Patient-reported symptom monitoring during chemotherapy

Between September 2007 and January 2011, Ethan Basch and colleagues at Memorial Sloan Kettering Cancer Center randomly assigned 766 patients receiving chemotherapy for metastatic breast, lung, genitourinary or gynecologic cancers either to usual care or to reporting their own symptoms.

Patients in the self-reporting group rated 12 common symptoms on a five-point scale through a web-based system called STAR, on tablet computers at clinic visits and, if they had home computers, from home between visits, prompted by weekly email reminders. A severe or worsening symptom set off an automatic email alert to the patient's nurse.

The first results appeared in the Journal of Clinical Oncology in December 2015. Health-related quality of life improved in 34 percent of the self-reporting group against 18 percent with usual care, and fewer patients went to the emergency room (34 against 41 percent). At one year, 75 percent of the self-reporting group were alive, against 69 percent, and mean quality-adjusted survival was 8.7 against 8.0 months. Some benefits appeared greater among patients without computer experience, who were older and frailer.

Survival was analyzed in June 2016 using the National Death Index, after 517 of the 766 patients had died and at a median follow-up of seven years. Median overall survival was 31.2 months with self-reporting and 26.0 months with usual care (hazard ratio 0.83). Patients in the self-reporting group also stayed on chemotherapy longer, 8.2 against 6.3 months. Basch presented the survival result in a plenary session of the American Society of Clinical Oncology annual meeting on 4 June 2017, and JAMA published it online the same day.