Morgan et al ^{Reference Morgan, Zolese, McNulty and Gebhardt1} highlight the high incidence of being a victim of intimate partner violence among female psychiatric patients in the UK. This is in keeping with a historic approach that has conceptualised domestic violence as something that men do to women and has only sought evidence for violence by men against women.

Partly this may be because women are more likely to report intimate partner violence than men. One study found that in the same sample of couples 28% of the women, but only 19% of their male partners, reported that their relationships were violent, suggesting underreporting in a third of men. ^{Reference Whitaker, Haileyesus, Swahn and Saltzman2}

In recent years researchers have approached populations without preconceptions as to the direction of violence. Large epidemiological studies have demonstrated that domestic violence is most commonly reciprocal and that when only one partner is violent there is an excess of violent women. Whitaker et al, ^{Reference Whitaker, Haileyesus, Swahn and Saltzman2} in a study of 14 000 young US couples aged 18-28 years, found that 24% of relationships had some violence and half of those were reciprocally violent. In 70% of the non-reciprocally violent relationships women were the perpetrators of violence. Reciprocal violence appears to be particularly dangerous, leading to the highest rate of injury (31.4%). This may be because reciprocal violence is more likely to escalate.

The International Dating Violence Study ^{Reference Strauss3} found that among students at 31 universities worldwide male and female students had similar rates of physically assaulting a partner (25% of men and 28% of women at the median university). There was parity for perpetrating severe assaults (used a knife or gun, punched or hit partner with something that could hurt, choked partner, slammed partner against a wall, beat up partner, burned or scalded partner on purpose, kicked partner) − 9% of male and female students at the median university. For severe injury (passed out, required medical attention or broke a bone) the perpetration rate was higher for males (median rate 3.1% by men and 1.2% by women).

A review of 62 empirical studies of female-perpetrated intimate partner violence ^{Reference Williams, Ghandour and Kub4} found rates of physical violence of 4-79% among adolescent girls, 12-39% among female college students and 13-68% among adult women. The researchers concluded that a significant proportion of females seeking help for victimisation are also perpetrators of intimate partner violence, and that those who treat battered women may need to consider addressing the perpetration of violence with their female clients.

Archer ^{Reference Archer5} attempted to resolve two competing hypotheses about partner violence, either that it involves a considerable degree of mutual combat or that it generally involves male perpetrators and female victims. His meta-analysis of 82 studies of gender differences in physical aggression between heterosexual partners showed that men were more likely to inflict an injury; 62% of those injured by a partner were women, but men still accounted for a substantial minority of those injured. However, women were slightly more likely than men to use one or more act of physical aggression and to use such acts more frequently. Younger aged couples showed more female-perpetrated aggression.

Only examining rates of violence perpetrated against women risks perpetuating an inaccurate stereotype of women as victims and men as aggressors. This may hinder women from receiving support to reduce their own perpetration of violence and may contribute to the underreporting of violence perpetrated by women against men.