Symptom Distress, Including Pain, Predicts Lung Ca Survival

October 1, 1995

WINNIPEG, Canada-Clinical trials have shown that the level of symptom distress at the time of lung cancer diagnosis can predict survival, and now a study performed in an ambulatory setting has confirmed this finding. Patients with higher symptom distress scores are likely to die more quickly than those with lower scores, report University of Manitoba researchers Lesley F. Degner, RN, PhD, and Jeffrey A. Sloan, PhD.

WINNIPEG, Canada-Clinical trials have shown that the level ofsymptom distress at the time of lung cancer diagnosis can predictsurvival, and now a study performed in an ambulatory setting hasconfirmed this finding. Patients with higher symptom distressscores are likely to die more quickly than those with lower scores,report University of Manitoba researchers Lesley F. Degner, RN,PhD, and Jeffrey A. Sloan, PhD.

They note that clinicians can use this information to counselpatients and families and discuss life completion issues. In addition,the data raise the interesting question of whether interventionsto palliate symptoms could influence survival.

The study involved 434 newly diagnosed cancer patients from twoclinics in Winnipeg, including 82 with lung cancer. Only 13% ofthe sample had been offered clinical trials, and 10% accepted.

Patients completed a symptom distress measure consisting of 13symptoms and scaled from 1 (normal or no distress) to 5 (severedistress). Survival analysis was conducted 5 years after the startof the initial data collection, at which time only 13 of the original82 lung cancer patients were alive.

Overall, symptom distress was very low in this group, with a meanscore of 23.06 out of a theoretical range of 13 to 65. The mosttroubling symptoms were fatigue, which nearly 40% of patientsscored as 3, 4, or 5; insomnia, with about 30% having high scores;and pain, with about 25% reporting distress (J Pain Symptom Manage10:423-431, 1995).

For lung cancer patients, the correlation between symptom distressand time of survival from diagnosis was -0.49
(P = .0001). Although there were patients with low symptomdistress who had both short and long survival times, virtuallyall of the patients with high symptom distress scores had shortsurvival times.

Of the three outliers (lung cancer patients with high distressclose to time of diagnosis but who were still alive at 5 years),chart review showed that their complaints involved pain in theirthoracotomy scars and/or general surgical site.

"The symptom distress scale is obviously sensitive to theeffects of postsurgical distress, and, as a result, its usefulnessin predicting survival is reduced for such patients," Drs.Degner and Sloan note.

The investigators conclude that measuring symptom distress couldhelp identify lung cancer patients with high distress who mayhave shortened survival regardless of treatment. Such knowledgecan be used to ensure that these patients receive appropriatepalliation and end-of-life counseling. Intervention studies areneeded, they say, to determine whether palliation can increasesurvival.