Cancer
Conditions
Brief summary
The main purpose of this research study is to evaluate the implementation and effectiveness of patient- and nurse-directed strategies across in-clinic and remote patient reported outcome (PRO) monitoring settings, using a three-arm pragmatic cluster randomized controlled trial. Additionally, the goal is to evaluate moderators of implementation effects on PRO monitoring. Eligible patients will be randomized independently to: (1) usual practice (i.e., encounter-based PRO administration via patient portal or tablet); (2) encounter-based PRO monitoring with patient reminders and nurse alerts; or (3) remote PRO monitoring with patient reminders and nurse alerts. The investigators hypothesize that nudges to patients and alerts to nurses will improve patient-level PRO completion and clinician-level PRO engagement.
Detailed description
Routine PRO monitoring is an evidence-based practice that improves patient outcomes. Patients with cancer commonly experience symptoms which go unrecognized by clinical teams up to half of the time. PROs are a powerful tool to bridge this gap, amplifying the patient experience and improving communication between patients and clinicians. A seminal clinical trial was conducted in which weekly PRO symptom assessments with care team alerts led to reduced care utilization, improved quality of life, and lengthened overall survival among patients receiving active treatment for advanced solid tumors. Subsequent randomized trials have corroborated the quality of life and survival advantages of PRO monitoring in other cancer populations and practice settings. Real-world adherence to PRO monitoring is lower than in seminal clinical trials, with potential racial/ethnic disparities. Multilevel barriers to PRO implementation exist at the patient, clinician, practice, and system levels. While patient adherence to PRO monitoring in seminal trials ranged from 70 to 90 percent, real-world adherence is much lower and more variable. Our preliminary research demonstrates that in a large, multi-site rollout of PRO monitoring, mean patient-level adherence was less than 50 percent. Moreover, compared to non-Hispanic White patients, Black and Hispanic patients were 9% and 4% less likely to complete PRO questionnaires, respectively, but more likely to report more severe symptoms. These results highlight that historically marginalized subgroups may benefit less from current methods of PRO administration, potentially reinforcing existing disparities in symptom management if not remediated. This will be the first study to apply behavioral economic strategies to align clinicians and patients in achieving sustained patient-level PRO adherence, and addresses limitations of current PRO approaches. Current approaches to PRO collection are limited by: (1) suboptimal clinician engagement, (2) encounter-based assessments which preclude longitudinal monitoring of PROs outside of the clinic, and (3) insufficient automated effectors actively linking reported symptoms with clinical response. This proposal addresses these limitations by employing behavioral strategies directly targeting clinician engagement, utilizing novel electronic remote monitoring methods to maximize real-time PRO capture, and linking these with pathway-driven automated effectors.
Interventions
OTHEREncounter-based PRO monitoring
This arm consists of usual practice plus the addition of patient reminders to complete PRO questionnaires and triage nurse alerts for severe symptoms. Patient reminders will be operationalized through the Epic patient portal. Triage nurse alerts will be routed to Epic symptom management pools in response to a patient reporting moderate or severe symptoms. Patients will have the opportunity to decline or opt-out of triage nurse support. Clinical interventions stemming from triage nurse alerts will be up to the discretion of clinical teams - i.e., clinical responses will be neither prescriptive nor mandatory. Notably, the entry point for triage nurse alerts (i.e., Epic symptom management pools), as well as resultant clinical responses, are existing standard operating procedures/practices for patients reporting symptoms by phone.
OTHERRemote PRO monitoring
This arm will consist of weekly PRO questionnaires administered via the patient portal, de-linked from clinical encounters. Patient reminders will be operationalized through the Epic patient portal. Triage nurse alerts will be routed to Epic symptom management pools in response to a patient reporting moderate or severe symptoms. Patients will have the opportunity to decline or opt-out of triage nurse support. Clinical interventions stemming from triage nurse alerts will be up to the discretion of clinical teams - i.e., clinical responses will be neither prescriptive nor mandatory. Notably, the entry point for triage nurse alerts (i.e., Epic symptom management pools), as well as resultant clinical responses, are existing standard operating procedures/practices for patients reporting symptoms by phone.
