trends and characteristics of youth IPH is missing, but it also jeopardises the success rate of any prevention programs and policy initiatives aimed at reducing severe IPV and IPH. In order to address this gap in the research literature, this study aims to investigate trends of IPHs in Sweden between 1990 and 2017, disaggregated by age-group and gender. By using poisson modelling, taking into account any shifts in population demographics, this study will explore if and how rates of youth IPH has changed over time for adolescent and young adult women and men, and compare these trends to rates of adult IPH over time.

Background

IPV is a global public health issue that affects thousands of women and children worldwide, often with detrimenal short-and long-term consequences (WHO, 2021). While IPV historically has been perceived as a private issue, often handled within the family unit, it is now acknowledged as a serious offense with complex implications (Matias et al., 2020). Victimization and perpetration of this offense has also been shown to occur at all ages, affecting youths as well as adults (Korkmaz et al., 2022; Reyes et al., 2021). Yet, until recently, youth IPV has been an overall neglected area in research and practice (Schuster & Tomaszewska, 2021), leading to major gaps in knowledge with regards to the incidence, trends and characteristics of youth IPV (NCCP, 2021). Lethal IPV, or IPH, is the most extreme form of this type of violence and while IPH is considerably less frequent in comparison to IPV, the consequences of IPH are still extreme and far-reaching, and is considered a public health problem in its own right (Rai et al., 2020). Similar to research on youth IPV, the scientific landscape of IPH has mainly investigated IPH as an adult phenomenon, in which little consideration has been taken to younger couples (Adhia et al., 2019). Consequently, youth IPH is neither well documented nor well understood (Bush, 2020).

The public health concerns of IPV and IPH have resulted in national and international resolutions and treaties. In 2013 (A/RES/68/191) and 2015 (A/RES/70/176) the United Nations General Assembly adopted two resolutions on gender-related killings of women and girls. The Commission for instance urged Governments to improve national legislation and incorporate mechanisms or policies with the aim to prevent, investigate and eradicate gender-based violence (A/RES/68/191). Governments in several countries have taken action in various ways in order to address and prevent IPV and IPH, for example by changing laws, adopting policies for early intervention, as well as ensuring the education and training of professionals working with these issues (UNODC, 2019). The Council of Europe Convention on preventing and combating violence against women and domestic violence, also called the Istanbul Convention, is a European legal instrument ensuring women’s right to live without being subjected to violence. It has been signed and ratified by 37 countries, meaning these governments are legally bound by these obligations. As such, the Istanbul convention constitutes a comprehensive and coordinated response to violence against women, including domestic violence, across and beyond Europe, aiming to improve prevention, protection, prosecution and coordinated policies. As such, great progress has been made with regards to efforts to combat and prevent violence against women. However, it remains unknown whether these efforts are reflected in figures concerning IPH. In fact, a considerable number of women are still being killed by their intimate partners, even after seeking support and protection by agencies (Swedish National Board of Health and Welfare, 2024; Vatnar et al., 2017). Empirical research exploring the distribution and characteristics of IPH can inform and improve the development of prevention programs as well as intervention strategies. A strong evidence base of IPH can not only improve these initiatives but also reduce the incidence rates of IPH against women (McPhedran et al., 2022). This underlines the importance of conducting further studies relating to IPH of women globally. As homicide in intimate partner relationships is not only a serious public health issue, but also a gendered one, exploring this severe crime further is imperative.

Previous research demonstrates that overall homicide rates in Western Europe and the U.S. increased in the 1960s, followed by a declining trend in the early 1990s (Aebi & Linde, 2014). Some studies have observed similar declining trends in IPH rates in Western Europe, Canada and the U.S. (e.g., Caman et al., 2017; Corradi & Stöckl, 2014; Dawson et al., 2009; Skott, 2019; Vatnar et al., 2022, UNODC, 2019). However, in contrast to other homicide types, rates of IPH and domestic homicides tend to be relatively constant over time (Caman et al., 2017; UNODC, 2019, 2022). Nonetheless, empirical research from Canada demonstrates that relative employment and divorce rates are associated with rates of IPH against women (Dawson et al., 2009). The study also shows that changes in divorce rates and men’s education were associated with rates of IPH against men (Dawson et al., 2009). It has also been suggested that the IPH rates are associated with availability of shelters to female victims of IPV (Browne et al., 1999).

While the majority of homicide victims are male, the vast majority of victims killed by a partner or another family member are women (Stöckl et al., 2013; UNODC, 2019, 2022; Vatnar et al., 2022). In Europe, the rate of IPH against women in 2017 was on average four times higher than the rates of IPH against men (UNODC, 2019). This means that while men tend to be killed by acquaintances in public places, women are most commonly killed in situations where they should feel the safest; in their homes, by an intimate partner. While about a third of all women who are killed worldwide have been killed by a current or former intimate partner, the proportion of IPHs against women varies greatly between regions as well as countries.

