Disclosures: Hallet reports personal fees (speaker honoraria) from AAA and Ipsen Biopharmaceuticals Canada outside the submitted work. Please see the study for all other authors' relevant financial disclosures.
April 01, 2022
3 min read

Patient-reported outcomes linked to nonfatal self-injury after cancer diagnosis

Disclosures: Hallet reports personal fees (speaker honoraria) from AAA and Ipsen Biopharmaceuticals Canada outside the submitted work. Please see the study for all other authors' relevant financial disclosures.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Self-reported signs of distress and an increasing total Edmonton Symptom Assessment System score within 180 days of cancer diagnosis appeared associated with higher odds of nonfatal, self-inflicted injury, study results showed.

The findings, published in JAMA Oncology, highlighted the need to monitor patient-reported outcomes and use them to direct management, according to researchers.

Odds ratios of non-fatal self-injury.
Hallet J, et al. JAMA Oncol. 2022;doi:10.1001/jamaoncol.2022.0203.

“Outside of randomized controlled trials, patient-reported outcomes are collected but are seldomly reviewed by health care providers and even less often used to tailor the management plan. Our data suggest that patient-reported outcomes can be used to identify patients at risk [for] dramatic events and in need of better support,” Julie Hallet, MD, MSc, FRCSC, surgical oncologist at Odette Cancer Center at Sunnybrook Health Sciences Centre and assistant professor of surgery at University of Toronto, told Healio.

Background and methodology

The researchers pursued the work based on their clinical observations in daily practice, having witnessed similar signs of distress among their patients dealing with cancer, Hallet said.

“We were wondering how we could identify those patients with very high levels of distress who may be in need for much more support through their cancer care,” Hallet said. “We used self-injury as a measure of severe distress and challenges in coping mechanisms.”

Through research, Hallet said they found self-injury after a cancer diagnosis occurs in three out of 1,000 patients, so they decided to use the symptom scores in patient-reported outcomes to examine those who might be at higher risk for self-injury.

The population-based matched case-control study included 2,030 adults with a new cancer diagnosis who reported an Edmonton Symptom Assessment System (ESAS) score within 36 months of diagnosis in Ontario, Canada, from 2007 to 2019. The ESAS, a validated patient-reported outcome measure, evaluates nine common cancer-associated symptoms.

Researcher matched 406 patients (55.2% women; 74.4% aged 50 years or older) who experienced a nonfatal self-injury and reported an ESAS score in the preceding 180 days with 1,624 control patients (55.2% women; 75.2% aged 50 years or older) who did not have a nonfatal self-injury.

Hallet and colleagues utilized multivariable conditional logistic regression to assess the association between moderate to severe ESAS symptom scores and total ESAS (range 0-90) score with a nonfatal self-injury in the following 180 days.

Key findings

Results showed the nonfatal self-injury cohort reported a higher proportion of moderate to severe symptoms and higher total ESAS score than controls before the event.

After adjustment for history of severe psychiatric illness and cancer treatment and accounting for all other symptoms, moderate to severe anxiety (OR = 1.61; 95% CI, 1.14-2.27), depression (OR = 1.66; 95% CI, 1.2-2.31) and shortness of breath (OR = 1.65; 95% CI, 1.18-2.31) as well as each 10-point increase in total ESAS score (OR = 1.51; 95% CI, 1.4-1.63) appeared independently associated with higher odds of subsequent nonfatal self-injury.

Hallet told Healio the association with shortness of breath was somewhat surprising.

“Anxiety and depression are often considered emotional symptoms, whereas shortness of breath is considered a physical one. When we looked into this further, we learned that shortness of breath is actually a very troubling symptom for patients with panic disorders,” Hallet said. “Overall, we think that this means that attention should be paid to patients reporting of higher severity of shortness of breath from a physical symptom management perspective, but also as an overlapping symptom potentially manifesting psychosocial distress.”


The results highlight the need to carefully monitor and utilize patient-reported outcomes in ongoing care, researchers wrote.

Julie Hallet, MD, MSc, FRCSC
Julie Hallet

“If patients report higher anxiety, depression or shortness of breath, they have a more than 60% increase in the risk [for] self-injury in the next 6 months. If their total symptom score increases, there is also a significant increase in self-injury in the following 6 months,” Hallet said. “Such reporting of symptoms should raise red flags that the patient needs better support, then tailored assessments and interventions can be put in place with the goals to reduce distress and its repercussions.”

Hallet and colleagues are continuing the research by looking at incidence of self-injury among patients living with cancer vs. those without cancer, health care costs associated with self-injury and the impact on patient survival. They also are creating a prognostic tool to assess individual risk for self-injury after a cancer diagnosis.

"Such a tool could be used on its own, via an online platform, for example, or integrated in electronic medical records to trigger red flags for intervention," Hallet said.

For more information:

Julie Hallet, MD, MSc, FRCSC, can be reached at Sunnybrook Health Sciences Centre, 2075 Bayview Ave., T2-102, Toronto, Ontario, Canada M4N 3M5; email: julie.hallet@sunnybrook.ca.