Sponsors
National Cancer Institute (NCI)
Abramson Cancer Center at Penn Medicine
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE (Investigator)
Intervention model description
Eligible patients will be identified pragmatically and will be independently randomized to study arms using a three-arm design.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Patients must initiate a new line of systemic cancer therapy (i.e., intravenous chemotherapy or immunotherapy) at a participating Implementation Lab site
Exclusion criteria
\- Patient opts out of pragmatic research
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| PRO Adherence | 3 months | The primary outcome will be PRO adherence, measured at the patient level as the proportion of expected PRO questionnaires completed per patient. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Percent of Patients With at Least One Note Documenting PROs Per Month During Study Period | 3 months | — |
| Time to Alert Response (for Arms 2 and 3 Only) | 3 months | — |
| Percent of Patients Who Trigger an Alert for Triage Nurses (for Arms 2 and 3 Only) | 3 months | — |
| Percent of Patients With at Least One Note Documenting PROs During Study Period | 3 months | — |
| Treatment Modifications | 3 months | Percent of patients with a treatment modification |
| Duration of Therapy | 3 months | Average duration of therapy during the study period |
| Acute Care Utilization | 3 months | Percent of patients with an emergency department visit or hospitalization |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Usual Practice This arm will consist of encounter-based PRO monitoring, in which patients have an opportunity to complete PRO questionnaires via patient portal in advance of clinical encounters or via tablet during clinical encounters. While clinicians will be encouraged to view and discuss PROs with patients during clinical encounters, they will not be prompted to do so in real-time, nor will there be alerts for escalating symptoms. | 171 |
| Encounter-based PRO Monitoring Encounter-based PRO monitoring (usual practice) plus patient reminders and triage nurse alerts
Encounter-based PRO monitoring: This arm consists of usual practice plus the addition of patient reminders to complete PRO questionnaires and triage nurse alerts for severe symptoms. Patient reminders will be operationalized through the Epic patient portal. Triage nurse alerts will be routed to Epic symptom management pools in response to a patient reporting moderate or severe symptoms. Patients will have the opportunity to decline or opt-out of triage nurse support. Clinical interventions stemming from triage nurse alerts will be up to the discretion of clinical teams - i.e., clinical responses will be neither prescriptive nor mandatory. Notably, the entry point for triage nurse alerts (i.e., Epic symptom management pools), as well as resultant clinical responses, are existing standard operating procedures/practices for patients reporting symptoms by phone. | 144 |
| Remote PRO Monitoring Remote PRO monitoring plus patient reminders and triage nurse alerts
Remote PRO monitoring: This arm will consist of weekly PRO questionnaires administered via the patient portal, de-linked from clinical encounters. Patient reminders will be operationalized through the Epic patient portal. Triage nurse alerts will be routed to Epic symptom management pools in response to a patient reporting moderate or severe symptoms. Patients will have the opportunity to decline or opt-out of triage nurse support. Clinical interventions stemming from triage nurse alerts will be up to the discretion of clinical teams - i.e., clinical responses will be neither prescriptive nor mandatory. Notably, the entry point for triage nurse alerts (i.e., Epic symptom management pools), as well as resultant clinical responses, are existing standard operating procedures/practices for patients reporting symptoms by phone. | 161 |
| Total | 476 |
Baseline characteristics
| Characteristic | Usual Practice | Encounter-based PRO Monitoring | Remote PRO Monitoring | Total |
|---|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 102 Participants | 93 Participants | 97 Participants | 292 Participants |
| Age, Categorical Between 18 and 65 years | 69 Participants | 51 Participants | 64 Participants | 184 Participants |
| Ethnicity (NIH/OMB) Hispanic or Latino | 161 Participants | 135 Participants | 154 Participants | 450 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 4 Participants | 6 Participants | 2 Participants | 12 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 6 Participants | 3 Participants | 5 Participants | 14 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 1 Participants | 0 Participants | 0 Participants | 1 Participants |
| Race (NIH/OMB) Asian | 5 Participants | 7 Participants | 10 Participants | 22 Participants |
| Race (NIH/OMB) Black or African American | 44 Participants | 27 Participants | 36 Participants | 107 Participants |
| Race (NIH/OMB) More than one race | 1 Participants | 3 Participants | 0 Participants | 4 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 1 Participants | 0 Participants | 1 Participants | 2 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 9 Participants | 5 Participants | 10 Participants | 24 Participants |
| Race (NIH/OMB) White | 110 Participants | 102 Participants | 104 Participants | 316 Participants |
| Region of Enrollment United States | 171 participants | 144 participants | 161 participants | 476 participants |
| Sex: Female, Male Female | 66 Participants | 58 Participants | 55 Participants | 179 Participants |
| Sex: Female, Male Male | 105 Participants | 86 Participants | 106 Participants | 297 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | 0 / 171 | 0 / 144 | 0 / 161 |
| other Total, other adverse events | 0 / 171 | 0 / 144 | 0 / 161 |
| serious Total, serious adverse events | 0 / 171 | 0 / 144 | 0 / 161 |
Outcome results
Primary
PRO Adherence
The primary outcome will be PRO adherence, measured at the patient level as the proportion of expected PRO questionnaires completed per patient.