As stated in the UNODC (2019) report; “in general, countries with relatively low female total homicide rates tend to have a relatively larger share of female intimate partner/family-related homicides, whereas in countries with relatively high female total homicide rates, the share of female intimate partner/family-related homicides is relatively smaller. This is because more women are killed outside the family sphere, although the actual intimate partner/family-related homicide rate may still be relatively high (UNODC, 2019, p. 13).” This is clearly illustrated in countries such as for example Lithuania and Hungary, which stand out with high rates of lethal violence against women outside of the partner- or family unit. Similarly, countries such as Switzerland, which has a very low overall rate of homicide, has a proportionally higher rate of lethal intimate partner violence against women (Markwalder et al., 2019). This could indicate that IPH, particularly against women, is more complex and difficult to reduce, even when other types of homicide occurring outside the home appears to have been successfully reduced.

Lethal violence against women perpetrated by intimate partners is consequently a widespread phenomenon that occurs in all regions and countries. In addition, the rates of IPH tend to be more homogenous than the overall homicide rates across regions (UNODC, 2022). The global rate of IPH against women in 2017 was estimated to 0.8 per 100,000 female population. In absolute numbers, most intimate partner killings occur in Asia, however, in consideration of population figures, women are at greatest risk of being killed by intimate partners in Africa (1.7 incidents per 100,000 female population) (UNODC, 2019). Scholars have previously emphasized that there is a lack of research on gender-specific trends of IPH across countries (Reckdenwald & Parker, 2012) underlining the need of examination of IPH from a European context (Corradi & Stöckl, 2014). While the average rate of IPH against women in Europe is 0.6 incidents per 100,000 female population, there are great variations in the IPH rates across European countries (UNODC, 2019). The Nordic countries (e.g., Sweden, Finland and Iceland) stand out in Europe as well as worldwide in terms of being characterized by higher gender equality in various ways. These countries have repeatedly over the years been positioned as leading countries in Europe (EIGE, 2022) and globally (World Economic Forum, 2022) with regards to gender parity, and were comparatively early in applying efforts to combat violence against women (Corradi & Stöckl, 2014). Yet, the overall rates of IPV and IPH against women have been outlined to be rather high in the Nordic countries (López-Ossorio et al., 2021). As such, this conundrum has been referred to as the “Nordic paradox”, a phenomenon yet to be understood and explained (Gracia & Merlo, 2016). Nonetheless, being the absolute leading country in Europe with regards to gender equality (EIGE, 2022), Sweden constitutes an interesting example in terms of investigating IPH patterns and rates. Such investigations should also explore different forms of IPH, divided by age and gender.

Previous research has underlined that the rates of various forms of IPV among adolescents and young adults are high (Korkmaz et al., 2022), and that the young population is particularly vulnerable for IPH (Seimer, 2004). As previous research in homicide more generally has also highlighted the importance of exploring disaggregated trends in lethal violence, where age, gender and intimate partner relationship should be examined specifically (Caman, 2017; Skott, 2019; Walby et al., 2016), this further stresses the importance of conducting further studies on youth IPH specifically. Yet, there is currently a dearth of research on youth IPH. Intimate partner killings tend to occur during the whole life course (Chopra et al., 2022). Yet, to our knowledge, no study has previously explored the trends of youth IPH, comparing youth IPH to adult IPH. While this lack of research might be related to the low number of cases of youth IPH, it may also be related to a different attitude and viewpoint when it comes to intimate relationships among youth. As IPH traditionally has been regarded as an ‘adult phenomenon’ (Adhia et al., 2019), violence occurring among young people tends to get downplayed and trivilialised, considering tumultuous reationships as something inherent and normal to youth. While IPV traditionally has tended to be theoretically understood either as a function of individual pathology, expressions of conflicts within the family or as an expression of gender-based oppression and inequality (Lawson, 2012), more recent theoretical work has also explored IPV from an intersectional perspective, examining how various intersection power orders such as gender, age, ethnicity, and class affect the perpetration and experiences of IPV (Boonzaier et al., 2020; Barrios et al., 2021). As the normalization of youth violence by adults has indeed been found in previous research relating all types of violence, and not just IPV (Skott, 2023), this trivialization could be related to societal power structures more generally concerning how young people are constructed and perceived.

As prevention measures implemented to combat IPH in adult relationships may not be suitable or compatible with violence involving younger victims, further studies examining youth IPH is not only relevant, but imperative if IPH among youths is to be successfully prevented. As such, the current study aims to address two gaps in the literature on IPH; (a) a documentation of gender-specific IPH rates over time, and (b) a consideration of subgroup variations in which rates of IPH against young victims are disaggregated and analyzed. This line of research can be fruitful for identifying particularly vulnerable groups of individuals, and inform direction for future prevention and intervention policies. In order to attempt to fill these knowledge gaps, the overarching objective of the current study is therefore to investigate and compare trends in rates of IPH in Sweden, disaggregated by gender and age. We aim to analyze gender-specific rates of IPH involving young victims, and compare these to rates of IPH involving adult victims.

