I appreciate Dr Fanslow’s editorial response1 to my Viewpoint entitled ‘Recognising and responding to partner abuse: challenging the key facts’ (N Z Med J. 2004 Sep 24). However she appears to have misunderstood some of my concerns. I completely agree that women are more likely to be physically and sexually assaulted by their partners and suffer significant injury than men. However considerable evidence indicates that women and men engage in physical abuse (such as slapping, hitting, or throwing objects) at about equal frequency, and the degree to which men and women verbally or psychologically abuse each other is unknown. While such non-physical abuse may well have serious health implications, this is a difficult area for a GP to assess and manage. Where relationship problems are detected before violence is entrenched, the use of a community-based communication and conflict resolution course for couples may be beneficial.2

Certainly, as GPs, we are well used to asking our patients about topics with which they may feel uncomfortable, and I agree that we should make sensitive enquiries whenever we have concerns that a patient may be suffering from abuse by another person. However I do not believe that current evidence supports us routinely asking all female patients aged over 15 years patients annually about partner abuse (as advocated by the Ministry of Health publication in question3). This is not because ‘some women feel uncomfortable with disclosing abuse’; it is because many women are not experiencing partner abuse and it is neither necessary nor appropriate to ask them about this possibility every year.

Dr Fanslow says it is likely that the Snively report4 under-estimates economic costs of partner abuse. However the Snively figures quoted in the Ministry of Health paper used a 14% lifetime prevalence (one in seven women and one in seven children) suffering from some form of family abuse (301,700 people) and then assumed that this was the number of adult women suffering abuse annually. According to the formulae and assumptions used, this included over 37,000 women needing treatment for dental injuries that year, whereas the actual number was less than 2000. All the costs included (such as GP and A & E visits, hospitalisations) are likely to be subject to a similar (18-fold) inflation.

Finally, while clearly there is an important overlap between child and partner abuse, the Edleson review article of 35 studies5 finding that the co-occurrence ranges from 6.5% to 97%, with diverse populations mostly of battered women or abused children not epidemiological samples, is not accurate evidence of the statement that ‘the co-occurrence of partner abuse with child abuse is 30–60%’.

I agree with Dr Fanslow that ‘all violence is potentially harmful’, but exaggerating or distorting the extent of the problem does those who are suffering its effects a disservice.