Time frame: 3 months
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Usual practice | PRO Adherence | 50.0 pct. assigned questionnaires completed |
| Encounter-based PRO monitoring | PRO Adherence | 56.9 pct. assigned questionnaires completed |
| Remote PRO monitoring | PRO Adherence | 34.9 pct. assigned questionnaires completed |
Secondary
Acute Care Utilization
Percent of patients with an emergency department visit or hospitalization
Time frame: 3 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual practice | Acute Care Utilization | 48 Participants |
| Encounter-based PRO monitoring | Acute Care Utilization | 45 Participants |
| Remote PRO monitoring | Acute Care Utilization | 51 Participants |
Secondary
Duration of Therapy
Average duration of therapy during the study period
Time frame: 3 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual practice | Duration of Therapy | 56.82 days | Standard Deviation 31.67 |
| Encounter-based PRO monitoring | Duration of Therapy | 55.94 days | Standard Deviation 33.91 |
| Remote PRO monitoring | Duration of Therapy | 53.66 days | Standard Deviation 32.62 |
Secondary
Percent of Patients Who Trigger an Alert for Triage Nurses (for Arms 2 and 3 Only)
Time frame: 3 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual practice | Percent of Patients Who Trigger an Alert for Triage Nurses (for Arms 2 and 3 Only) | 71 Participants |
| Encounter-based PRO monitoring | Percent of Patients Who Trigger an Alert for Triage Nurses (for Arms 2 and 3 Only) | 84 Participants |
Secondary
Percent of Patients With at Least One Note Documenting PROs During Study Period
Time frame: 3 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual practice | Percent of Patients With at Least One Note Documenting PROs During Study Period | 67 Participants |
| Encounter-based PRO monitoring | Percent of Patients With at Least One Note Documenting PROs During Study Period | 82 Participants |
| Remote PRO monitoring | Percent of Patients With at Least One Note Documenting PROs During Study Period | 96 Participants |
Secondary
Percent of Patients With at Least One Note Documenting PROs Per Month During Study Period
Time frame: 3 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual practice | Percent of Patients With at Least One Note Documenting PROs Per Month During Study Period | 20 Participants |
| Encounter-based PRO monitoring | Percent of Patients With at Least One Note Documenting PROs Per Month During Study Period | 16 Participants |
| Remote PRO monitoring | Percent of Patients With at Least One Note Documenting PROs Per Month During Study Period | 32 Participants |
Secondary
Time to Alert Response (for Arms 2 and 3 Only)
Time frame: 3 months
Population: Clinical teams documented responses to patients' questionnaire submissions within the electronic health record.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual practice | Time to Alert Response (for Arms 2 and 3 Only) | 1.24 days | Standard Deviation 1.06 |
| Encounter-based PRO monitoring | Time to Alert Response (for Arms 2 and 3 Only) | 1.20 days | Standard Deviation 1.01 |
Secondary
Treatment Modifications
Percent of patients with a treatment modification
Time frame: 3 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual practice | Treatment Modifications | 43 Participants |
| Encounter-based PRO monitoring | Treatment Modifications | 35 Participants |
| Remote PRO monitoring | Treatment Modifications | 29 Participants |