More specifically, the following research questions will be addressed:

  1. 1

    What are the overall rates and trends of opposite-sex IPH against men and women respectively?

  2. 2

    What are the overall rates and trends of IPH against young women (≤ 25 year), young men (≤ 25 year), adult women (≥26 year) and adult men (≥26 year)?

  3. 3

    Are there differences in trends of IPHs over time depending on gender and age-group?

Methods and Materials

This population-based study involves all solved IPHs in Sweden between January 1st, 1990 and December 31st 2017. A dataset was obtained from a research project at the Swedish National Council for Crime Prevention, encompassing all homicides committed in Sweden. Information from police files, court verdicts and forensic psychiatric reports have been systematically coded according to a specific coding scheme. As such, data has been triangulated from several data sources. In order to ensure that all cases identified as homicide by the judicial system are included, and that no case has been duplicated, the dataset has been checked repeatedly. For example, comparisons have been made between the current dataset and the Cause of Death registry from the National Social Board of Health and Welfare, as well as official crime statistics from the Swedish National Council for Crime Prevention. The statistics in the various sources demonstrate high congruency, and the high quality of the current dataset has been previously established (Liem et al., 2013). There are however some minor differences between the data sources; while the Cause of Death registry encompasses incidents committed outside the borders of Sweden, the current dataset solely includes homicides committed in Sweden. This dataset is especially valuable for analyzing trends over time and general patterns, as it allows for a wide time-frame and robust analyses between subgroups.

Homicide is defined as an intentional criminal act of violence by one or more human beings resulting in the death of one or more other human beings. According to the Swedish Penal Code, homicide covers cases of murder, manslaughter, infanticide, as well as assault leading to death. The original dataset includes both solved and unsolved homicides. A solved case refers to cases in which a perpetrator has been identified, charged and convicted, or cases in which a prosecutor has identified a perpetrator who could not be charged, for example due to the perpetrator committing suicide in the course of perpetrating the offense. As the relationship between the offender and the victim is a key variable in current study, unsolved cases were excluded. For the same reasons, cases in which the offender-victim relationship was unknown were omitted from the analyses. As the sole focus of the study is IPH, only incidents in which the offender and victim were or had been in an intimate relationship were included in the analyses. The definition used in current study of ‘intimate partners’ is incidents in which the offender and the victim were or previously had been married, engaged, cohabitants or boyfriend-girlfriend. The same definition was used for both young and adult victims. A key variable from the dataset is type of homicide, in which the typology from the European Homicide Monitor (Liem et al., 2013) has been adopted to categorize homicides. The typology is based on the combination of information regarding relationship and motive, in which relationship has greater importance.

As the age of criminal responsibility is 15 years, the study includes offenders aged 15 and older, and incidents involving offenders with a completely unknown age were excluded. Incidents involving victims with unknown ages were included in analyses of overall trends of IPH, but excluded from analyses of age-specific IPH trends. Youth IPH is defined as IPH against victims who are 25 years old or younger, while adult IPH constitutes IPHs committed against victims who are 26 years old or older.

In order to gain more accurate measures of trends, scholars have emphasized that rates, rather than number of incidents, in a given time and place should to be disclosed, as it also takes the total population into consideration (Mazerolle et al., 2015). Correspondingly, rates per 100,000 inhabitants for each subgroup were estimated, based on age-specific population statistics for each year, retrieved from Statistics Sweden. In order to estimate rates of youth IPH against women, population figures for women between the ages of 15 and 25 were obtained. Similarly, estimations of rates of youth IPH against men were calculated by obtaining population figures of men between the ages of 15 and 25. In order to calculate rates of adult IPH against women, population figures of women between the ages of 26 and 99 years were retrieved, and the corresponding figures for men were collected. The study has been approved by the Regional Ethical Review Board in Sweden (Protocol-ID: 2020–02534).

Data

The advantage of investigating IPH is that homicides (including IPHs) are less characterized by dark figures. In other words, in comparison to non-lethal violence and other types of crimes, homicides are less likely to remain undetected by or unreported to the authorities. It is also less likely to be related to reporting errors or police practices (van Breen et al., 2023). In total, there were 2,358 solved incidents of homicide in Sweden between 1990 and 2017. Of these, 22% (n = 521) were classified as IPH, and involved 516 offenders and 521 victims of IPH. A few (n = 3) offenders committed IPH more than once, constituting repeat IPH offenders. In 3 cases, the offender gender was unknown and these cases were subsequently removed, generating a total IPH sample of 518 cases. Overall, there were 431 (83%) cases of male-perpetrated IPH, and 87 (17%) cases of female-perpetrated IPH.

