Clinical Decision Support Algorithm Leads to Symptom Improvements

Larry Hand

October 29, 2014

BOSTON — A specially developed clinical decision support intervention based on analysis of patient-reported symptoms has led to significant improvement in patient well-being compared with usual care for patients with cancer, according to a feasibility study presented here at the Palliative Care in Oncology Symposium.

Led by Mary E. Cooley, PhD, RN, a nurse scientist the Dana-Farber Cancer Institute and a lecturer on psychiatry at Harvard Medical School, Boston, Massachusetts, the study enrolled 20 physicians and their patients and randomly assigned them to usual care or the intervention, called symptom assessment and management intervention (SAMI).

The study targeted five symptoms: pain, fatigue, depression, anxiety, and dyspnea. An expert panel analyzed guidelines for treating the target symptoms and developed a basic report that researchers used to tailor information and recommendation to give to treating physicians for each study patient.

During the study, 179 patients completed a Web-based symptom assessment in the waiting room before each visit for 6 months. Research staff then added laboratory values and medications to the information. The researchers then used SAMI to develop individual SAMI reports for clinicians.

Quick to Use

"The symptom assessment tool completed by patients took approximately 10 minutes to complete, it took less than 2 seconds to generate the report and 1 minute for clinicians to use the intervention," Dr Cooley told Medscape Medical News. "The first page of the report has specific suggestions that would optimize the patient's care, and the second page shows their symptom trajectory over time."

"The intervention consisted of guideline-based care and was translated into computer-based algorithms to provide tailored suggestions to enhance management of pain, depression, anxiety, fatigue, and dyspnea," she added. "These suggestions consisted of supportive care medications or referrals. The clinician had the report before the visit and received information about patients' symptoms and suggestions to improve symptom management."

The patients had a mean age of 63; 58% were female and 88% were white; and 32% had less than a high school education.

Researchers evaluated effects of the trial using questionnaires at baseline, and at 2, 4, and 6 months. They used the treatment outcome index (TOI) to assess health-related quality of life.

Patients in the intervention group reported significant improvements physical well-being (P = .007) as well as a clinically significant different in the TOI (62 vs 68) at 4 months compared with the usual care (UC) group.

Compared with the UC group, the SAMI group had higher odds ratios (ORs) for managing depression (OR, 1.6; 95% confidence interval [CI], 1.0 - 2.5), anxiety (OR, 1.7; 95% CI, 1.0 - 3.0), and fatigue (OR, 1.6; 95% CI, 1.1 - 2.5). The odds of a patient seeking palliative care for pain also appeared higher in the SAMI group (OR, 3.2; 95% CI, 0.7 - 13.4).

This helps to drive the conversation. Dr Mary E. Cooley

"When trying to integrate symptomatics into the work flow of providers, this is one easy way to do it," Cooley said. "This helps to drive the conversation."

Practical Applicability

"I think Mary's study and the technology that she's developed is an important advance," Carrie T. Stricker, PhD, CRNP, a nurse practitioner at Penn Medicine at the University of Pennsylvania in Philadelphia, told Medscape Medical News. She also is chief clinical officer and cofounder of OnQ Health, a company that collects and analyzes patient-reported and clinical data to devise plans for advising clinical oncologists on initial, supportive, and survivorship care.

"We know that there is a very real challenge in attending to patient symptoms and supportive care needs during oncology treatment and care," Dr Stricker continued. "We as clinicians chronically underestimate the symptoms and our patients are chronically having more trouble than we realize."

She added that the intervention is intended to address gaps by "engaging patients in reporting their symptoms, but not just leaving it there."

"I do think that it has practical applicability. There are big knowledge gaps in what are the appropriate strategies and agents, and I think the findings of this study represent that more work needs to be done in terms of figuring out how to integrate this more effectively into the clinic," she said.

Dr Cooley emphasized that the study was a feasibility study and that an effectiveness study now needs to be done. The research group has published one study using the algorithm decision support tool in outpatient thoracic oncology and another on lung cancer.

Dr Cooley has disclosed no relevant financial relationships. One coauthor reported receiving, through his institution, research funding from GlaxoSmithKline; one coauthor has reported receiving research funding from ImClone Systems, and another coauthor has reported owning stock in Fidelity Investments. Dr Stricker said that other than her position with OnQ Health, she had no relevant financial relationships.

Palliative Care in Oncology Symposium. Abstract 1. Presented October 24, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.