Diabetes, Hypertension
Conditions
Brief summary
The medical encounter can be overwhelming in term of the amount of information discussed, its technical nature, and the anxiety it can generate. Easy access to a secure audio recording from any internet enabled device is an available low cost technology that allows patients to revisit the visit either alone or sharing with caretakers and family. It has been introduced and tested outside the VA with evidence that it increases patient recall and understanding and may even improve physician performance. Little is known, however, about whether and to what extent these effects lead to better outcomes, such as improved treatment plan adherence and chronic disease self-management. This study is a randomized controlled trial designed ascertain whether easy access to audio recordings of the medical visit improves patients perception that they understand and can manage their own care, and leads to a variety of improved outcomes, such as better blood pressure and diabetes control, and fewer emergency department visits and hospitalizations.
Detailed description
The study aims to assess (1) the impact of an open access audio (OAA) program on two behaviors (patient activation, treatment plan adherence), and two chronic condition measures (glycosylated hemoglobin, blood pressure); (2) the impact of open access audio on provider communication and on their attention to patient contextual factors (i.e. individual Veteran's needs and circumstances relevant to planning effective care); and (3) patient, provider, and leadership perceptions of the extent to which the program is safe, not burdensome, and worthwhile at both the start and at two years into the program. A secondary analysis will descriptively measure the effect size of OAA on ED visits and hospital admissions.
Interventions
OTHEROpen Access Audio
Patient obtains secure access to an audio recording of the medical encounter post-visit that is accessible from any internet enabled device.
OTHERNo Access, Patient and Provider Aware of Recording
Patient does not obtain access to audio recording but both patient and provider are aware visit is being recorded.
OTHERNo Access, Provider Unaware of Recording
Patient does not obtain access to audio recording and provider is unaware visit is being recorded.
Sponsors
VA Office of Research and Development
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
TRIPLE (Subject, Caregiver, Outcomes Assessor)
Masking description
Arm 1 is the intervention, with both provider and patient aware of the recording. It is designed to represent how OAA would be utilized in actual practice. In Arm 2 all parties are also aware of the audio recording, but this time neither gets access to the audio after the visit (although the research team does). Arm 2 is a control for isolating the effect of patient access after the visit to the audio in Arm 1. Both patient and physician are blind to whether they are in Arm 1 or 2 until after the visit. Arm 3 is a second kind of control, as the audio is again not shared with the patient. In this arm, however, the patient conceals the audio recorder, so that the provider is unaware they are being audio recorded until after the visit.
Intervention model description
The setting will be primary care and diabetes clinics, at two facilities for generalizability. To achieve aims 1 and 2, the investigators plan a randomized controlled three arm design: (1) the encounter is recorded, with provider and patient aware, and uploaded to a server the Veteran, provider, and research team can access post visit; (2) the encounter is recorded, with both parties aware, and uploaded to a server only the research team can access; and (3) the encounter is recorded, with only the patient aware, and uploaded to a server only the research team can access. Resource utilization and disease measures indicated in aims 1 and 2 will be collected in all arms.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* All patients with scheduled appointments at the participating sites: primary care and diabetes clinics at Jesse Brown VA Medical Center and the Louis Stokes Cleveland VA Medical Center
Exclusion criteria
* None
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Contextualization of Care | Recorded visit | This outcome tests the hypothesis that physicians are more likely to contextualize the care plan when they are aware the visit is being recorded for the patient to when they are not. Contextualization of care will be rated off of the audio recording by a research assistant trained in Content Coding for Contextualization of Care (4C Coding), which determines each contextual red flag audible during the visit, whether the physician probed each red flag, each contextual factor identified by physician probe, and whether the physician addressed the contextual factor in the care plan. Outcome measure is the predicted probability of an encounter in which the care plan was contextualized (either the physician found no contextual factors present after investigating red flags, or physician found a contextual factor(s) and incorporated them into the plan), adjusted for clinic. |
| Blood Pressure | 6 months | This outcome tests the hypothesis that blood pressure control improves in patients with blood pressure \> 140/90 when they have access to an audio recording of their visit with their provider regarding blood pressure management compared to those who do not. An improvement in blood pressure is defined as a reduction of 10 or more mmHg in either the systolic or diastolic blood pressure. |
| Glycosylated Hemoglobin (HgB A1c) | 4-6 months | This outcome tests the hypothesis that HgB A1c improves in patients with Hgb A1c \> 7 when they have access to an audio recording of their discussion with their provider regarding diabetes management compared to when they do no. An improvement is defined as a decrease of 1% or more. |