There are incidents in the dataset in which the age of the victim or the offender is unknown. If it was unknown whether the offender was older than 15 years, the incident was excluded from the analyses (n = 2). When the victim age was completely unknown (n = 56), the incident was removed from the age-specific analyses, but included in analyses of overall IPH trends.

There were in total 65 incidents of youth IPH, which constitute 13% of all IPHs between 1990 and 2017. One of these cases involved (female-to-female) same-sex IPH, which constitutes 1,5% of all youth IPHs. The corresponding figures in adult IPHs is five cases of same-sex IPH and 0,9%. As previous research has shown that same-sex IPHs exhibit different characteristics than opposite-sex IPHs (Gannoni & Cussen, 2014; Mize & Shackelford, 2008), same-sex IPHs ought to be analyzed separately. Due to the small number of cases, these were omitted from the analyses.

In total, the overall sample consists of 457 incidents of opposite-sex IPH, 383 (84%) of these cases were perpetrated against a female victim and 74 (16%) of these cases were perpetrated against a male victim. In consideration of age, there were 64 incidents of opposite-sex youth IPH, in which 62 (97%) involved female victims, and 393 incidents of adult IPH, in which 321 (82%) were female victims.

Statistical Analyses

Trends in incidence rates over time were analyzed by conducting poisson regression models, in which the specific population was regarded as the exposure factor, time (year) was regarded as the explanatory variable and number of IPH incidents was regarded as the dependent variable. Poisson regression was chosen since it allowed for the examination of the difference in average change over time between different groups, while simultaneously adjusting for population and exposure over time. Poisson modelling also does not assume linearity or normality of the data. We chose to separate the dataset into the specific subgroup (male/female young/old) instead of including them as explanatory variables in the model, as the main focus is the development within each subgroup. The analyses resulted in outcome measures called Incidence Rate Ratios (IRR). An IRR of 0.8 implies a 20% yearly decrease, while 1.2 suggests a 20% yearly increase in incidence rate. Tests for overdispersion was conducted by the dispersion test in R in order to explore heterogeneity in the data, with good fit for all subgroups. Residuals analysis was performed with Shapiro-tests (for normal distribution), and chi-square tests for deviance residuals (see Appendix). Statistical analyses were performed using R.

Results

Firstly, we found that 22% of all homicides in Sweden between 1990 and 2017 constitute IPH; meaning that little more than fifth of all homicides in Sweden are committed against a current or former intimate partner. The majority of the victims were female, regardless of age-group; however, the proportion of female victims was somewhat higher in youth IPHs (97%; n = 62) compared to adult-IPHs (82%, n = 321). The youngest victim was 15 years old, and the mean age of female victims of youth IPH was 22 years (range 15–25, SD = 2.9), and the mean age of male offenders was 28 years (range 16–51, SD = 7.8). The mean age in female victims of adult IPH was 47 years (range 26–88, SD = 15.0), and 49 years (range 24–88, SD = 15.0) in male offenders. With regards to female-perpetrated IPH among adults, the mean age in adult male victims was 47 years (range 28–81, SD = 11.7), and 41 years (range 19–65, SD = 10.2) in female offenders. Due to the limited number of cases of IPHs against young men (n = 2), analyses of such could not be calculated.

Regardless of age, the average rate of IPH against women in recent years (between 2014 and 2017) is 0.34 incidents per 100,000 female population. The age-specific analyses demonstrate that the corresponding rates of IPH against young and adult women were equivalent; 0.32 incidents per 100,000 female population involving young women, and 0.34 incidents per 100,000 female population involving adult women. As there were two incidents involving young men, the corresponding figures for overall rate of IPH against men and rate for adult men were the same, with 0.05 incidents per 100,000 male population.

With regards to trends in rates over time, the figures below demonstrate annual rates per 100,000 inhabitants (blue line), and a trendline based on poisson regression analyses (red line), illustrating the trends in rates over time in the distinct subgroups. In Fig. 1, trend in overall rates of IPH are presented, regardless of gender or age. The average annual decrease of IPHs between 1990 and 2017 in Sweden was 1.7% (IRR = 0.983, p = .002).

Fig. 1
figure 1

Trend in rates of intimate partner homicide per 100,000 inhabitants

Full size image

In the next step, gender was accounted for in the analyses. Consequently, Fig. 2 presents trends regarding IPHs against women, illuminating an annual decrease of 1.4% in IPHs against women, regardless of age (IRR = 0.986, p = .014).

Fig. 2
figure 2

Trend in rates of intimate partner homicides against women per 100,000 female population

Full size image

The trends in rates of IPHs against men is presented in Fig. 3. A visual inspection of the graph indicates a declining trend; however, this trend was not statistically significant, possibly due to the low number of incidents (IRR=, 976, p = .060).

