General medicine

In reply:In their comment on our article,1 Adler and Adler are correct in noting that sexual harassment is a significant issue facing female doctors, as it is for women in many occupations.2 Those who experience sexual harassment are also more likely to experience bullying and other forms of workplace incivility.3 Workplace bullying has also been found to be related to higher risk of suicidality.4 In view of this, we agree that these would be worthwhile targets for prevention initiatives. But the question remains as to whether sexual harassment and the problems regarding gender role stress discussed in our article1 all stem from the same set of causes. A meta-analysis3 noted two systematic sources of sexual harassment: (i) organisational context, including a climate that permits sexual harassment, and policies that do not support the reporting of and action against perpetrators; and (ii) the context of gender in the workplace — for example, the extent to which traditional gender roles are able to be maintained, such as through gender differences in the organisation of work, and the overall proportion of women employed in the job. In our article, we specifically highlighted the maintenance of gender-normative behaviour in medicine as potential stressors for suicide. While we agree that sexual harassment is a significant issue that needs to be addressed, we suggest that this be incorporated into a wider and overall strategy to overcome gender inequality in medicine. This strategy should target the negative personal and organisational (eg, organisation commitment, work withdrawal, job satisfaction) outcomes of harassment,3 and include measures to allow both male and female doctors to care for their families. In addition, to align with best practice in workplace suicide prevention,5 we would also support a stigma reduction campaign, access to treatment for health practitioners who experience poor mental health and suicidal ideation, and support for colleagues and families following bereavement from suicide within the medical profession.

Allison J Milner · Matthew J Spittal · Marie M Bismark

To the Editor:Advance care planning (ACP) promotes conversations about future health care, in case a person should lose capacity for decision making. Advance care directives (ACDs) provide written documentation of these conversations. Yet, although the Australian Medical Association advocates ACP within routine clinical practice,1 ACD completion rates remain low.2 While recognising numerous barriers to ACP, including patient, practitioner and health care system factors,3 a dedicated ACP Medicare Benefits Schedule (MBS) item number was mentioned in a number of general discussions at the 2016 Advance Care Planning Australia national conference (Melbourne, 15–17 November) as a potential incentive to increase the use of ACDs in primary care. In general, we support this recommendation. Health economics tells us that where there is a shortage of a specific service, as could be argued for ACP, the supply of this service will increase under fee-for-service payment.4 Patients are less knowledgeable about prices, services and associated benefit than providers; however, the existence of an agency relationship5 (where the patient assigns decisional authority on the basis that they have less information) between general practitioners and patients would facilitate ACDs from which patients are most likely to benefit. On the other hand, there is some risk that financial incentives will motivate GPs to do more than is optimal or desired, especially if the service is priced above standard consultation fees. There is also an opportunity cost; if GPs are providing more ACP services they will have less time to provide other primary care services. However, the risk of financial incentives may be mitigated by the non-financial barriers to ACP (fear or reluctance of patients, insufficient GP skills or organisational factors). Therefore, in designing policies to increase uptake, especially where financial incentives are the driver, a multifaceted approach should be considered. Service reimbursement mechanisms and potential barriers should also be balanced against maintenance of ACP values such as patient autonomy and informed decision making. As a starting point, a two-tiered MBS item number could be beneficial, the first item billable for initiation of ACP and the second used for revisitation of ACDs after a given time period (eg, annual review). Subsequent review of policy responses, including patient and medical practitioner input, will be needed to ensure appropriate directions surrounding MBS billing. Further accompanying strategies may be required to address non-financial barriers to ACP uptake.

Amanda Pereira-Salgado · Jennifer J Watts