| Return Visit Adherence (RVA) | From index visit to end of study data collection, an average of 512 days | This outcome tests the hypothesis that return visit adherence (RVA) improves in patients with non-adherence when they have access to an audio recording of their discussion with their provider. The outcome is the predicted proportion of visits scheduled between the time of the index visit and the end of the study that the patient attends (adjusted for clinic). For example, a value of 0.73 means that, adjusted for clinic, a patient in the given arm attends 73% of visits scheduled between their index visit and the study end date. |
| Prescription Refill Rate (Probability That Proportion Days Covered >= 80%) | 6 months | This outcome tests the hypothesis that prescription adherence improves in patients when they have access to an audio recording of their discussion with their provider compared to when they do not. Adherence is defined as having at least 80% proportion days covered (PDF). PDC is the total number of days covered by refills in a measurement period divided by the number of days between the first fill and the end of the measurement period. Non-adherence is defined as a PDC\<80% during the 6 months prior to the audio recorded visit. The outcome is the probability that a patient's PDC is at least 80% following their participation, adjusted for clinic. |
| Patient Activation Measure (PAM) | 2 weeks | This outcome tests the hypothesis that patient activation is higher in patients when they have access to an audio recording of their visit with their provider compared to when they do not. Patients are called by phone post visit to answer PAM 10, a 10 item questionnaire with a 5 point Likert response that scores their perceived knowledge, skills, and confident in their capacity to follow their treatment plan with scores from 0-100. A higher score indicates higher patient activation. |
| SEGUE Framework for Evaluating and Scoring Communication Behavior. | Recorded visit | This outcome tests the hypothesis that physicians communicate more effectively when they are aware the visit is being recorded for the patient to when they are not. Communication behavior will be rated off the audio recording by a research assistant utilizing the SEGUE checklist. Outcome measure is total proportion achieved score on 32 item SEGUE instrument, ranging from 0 to 1 with a higher score indicating more effective communication. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Emergency Department Utilization | 6 months | This outcome tests the hypothesis that emergency department visit rates are lower following encounters in which patients have access to an audio recording of their visit with their provider compared to when they do not. Outcome measure is the average number of ED visits made by the patient in the 6 months following the audio recorded visit, adjusted for clinic. |
| Hospital Admission Rate | 6 months | This outcome tests the hypothesis that inpatient admission rates are lower following encounters in which patients have access to an audio recording of their visit with their provider compared to when they do not. Outcome measure is the average number of inpatient hospital admissions of the patient (of any duration) during the 6 month period following the audio recorded visit. |
| Glycated Hemoglobin (Access vs. Non-access) | Six month | Comparison of HgA1c among Arm 1 patients who did and did not access their visit recording |
| ED Visit Rate (Access vs. Non-access) | Six month | Comparison of ED visit rate among Arm 1 patients who did and did not access their visit recording. The ED visit rate is the number of emergency department visits in the 6 months following the patient's recorded visit, adjusted for clinic. |
| Return Visit Attendance (Access vs. Non-access) | From recorded visit to study end date, an average of 512 days | Comparison of RVA among Arm 1 patients who did and did not access their visit recording. The outcome is the predicted proportion of visits scheduled between the time of the index visit and the end of the study that the patient attends (adjusted for clinic). For example, a value of 0.73 means that, adjusted for clinic, a patient in the given arm attends 73% of visits scheduled between their index visit and the study end date. |
| Probability That Proportion of Days Covered by Prescription >= 80% (Access vs. Non-access) | 6 months | Comparison of probability of prescription refill adherence (PDC\>=80%) among Arm 1 patients who did and did not access their visit recording. Adherence is defined as having at least 80% proportion days covered (PDF). PDC is the total number of days covered by refills in a measurement period divided by the number of days between the first fill and the end of the measurement period. Non-adherence is defined as a PDC\<80% during the 6 months prior to the audio recorded visit. The outcome is the probability that a patient's PDC is at least 80% following their participation, adjusted for clinic. |
| Blood Pressure (Access vs. Non-access) | Six month | Comparison of blood pressure among Arm 1 patients who did and did not access their visit recording |
Countries
United States
Participant flow
Pre-assignment details
Although providers are also engaged as part of the intervention, all trial enrollment, randomization, assignment, and analysis is by patient visit. It is only the patient's assignment that affects which study arm a visit with a provider falls into. Accordingly, providers are not considered participants enrolled for purposes of study result reporting.