Fig. 3
figure 3

Trend in rates of intimate partner homicides against men per 100,000 male population

Full size image

In order to investigate whether there are differences in trends over time depending on victim age, trend analyses were conducted in which both age and gender were considered. Figure 4 reveals that the trend in rates of IPHs against young women between the ages of 15 and 25 years has remained stable between 1990 and 2017 (IRR = 1.001, p = .932). As such, youth IPHs against women has neither decreased nor increased.

Fig. 4
figure 4

Trend in rates of intimate partner homicides against women between 15 and 25 years old per 100,000 female population

Full size image

In contrast, there has been an annual decrease of 1.7% in IPHs against adult women aged 26 years and older, as shown in Fig. 5 (IRR = 0.983, p = .006). In other words, the decrease in IPHs against women (shown in Fig. 2) is predominantly propelled by a decrease in rates involving adult women, while the rates involving young women have remained constant. The nonsignificance of the trends of young female IPH could be related to low n, however, as the IRR demonstrate no decline (1.001), our findings do suggest a constant trend in IPH against adolescent and young adult women .

Fig. 5
figure 5

Trends in rates of intimate partner homicides against women 26 years and older per 100,000 female population

Full size image

As previously mentioned, there were only two incidents of IPH against young men between 1990 and 2017. As such, it was not possible or relevant to calculate trends in youth IPH involving male victims. However, with regards to IPHs against adult men, the graph (Fig. 6) indicates a declining trend. In line with the analyses presented in Fig. 3 (in which only two cases differentiate these analyses), the trend was not quite statistically significant (IRR = 0.975, p = .053).

Fig. 6
figure 6

Trends in rates of intimate partner homicide against men 26 years and older per 100,000 male population

Full size image

In summary, our results demonstrate that overall rates of IPH in Sweden have declined between 1990 and 2017. However, disaggregation of gender demonstrates that the rates involving female victims have declined significantly over time, while the declining trend involving male victims could not be established statistically, likely due to the low number of incidents. In consideration of both gender and age-group, our analyses reveal two important findings. (1) IPH against young men is a rare phenomenon, while the proportion of male victims is higher in adult IPHs. (2) The trend in rates of IPH against young women have remained stable during the study period.

Discussion

There are major gaps of knowledge in Sweden regarding incidence, trends and characteristics of youth IPV, especially with lethal outcome (NCCP, 2021). There are, to date, no Swedish and very few international studies on the scope and characteristics of youth IPH (Bush, 2020; Kistin et al., 2019). The objective of current study was to analyze and compare trends in rates of IPH in Sweden, in which gender-specific rates of IPH involving young victims were compared to rates of IPH involving adult victims. Research on rates and trends of IPH, disaggregated by gender and age, is essential for evidence-based development of prevention and intervention policies.

First and foremost, our findings highlight that a substantial proportion of all solved homicides are perpetrated by a current or former intimate partner of the victim. As such, IPH warrants serious sociopolitical attention and evidence-based strategies. Our results also demonstrate that, regardless of age, the average rate of female IPH between 2014 and 2017 in Sweden is approximately six times higher than the rate of male IPH. Meanwhile, the average rate of female IPH in Europe is four times higher than the average rate of male IPH (UNODC, 2019).

Previous research has highlighted that the rates of IPH against women are particularly high in the Nordic countries such as Sweden, Iceland and Finland (López-Ossorio et al., 2021). Our study elucidates that, irrespective of age, the Swedish average rate of IPH against women the past years (2014–2017) is 0.34 incidents per 100 000 female population. As such, the female rate of IPH is lower in Sweden compared to some European countries, such as Iceland (0.6), Finland (0.4) and Albany (0.7), while higher than the rates in for example Spain (0.2) and Italy (02). Instead, the average rate in Sweden is similar to the rates of IPH against women reported in France (0.3) and Switzerland (0.3) (UNODC, 2019).

