Sir,

We agree with Dekker et al (2013) that resiliency is likely an important construct to take into account when detecting and treating psychological distress, and it may be helpful to identify risk factors for less resiliency in order to target ‘at-risk’ populations.

The authors argue that screening the entire cancer population for distress, when the majority of them are resilient and will likely not be in need of specialised care, may not be the most efficient approach. Instead, they propose using a matrix of empirically derived stable demographic characteristics (i.e., risk factors) to identify those most likely to suffer from considerable distress. Then, more specific case-finding tools could be applied to maximise specificity. The specific risk factors suggested by this group included absence of informal care provided by family members or friends, younger age and graft-vs-host disease. They suggest most of these could be identified by simple chart review.

We have ourselves pursued this idea of risk factors for continued trajectories of distress, and found that factors predicting a continued distress trajectory across 12 months following diagnosis were being younger, female, having chemotherapy or radiation therapy treatments and suffering from head and neck, gynaecological or gastrointestinal cancers (Enns et al, 2013).

While this approach undoubtedly has some merit, the main problem is that while these characteristics may confer elevated risk for distress, they do not fully account for all elevated distress; certainly there are many older individuals, for example, who suffer from elevated distress and could benefit from treatment. Would we be willing to risk completely missing their needs by excluding them from screening for distress or case-finding programs? Also, while many of the demographic and treatment-related factors suggested by Dekker et al (2013) may be available on chart review, someone would have to do that chart review, incurring potentially considerable expense, and arguably other risk factors (for example, social support) would not be easily discernable from charts alone. We are not sure that process would be any quicker or easier than rapid, low cost, automated screening for distress across the entire population. Finally, using risk-based assessment cannot take into account disease progression or treatment side effects, all of which may change the patient distress profile over time. The model for Screening for Distress as the 6th Vital Sign calls for repeated assessments of distress over the disease and treatment trajectory (Bultz and Johansen, 2011), and hence is responsive to changes in patient needs over time.

The authors suggest there are, therefore, two alternative approaches towards detecting and treating psychological distress in patients with cancer: ‘screening for psychological distress’ and ‘supporting resilience and case finding’. The latter would involve applying risk factors as exemplified above to identify those at high risk, then further case-finding and treatment within this group.

Both approaches undoubtedly have merit, but while the authors suggest there is limited empirical support for the benefits of screening for distress as we have applied it, there appears to be even less support for the idea of ‘supporting resilience and case finding’. It would be most interesting to see the authors or other qualified researchers compare these two approaches to see which better detects distress and results in appropriate treatment and decreased overall population-level distress over time.

Based on the results recently published in this journal, we ourselves have suggested something of a hybrid screening programme which relies on direct triage of high-risk patients, coupled with computerised referrals for those in low-risk subgroups (Carlson et al, 2012). To us, routine screening for distress as a vital sign seems a simple, quick, low cost, responsible and effective way of approaching the complexity of cancer care. Perhaps the next steps suggested by this conversation would be comparison of these various models of care. Until that research is done, this remains but an interesting theoretical debate.