Participants by arm
| Arm | Count |
|---|---|
| OAA Intervention Visit recorded, both patient and provider aware, both patient and provider have access to audio post-visit
Open Access Audio: Patient obtains secure access to an audio recording of the medical encounter post-visit that is accessible from any internet enabled device. | 691 |
| OAA Physician Aware Control Visit recorded, both patient and provider aware, neither patient nor provider have access to audio post-visit
No Access, Patient and Provider Aware of Recording: Patient does not obtain access to audio recording but both patient and provider are aware visit is being recorded. | 694 |
| OAA Physician Unaware Control Visit recorded, patient aware but provider unaware, neither patient nor provider have access to audio post-visit
No Access, Provider Unaware of Recording: Patient does not obtain access to audio recording and provider is unaware visit is being recorded. | 168 |
| Total | 1,553 |
Baseline characteristics
| Characteristic | OAA Physician Aware Control | OAA Physician Unaware Control | OAA Intervention | Total |
|---|---|---|---|---|
| Age, Continuous | 61.9 years STANDARD_DEVIATION 11.6 | 61.3 years STANDARD_DEVIATION 10.8 | 62.8 years STANDARD_DEVIATION 10.8 | 62.2 years STANDARD_DEVIATION 11.2 |
| Clinical setting Diabetes clinic | 146 Participants | 0 Participants | 146 Participants | 292 Participants |
| Clinical setting Primary care clinic | 548 Participants | 168 Participants | 545 Participants | 1261 Participants |
| Ethnicity (NIH/OMB) Hispanic or Latino | 27 Participants | 7 Participants | 25 Participants | 59 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 649 Participants | 158 Participants | 639 Participants | 1446 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 18 Participants | 3 Participants | 27 Participants | 48 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 4 Participants | 3 Participants | 2 Participants | 9 Participants |
| Race (NIH/OMB) Asian | 5 Participants | 1 Participants | 1 Participants | 7 Participants |
| Race (NIH/OMB) Black or African American | 504 Participants | 132 Participants | 517 Participants | 1153 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 8 Participants | 0 Participants | 6 Participants | 14 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 22 Participants | 8 Participants | 25 Participants | 55 Participants |
| Race (NIH/OMB) White | 151 Participants | 24 Participants | 140 Participants | 315 Participants |
| Sex: Female, Male Female | 35 Participants | 3 Participants | 34 Participants | 72 Participants |
| Sex: Female, Male Male | 659 Participants | 165 Participants | 657 Participants | 1481 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | 0 / 691 | 0 / 694 | 0 / 168 |
| other Total, other adverse events | 0 / 691 | 0 / 694 | 0 / 168 |
| serious Total, serious adverse events | 0 / 691 | 0 / 694 | 0 / 168 |
Outcome results
Primary
Blood Pressure
This outcome tests the hypothesis that blood pressure control improves in patients with blood pressure \> 140/90 when they have access to an audio recording of their visit with their provider regarding blood pressure management compared to those who do not. An improvement in blood pressure is defined as a reduction of 10 or more mmHg in either the systolic or diastolic blood pressure.