In consideration of age, our study discloses several noteworthy and novel findings. First, the past years’ average rates of IPH against adolescent and young women are similar to the average rates involving adult women. So, while IPH has been perceived as an adult phenomenon and the literature almost exclusively focus on adults (Adhia et al., 2019), our findings in which population figures are accounted for, clearly show that the rates of youth and adult IPH involving female victims, in fact, are almost identical. These findings support the statement by Chopra et al. (2022) that IPH can occur during the whole life course. Second, while there has been an annual decrease of 1.7% in adult female victimization of IPH in Sweden, the trend of IPH against young women have remained stable the past 27 years. This means that the decreasing trend in IPH appears to be driven by a decrease in adult IPH, whereas IPH involving young female victims has not presented any indications of decreasing over time. Last but not least, it has been reported that youth IPV more frequently tends to be bidirectional (Schwartz & Kallumkal, 2022) and to be relatively gender symmetrical with regards to perpetration (Bender et al., 2021). Unlike youth IPV however, IPH involving young victims almost exclusively affects females. In line with previous studies (Adhia et al., 2019; Bush, 2020), our figures illustrate that IPH is a strongly gendered phenomenon, particularly evident in youth IPH. Youth IPH rarely involves male victims and the proportion of female victims is higher in youth IPH, compared to adult IPH. Taking an intersectional perspective, this could suggest that young women are particularly vulnerable in relation to IPH, both in relation to gender but also in relation to age. As previously suggested, the lack of research in relation to youth IPV and IPH could be related to a different attitude or viewpoint when it comes to intimate relationships among youth where violence in such relationships is normalized and trivialized. Contextualising this normalization of youth violence in broader societal power structures and hierarchies of both gender and age would suggest that young women might be particularly vulnerable to having their experiences of violence being downplayed or minimized. This might indicate that a shift in how youth IPV and IPH is regarded is needed in order to mobilise adequate response to this problem. This issue might also be related to societal norms relating to romantic relationships, where violence and controlling behaviour is romanticized within the context of romantic relationships (Papp et al., 2017; Cava et al., 2020). As such romanticisations of IPV are also very prevalent in popular culture (Byrne & Taddeo, 2019; Maas & Bonomi, 2021), future studies should explore the possible reciprocal relationship between IPV and popular cultural artefacts.

While there are many similarities between IPH involving young and adult women (Glass et al., 2008), there are however fundamental differences with regards to challenges and barriers that are specific to young victims of IPV (Kistin et al., 2019). While Sweden has been identified as an ‘early bird’ country in terms of policy-making and government actions concerning non-lethal and lethal violence against women (Corradi & Stöckl, 2014), much remains to be done with regards to addressing youth IPV, as IPV is still essentially perceived an adult issue. In turn, this view is reflected in societies’ policies, reports and responses to IPV, which mainly focus on violence in adult relationships. As a result, young victims of IPV and IPH tend to become invisible in the knowledge-base and in the responses to combat IPV (NCCP, 2021; Statskontoret, 2021). Professionals in the field testify that awareness and competency regarding youth IPV is low for professionals who work on strategic as well as operational levels (NCCP, 2021). In an evaluation of Swedish authorities’ work with the Swedish national strategy to combat men’s violence against women, the Swedish Agency for Public Management point out that young female victims are invisible (Statskontoret, 2021). Instead, the responsibility for IPV involving younger victims has been given to non-profit organizations (NCCP, 2021). In sum, the policies and responses of authorities and municipalities regarding IPV are not tailored or appropriate for the younger population. This could be one of the explanations why IPH rates involving young female victims have remained constant, while the rates of IPH against adult women have declined.

By narrowing the focus on violence in adult relationships, young people can find it difficult to understand as well as relate to available information about support services. As a result, this may affect how they perceive their own experiences of violence and consequently their willingness to disclose or report IPV experiences (Love & Richards, 2013). The adolescent years are characterized by identity formation and exploration of intimacy and sexuality. The cognitive and psychosocial development in combination with lack of previous experiences of intimate relationships makes young individuals particularly vulnerable to violent victimization. Therefore, early experiences of victimization or perpetration of IPV can be more harmful and have detrimental impact on the transition into adulthood, with long-term consequences (NCCP, 2021). For example, a study shows that young individuals who have early experiences of IPV are more likely to report victimization 12 years later (Exner-Cortens et al., 2017). In a similar vein, there is a risk that early experiences of IPV perpetration become persistent behavioral patterns in future intimate relationships (Fernández-González et al., 2020; Garthe et al., 2017), or ultimately lead to deadly violence. It has been suggested that young women have a higher risk than adult women to become victims of lethal violence, due to a higher risk to confuse jealousy and controlling behaviors with commitment and love, and to normalize and accept violence in the course of their intimate relationships (Seimer, 2004). It is therefore of great importance to address youth IPV and to intervene as early as possible.

Policy Implications

While efforts to combat men’s violence against women have intensified during the past decades, these efforts are not noticeably reflected in the trends concerning rates of IPH against young women, which continue to be persistent. As the recent rates of IPH against young women is nearly identical to the rates of IPH against adult women, it is urgent to recognize the possible severe consequences of youth IPV and to cease to perceive IPH as an adult phenomenon. In order to efficiently combat youth IPH, prevention and intervention strategies are required on multiple levels simultaneously. Moreover, these efforts are required to be tailored to the needs of the younger population, for example increased public awareness (primary level) and competency among professionals (secondary level) regarding youth IPV is essential, as is tailored information and responses.

