When a Cancer Patient Tests Positive for COVID-19
Photo: Getty Images
The COVID-19 pandemic has placed unprecedented demands on health care providers, among them the need to monitor patients remotely. Among the most vulnerable populations are cancer patients, who are at high risk of developing potentially lethal complications of COVID-19.
Our team at Memorial Sloan Kettering Cancer Center (MSK) needed to urgently develop a way to care for our patients with COVID-19 at home to detect escalating symptoms that would require immediate care. In March, the Hospital Incident Command System, which focuses on emergency planning and response, commissioned a team to fast-track a solution. Six days later we launched the COVID-19 Cohort Monitoring Program, a team and set of technologies for safely managing cancer patients with COVID-19 at home.
Team and Intervention
Building on an existing program for remotely monitoring patients on chemotherapy, the new program automatically enrolled any MSK patient testing positive for the virus. A group of physicians, nurse practitioners, physician assistants, and RNs called the COVID-19 Cohort Monitoring Team (CMMT) served as a command center team, leveraging five digital platforms to enroll patients, track their status and rapidly intervene if their symptoms worsened.
Patients in the program received an educational video to orient them and introduce the technology interface, which integrated with our existing patient portal. A subset of highest-risk patients also received pulse oximeters which, placed on a fingertip, give a reading of patients’ heart rate and blood oxygen levels. Each day, all patients receive a 10-question electronic survey that asks them to report on any COVID-19 symptoms. Those who can’t connect through the platform are called by a staff member. Questions include:
- Have you had any difficulty breathing while not moving that is new or has gotten worse?
- Have you been coughing?
- Do you feel like you have had a fever?
- What was the highest temperature you had?
- Compared to yesterday, how do you feel?
The 20% of patients with pulse oximeters also reported on their heart rate and blood oxygen reading.
The automated system alerts the care team when a patient reports mild-to-moderate symptoms (a yellow alert) or a red alert for severe symptoms, including a fever of 102 or above or an oxygen saturation below 92%. Red alerts required an immediate response. In those cases, team clinicians reach the patient by phone or video call to determine whether he or she needs urgent care, such as a trip to the ED or an immediate visit from emergency medical services. Team members actively monitor patient responses from 8 AM to 8 PM; an overnight patient support line is staffed by acute-care clinicians, providing effective 24/7 coverage.
In addition to addressing urgent-care needs, the system allows the team to track symptoms over time and monitor recovery. A dashboard also allows the leadership team to see patterns in the larger cohort such as survey completion rates, duration and severity of symptoms, and demographics.
During the period between March 26 and June 17, the CCMT enrolled 763 patients who filled out 10,044 questionnaires. The response rate for the daily questionnaire was about 53%; the other 47% received telephone assessments. Thirteen percent of patients filling out the survey triggered red alerts and required immediate intervention; 3% of those engaged by phone were referred to an ED for assessment.
By and large, patients have been highly satisfied with the program. Of 239 patients who have exited the program and completed satisfaction surveys:
- 92% would recommend this program to similar patients;
- 92% felt the time and effort to report symptoms was worth it;
- 93% of those with a pulse oximeter agreed that it made them feel more comfortable being at home;
- 90% felt connected and safe with the Covid-19 Cohort Management team (despite only interacting with them remotely); and
- 62% felt that taking part in the program helped prevent visits to the emergency room or urgent care center.
Challenges and Lessons
In rapidly launching a program with high clinical stakes, the CCMT had many successes and encountered numerous challenges. The lessons learned will help the program evolve and we hope help other institutions create similar programs for high-risk patients.
Building the program in six days required robust resourcing and efficient decision-making to eliminate obstacles. Key leaders with skills in infectious disease, informatics, and clinical operations were charged with creating the program and provided with technical and human capital resources. The program’s novelty required creating new policies and procedures, a slow and cumbersome process under normal circumstances that was streamlined by having this leadership in place and aligned at the outset, allowing rapid, high-level decision making.
While some of the clinicians had prior experience with the digital platforms involved, the need for speed meant assembling a large clinical team, many with no experience with elements of the platforms. We quickly developed training materials and sessions for these clinicians on conducting symptom interviews, remote monitoring, and virtual team communications. We began daily team video and phone briefings to address program obstacles (such as closed pharmacies and transportation and getting help for patients in the home), ensure there were no emerging patient safety issues, and, critically, to build a sense of camaraderie.
It certainly helped us move fast to be able to adapt an existing program for monitoring patient-reported symptoms from remote managing patients’ chemotherapy side-effects at home. Providers wanting to build similar new COVID-specific remote monitoring and care programs should look first within their own organizations to see if elements of existing programs and digital platforms can be adapted. Assembling a multidisciplinary team that can cut through red tape in the process will be an important enabler.
Prior to the pandemic, many of our patients connecting with us remotely had relied on an adult child or other caregiver to assist with technology. With social distancing, this support often became limited or disappeared. Thus, for approximately half of the patients in this program we had to call rather than rely on them to send symptom survey responses through the portal. Providers launching similar programs will need to evaluate the limits of technology for communicating with patients, and patients’ own limitations during the pandemic. Telephoning patients requires more staff but we found could be expedited with training. A benefit of this more traditional approach is that patients valued the human connection provided by the daily phone check-ins.