Time frame: 6 months
| Arm | Measure | Group | Value (MEAN) |
|---|---|---|---|
| OAA Intervention | Blood Pressure | Systolic | 131.6 mmHg |
| OAA Intervention | Blood Pressure | Diastolic | 78.2 mmHg |
| OAA Physician Aware Control | Blood Pressure | Systolic | 131.8 mmHg |
| OAA Physician Aware Control | Blood Pressure | Diastolic | 79.5 mmHg |
Comparison: Diastolic BPp-value: 0.06Regression, Linear
Comparison: Systolic BPp-value: 0.9Regression, Linear
Primary
Contextualization of Care
This outcome tests the hypothesis that physicians are more likely to contextualize the care plan when they are aware the visit is being recorded for the patient to when they are not. Contextualization of care will be rated off of the audio recording by a research assistant trained in Content Coding for Contextualization of Care (4C Coding), which determines each contextual red flag audible during the visit, whether the physician probed each red flag, each contextual factor identified by physician probe, and whether the physician addressed the contextual factor in the care plan. Outcome measure is the predicted probability of an encounter in which the care plan was contextualized (either the physician found no contextual factors present after investigating red flags, or physician found a contextual factor(s) and incorporated them into the plan), adjusted for clinic.
Time frame: Recorded visit
Population: Arm 2 random subsample vs. Arm 3, excluding visits with recorder failure or on phone
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| OAA Intervention | Contextualization of Care | 0.51 Predicted probability of contextual plan |
| OAA Physician Aware Control | Contextualization of Care | 0.41 Predicted probability of contextual plan |
p-value: 0.01Regression, Logistic
Primary
Glycosylated Hemoglobin (HgB A1c)
This outcome tests the hypothesis that HgB A1c improves in patients with Hgb A1c \> 7 when they have access to an audio recording of their discussion with their provider regarding diabetes management compared to when they do no. An improvement is defined as a decrease of 1% or more.
Time frame: 4-6 months
| Arm | Measure | Value (MEAN) |
|---|---|---|
| OAA Intervention | Glycosylated Hemoglobin (HgB A1c) | 6.95 percent |
| OAA Physician Aware Control | Glycosylated Hemoglobin (HgB A1c) | 6.83 percent |
Primary
Patient Activation Measure (PAM)
This outcome tests the hypothesis that patient activation is higher in patients when they have access to an audio recording of their visit with their provider compared to when they do not. Patients are called by phone post visit to answer PAM 10, a 10 item questionnaire with a 5 point Likert response that scores their perceived knowledge, skills, and confident in their capacity to follow their treatment plan with scores from 0-100. A higher score indicates higher patient activation.
Time frame: 2 weeks
Population: Random samples from arms 1 and 2
| Arm | Measure | Value (MEAN) |
|---|---|---|
| OAA Intervention | Patient Activation Measure (PAM) | 9.44 score on a scale |
| OAA Physician Aware Control | Patient Activation Measure (PAM) | 9.54 score on a scale |
p-value: 0.44Regression, Linear
Primary
Prescription Refill Rate (Probability That Proportion Days Covered >= 80%)
This outcome tests the hypothesis that prescription adherence improves in patients when they have access to an audio recording of their discussion with their provider compared to when they do not. Adherence is defined as having at least 80% proportion days covered (PDF). PDC is the total number of days covered by refills in a measurement period divided by the number of days between the first fill and the end of the measurement period. Non-adherence is defined as a PDC\<80% during the 6 months prior to the audio recorded visit. The outcome is the probability that a patient's PDC is at least 80% following their participation, adjusted for clinic.