Moreover, a priority moving forward should be ensuring efficient risk assessment, management and safety planning involving young women who face a risk of severe or lethal IPV (tertiary level). The Istanbul convention stipulates that support services and domestic violence shelters should be available and accessible to victims, regardless of background characteristics, such as age. Yet, a barrier that is specific to younger victims concerns access to protection and safety (Adhia et al., 2019; Kistin et al., 2019). For example, interviews conducted with several professionals in Sweden reveal issues related to access to protection and domestic violence shelters for young women who need protection from a violent partner, especially if the victim is under the age of 18 years old (NCCP, 2021). Numerous cases with high risk of severe or lethal violence, where the parents were not able to sufficiently protect the victim, have led to insufficient protection for young women who have been denied access to shelters (NCCP, 2021). Meanwhile, previous research has demonstrated an association between the rates of IPH and availability of shelters for female IPV victims (Browne et al., 1999). As such, there is a need of improvement with regards to non-discriminatory implementation of the Istanbul convention and making sure that responses to end violence against women are extended to a younger population, available to them and tailored to their specific needs. Moreover, it is likely that lack of access to protection and domestic violence shelters have greater impact on low-income victims who do not have the resources to seek other alternatives (Kistin et al., 2019).

Furthermore, it has been highlighted that youth IPV does not take place ‘behind closed doors’, but rather in arenas where adults are present, like schools and home settings (Korkmaz et al., 2022). Youths are also likely to interact with clinicians and counsellors, in for example pediatric and adolescent clinics, school-based clinics, STD-clinics and child and adolescent psychiatry, which gives opportunities to detect perpetration and victimization of IPV (Kistin et al., 2019). It is therefore necessary for professionals who work with youths to be provided training about signs of youth IPV and knowledge about risk markers for youth IPH. This could for instance be realized through programmes such as Ask, Support, Care (Hassan, 2020) that aims to equip a wide range of professionals with skills necessary to both identify IPV among their patients, clients and colleagues as well as the skills to safely raise the issue. When professionals identify these signs and risk factors, they ought to provide support, advice and discuss safety of the victim (Overstreet et al., 2021). In terms of cooperation and coordination of community responses, high-risk cases involving young victims may require involvement of other agencies than the ones that are traditionally involved in cases with adult IPV victims. It is of great importance to ensure efficient cooperation and coordination of responses in order to gain necessary insight, and to ensure victim safety and offender accountability. In addition, scholars argue that coordinated interventions need to be customized for at-risk cases (Vatnar et al., 2022). The fragmentation and lack of coordination of responses can put victims in increased danger, as “there is a growing appreciation that a victim’s risk may increase if one element of the system improves responsiveness, while others do not” (Buzawa & Buzawa, 2013, p. 134).

Future Research

Our findings demonstrate that it is meaningful to analyze age and gender-specific trends. Thus, we call for further research from European and non-European countries that specifically investigate trends involving young victims of IPH, disaggregated by gender. Beyond the investigation of trends in rates over time, the next step is to explore the characteristics of IPH against women, and investigate if and how IPH involving adolescent/young women differs from IPH against adult women. There are great similarities between cases of youth and adult IPH (Glass et al., 2008). However, the dearth of research on youth IPH has led to a gap of knowledge about characteristics and unique risk factors for IPH against adolescent and young adult women (Kanbur et al., 2021). As such, existent risk assessment tools may not be as tailored or sufficient in identifying risk for severe or lethal violence against younger women (NCCP, 2021).

In addition, more studies on risk factors of IPH are needed from a European context. A meta-analysis by Spencer & Stith (2020) establishes that the two strongest predictors of IPH are the perpetrator having direct access to guns and the perpetrator threatening the victim with a gun. Accordingly, gun control and banning access to guns are policies that have been highlighted (Adhia et al., 2019; Vigdor & Mercy, 2006). In fact, states that ban offenders from owning guns in connection to IPV-related restraining orders were correlated with having nearly 10% decrease in IPH and 14% decrease in firearm-related IPH compared to states without such laws (Díez et al., 2017). However, while the majority of IPHs in the US are committed using handguns or other firearms (Overstreet et al., 2021), the majority of IPHs in Sweden are commited using a knife, followed by strangulation. As such, the relevance of gun-related risk factors in the Swedish context are less clear (Caman & Dufort, 2022; Caman & Skott, 2024). Simultaneously, as firearms are imported from Ukraine, Syria and Balkans, the access to illegal firearms in Sweden are increasing. In line, firearm-related homicides committed by young men in a criminal milieu have increased significantly in recent years (Sturup, 2021). As such, the prevalence of firearm-related IPHs should be closely monitored in future research, as there is a risk of increase of firearm-related IPHs in Sweden, especially involving younger couples. In a similar vein, future studies should investigate the importance of the gun-related risk factors in a European context. A potential increase in IPHs involving illegal firearms would however not benefit from gun laws similar to the U.S, but would require other approaches.