Time frame: 6 months
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| OAA Intervention | Prescription Refill Rate (Probability That Proportion Days Covered >= 80%) | 0.05 Probability of >=80% PDC |
| OAA Physician Aware Control | Prescription Refill Rate (Probability That Proportion Days Covered >= 80%) | 0.04 Probability of >=80% PDC |
p-value: 0.1Regression, Logistic
Primary
Return Visit Adherence (RVA)
This outcome tests the hypothesis that return visit adherence (RVA) improves in patients with non-adherence when they have access to an audio recording of their discussion with their provider. The outcome is the predicted proportion of visits scheduled between the time of the index visit and the end of the study that the patient attends (adjusted for clinic). For example, a value of 0.73 means that, adjusted for clinic, a patient in the given arm attends 73% of visits scheduled between their index visit and the study end date.
Time frame: From index visit to end of study data collection, an average of 512 days
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| OAA Intervention | Return Visit Adherence (RVA) | 0.733 Proportion of future visits attended |
| OAA Physician Aware Control | Return Visit Adherence (RVA) | 0.732 Proportion of future visits attended |
p-value: 0.29Regression, Linear
Primary
SEGUE Framework for Evaluating and Scoring Communication Behavior.
This outcome tests the hypothesis that physicians communicate more effectively when they are aware the visit is being recorded for the patient to when they are not. Communication behavior will be rated off the audio recording by a research assistant utilizing the SEGUE checklist. Outcome measure is total proportion achieved score on 32 item SEGUE instrument, ranging from 0 to 1 with a higher score indicating more effective communication.
Time frame: Recorded visit
Population: Arm 2 random subsample compared with Arm 3, excluding visits with no recordings due to recorder failure.
| Arm | Measure | Value (MEAN) |
|---|---|---|
| OAA Intervention | SEGUE Framework for Evaluating and Scoring Communication Behavior. | 0.85 score on a scale |
| OAA Physician Aware Control | SEGUE Framework for Evaluating and Scoring Communication Behavior. | 0.82 score on a scale |
p-value: 0.01Regression, Linear
Secondary
Blood Pressure (Access vs. Non-access)
Comparison of blood pressure among Arm 1 patients who did and did not access their visit recording
Time frame: Six month
| Arm | Measure | Group | Value (MEAN) |
|---|---|---|---|
| OAA Intervention | Blood Pressure (Access vs. Non-access) | Systolic | 130.7 mmHg |
| OAA Intervention | Blood Pressure (Access vs. Non-access) | Diastolic | 78.0 mmHg |
| OAA Physician Aware Control | Blood Pressure (Access vs. Non-access) | Systolic | 133.0 mmHg |
| OAA Physician Aware Control | Blood Pressure (Access vs. Non-access) | Diastolic | 79.5 mmHg |
| OAA Physician Aware Control | Blood Pressure (Access vs. Non-access) | Systolic | 132.7 mmHg |
| OAA Physician Aware Control | Blood Pressure (Access vs. Non-access) | Diastolic | 80.5 mmHg |
Comparison: Systolic BPp-value: 0.2Regression, Linear
Comparison: Systolic BPp-value: 0.82Regression, Linear
Comparison: Diastolic BPp-value: 0.12Regression, Linear
Comparison: Diastolic BPp-value: 0.1Regression, Linear
Secondary
ED Visit Rate (Access vs. Non-access)
Comparison of ED visit rate among Arm 1 patients who did and did not access their visit recording. The ED visit rate is the number of emergency department visits in the 6 months following the patient's recorded visit, adjusted for clinic.
Time frame: Six month
| Arm | Measure | Value (MEAN) |
|---|---|---|
| OAA Intervention | ED Visit Rate (Access vs. Non-access) | 0.69 ED visits |
| OAA Physician Aware Control | ED Visit Rate (Access vs. Non-access) | 1.05 ED visits |
| OAA Physician Aware Control | ED Visit Rate (Access vs. Non-access) | 0.93 ED visits |
p-value: <0.001Regression, negative binomial
p-value: 0.03Regression, negative binomial
Secondary
Emergency Department Utilization
This outcome tests the hypothesis that emergency department visit rates are lower following encounters in which patients have access to an audio recording of their visit with their provider compared to when they do not. Outcome measure is the average number of ED visits made by the patient in the 6 months following the audio recorded visit, adjusted for clinic.
Time frame: 6 months
| Arm | Measure | Value (MEAN) |
|---|---|---|
| OAA Intervention | Emergency Department Utilization | 1.03 ED visits |
| OAA Physician Aware Control | Emergency Department Utilization | 0.96 ED visits |
p-value: 0.54Regression, negative binomial
Secondary
Glycated Hemoglobin (Access vs. Non-access)
Comparison of HgA1c among Arm 1 patients who did and did not access their visit recording
Time frame: Six month
| Arm | Measure | Value (MEAN) |
|---|---|---|
| OAA Intervention | Glycated Hemoglobin (Access vs. Non-access) | 6.53 percent |
| OAA Physician Aware Control | Glycated Hemoglobin (Access vs. Non-access) | 6.95 percent |
| OAA Physician Aware Control | Glycated Hemoglobin (Access vs. Non-access) | 6.73 percent |
p-value: 0.03Regression, Linear
p-value: 0.06Regression, Linear
Secondary
Hospital Admission Rate
This outcome tests the hypothesis that inpatient admission rates are lower following encounters in which patients have access to an audio recording of their visit with their provider compared to when they do not. Outcome measure is the average number of inpatient hospital admissions of the patient (of any duration) during the 6 month period following the audio recorded visit.
Time frame: 6 months
| Arm | Measure | Value (MEAN) |
|---|---|---|
| OAA Intervention | Hospital Admission Rate | 0.26 Hospital admissions |
| OAA Physician Aware Control | Hospital Admission Rate | 0.28 Hospital admissions |
p-value: 0.74Regression, negative binomial
Secondary
Probability That Proportion of Days Covered by Prescription >= 80% (Access vs. Non-access)
Comparison of probability of prescription refill adherence (PDC\>=80%) among Arm 1 patients who did and did not access their visit recording. Adherence is defined as having at least 80% proportion days covered (PDF). PDC is the total number of days covered by refills in a measurement period divided by the number of days between the first fill and the end of the measurement period. Non-adherence is defined as a PDC\<80% during the 6 months prior to the audio recorded visit. The outcome is the probability that a patient's PDC is at least 80% following their participation, adjusted for clinic.
Time frame: 6 months
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| OAA Intervention | Probability That Proportion of Days Covered by Prescription >= 80% (Access vs. Non-access) | 0.09 Probability of at least 80% PDC |
| OAA Physician Aware Control | Probability That Proportion of Days Covered by Prescription >= 80% (Access vs. Non-access) | 0.04 Probability of at least 80% PDC |
| OAA Physician Aware Control | Probability That Proportion of Days Covered by Prescription >= 80% (Access vs. Non-access) | 0.05 Probability of at least 80% PDC |
p-value: <0.001Mixed Models Analysis
p-value: 0.98Mixed Models Analysis
Secondary
Return Visit Attendance (Access vs. Non-access)
Comparison of RVA among Arm 1 patients who did and did not access their visit recording. The outcome is the predicted proportion of visits scheduled between the time of the index visit and the end of the study that the patient attends (adjusted for clinic). For example, a value of 0.73 means that, adjusted for clinic, a patient in the given arm attends 73% of visits scheduled between their index visit and the study end date.
Time frame: From recorded visit to study end date, an average of 512 days
Population: Patients in OAA intervention arm (accessed and not access) exclude 11 patients who did not have the opportunity to access a recording due to recorder failure.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| OAA Intervention | Return Visit Attendance (Access vs. Non-access) | 0.78 Proportion of return visits attended |
| OAA Physician Aware Control | Return Visit Attendance (Access vs. Non-access) | 0.71 Proportion of return visits attended |
| OAA Physician Aware Control | Return Visit Attendance (Access vs. Non-access) | 0.73 Proportion of return visits attended |
p-value: <0.001Regression, poisson
p-value: 0.08Regression, poisson