Last but not least, IPV victimization has many detrimental consequences. While IPV may result in IPH, it can also lead to attempted or completed suicide. Recent studies from the U.S. have studied the connection between IPV victimization and suicide (Bush, 2020; Graham et al., 2022). It has for example been reported that adolescents who have been exposed to physical IPV are two times more likely to experience suicidal ideations, and nearly 2.5 times more likely to attempt suicide in comparison to adolescents without experiences of IPV (Baiden et al., 2021; Nahapetyan et al., 2014). It has also been shown that suicidality as a result of IPV victimization persists for years after the victimization (Exner-Cortens et al., 2017). Bush (2020) reports that problems related to intimate relationships precipitated 27% of youth suicides, and 36% of suicides among females between 15 and 44 in the U.S. In order to better comprehend the magnitude of deaths in European countries that are related to IPV in youths (and adults), future studies should also explore rates and trends of suicides that are linked to IPV. A more comprehensive examination of IPV-related deaths in future studies could include incidents of IPH, IPH-suicide and suicides (Graham et al., 2022).

Limitations and Strengths

One of the limitations to the current study is the relatively small sample size. However, the Swedish dataset that is used in the current study can still provide valuable insights to patterns and trends of the IPH the phenomenon as the data is population-based and therefore to be considered representative of the population as a whole. An advantage of the dataset is that it includes samples that are otherwise often excluded. For example, offenders who cannot be charged or convicted due to suicide in connection to the offense tend to be excluded in other studies (Giesbrecht et al., 2023), but are included in the present study. Furthermore, offenders who commit the offense under the influence of a severe mental disorder are also included in the current study, as the Swedish legislation stipulates that all offenders, irrespective of mental state, ought to be convicted to either prison or forensic psychiatric careFootnote 1.

Attention has been brought to the issue of relatively few countries collecting and making national data available on gender-related killings of women and girls (UNODC, 2019). For example, four in ten female homicides that occurred worldwide in 2021 remain unclassified (UNODC, 2022). While there have been improvements by a number of countries in terms of collecting sex-disaggregated homicide data, there are still widespread limitations with regards to collecting and presenting data disaggregated by offender-victim relationship. It has also been highlighted that a substantial percentage of incidents involve missing values with regards to offender-victim relationships (Langford et al., 1998). As a result, there are limitations in data availability to measure killings perpetrated by intimate partners, and the available statistics may underreport the true magnitude of the problem. An overwhelming majority of cases (99.3%) in the dataset used in the current study involves known offender-victim relationships (Caman et al., 2017). However, there is a risk of misclassification of relationships, especially involving younger individuals, in which some offenders may have been categorized as an acquaintance instead of a current or former partner. Intimate relationships among youths are more likely to be fluid and short-lived, which increases the risk of misclassification and, ultimately, underestimation of youth IPH (Coyne-Beasley et al., 2003).

An additional limitation of current study is the heteronormative perspective, with sole focus on opposite-sex relationships. While transgender individuals (Peitzmeier et al., 2019) and members of the LGBTQI + community (Chen et al., 2020) tend to have a higher risk of being victimized of IPV, these marginalized populations tend to be overlooked in homicide datasets and IPH research (AbiNader et al., 2023). While our findings suggest that 1.5% of youth IPH and 0.9% of adult IPH involve same-sex relationships, there is, again, a risk of misclassification, in which intimate relationships have wrongfully been categorized as acquaintances. An additional caveat in the current study is that a minority of IPH cases involved missing values related to the victims’ age, which may have an impact on the age-specific analyses of trends over time, underestimating the rates of youth IPH.

Finally, while poisson regression analysis is a suitable technique for analysing changes over time, adjusted for population groups, there is a risk of type II errors due to the low number of incidents in certain groups. As this means that the study has an inflated risk of ‘false negatives’, the nonsignificance of particularly the smaller groups, such as IPH against young women and IPH against adult men, needs to be considered with caution. However, as no previous study has previously explored the change of IPH among youths over time, and as the IRR of the subgroup of young women is still 1.001, indicating no change in trends, the current study still holds important findings relevant to the understanding of youth IPH.

Conclusions

Our study highlights that IPH occur among youths as well as adults, and that this is not only an adult phenomenon. In fact, the recent rates of IPH against young women are nearly identical to the rates of IPH against adult women. However, the gender disparity is even more pronounced in youth IPH, in which almost all victims were female. While the trend of IPH against adult women in Sweden has declined by 1.7% percent annually since 1990, the trend of IPH involving young female victims remains persistent. The dearth of knowledge regarding youth IPH may in practice mean that society is not as equipped with regards to risk assessment, management and protection of young women at risk of being severely injured or killed. Our study suggests that a point of direction for future policies and strategies to combat men’s violence against women is to intensify focus on young female victims, and to better tailor prevention and intervention efforts to this neglected and invisible subgroup of victims.