Electronic Health Records
Conditions
Keywords
Medication Therapy Management, Clinical Decision Support Systems, Medical Informatics, Patient Care Management, United States Department of Veterans Affairs
Brief summary
The purpose of this study is to advance the science of healthcare informatics and to improve medication management through the development of a new approach to the electronic medical record called the Integrated Medication Manager (IMM).
Detailed description
In an attempt to address problems patient non-compliance with quality goals barriers to access and integration of health information that impede achievement of treatment goals, the VA is developing a new approach to the electronic medical record. The VA is moving away from the paper-chart metaphor and towards an integrated representation of the patient's status and care process across time. One of the first steps in the development phase has been to explicitly relate patient conditions, therapies, and goals in the domain of pharmacotherapy. This is called Integrated Medication Management and draws on Hollnagel's Contextual Control Model. Providers will be able to plan care and create orders directly in the context of these explicit relationships. This application will be implemented nationwide through a web interface embedded within the existing Computerized Patient Record System (CPRS), the graphical user interface to VA Information Systems (VistA). Aim 1: Identify cognitive components of providers' therapeutic decision making in the field. Aim 2. Refine and evaluate the Integrated Medication Manager using simulation studies. * Aim 2.a. Refine interfaces and logic of the Integrated Medication Manager. * Aim 2.b. Compare the performance of the Integrated Medication Manager and usual CPRS. All hypotheses (below) test the use of IMM versus usual electronic medical record (EMR). * Speed of decision-making will be faster. * Accuracy of data interpretation (clinical assessment) will be higher. * Appropriateness of therapeutic plans will be higher. * Efficiency of gathering information will be higher. * Common ground measures will be higher. * Appropriateness of therapeutic plans will be higher when relevant data is outside the usual time horizon. * Appropriateness of therapeutic plans will be higher when complex associations among patient therapies and goals exist. * Appropriateness of therapeutic plans will be no lower when relevant data is not captured by the displays of the IMM. * Appropriateness of therapeutic plans will be higher when highly salient data is not germane to the most important problem. * Appropriateness of therapeutic plans will be higher when cognitive load is high due to interruptions.
Interventions
OTHERIntegrated Medication Manager
A theory based electronic health record. Half of the provider participants were assigned the IMM to use. The other half were assigned the VA's CPRS EHR to use for the simulation. Providers were randomly assigned to a EHR to use.
Sponsors
Agency for Healthcare Research and Quality (AHRQ)
University of Utah
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Practiced in primary care for at least two years * Third year residents with two years of residency in internal medicine or family practice * Do not have to be currently practicing
Exclusion criteria
* None
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Amount of Time to Complete Assessment and Plan | 10 minutes | Each participant had 10 minutes maximum to review the patient case and write an Assessment and Plan. |
| Accuracy of Written Assessment and Plan in Terms of Control and Status | 10 minutes | Each participant had 10 minutes maximum to review the patient case and write an Assessment and Plan. The primary outcome evaluated participants' recommendations for treatment of patient conditions. Participants reviewed a total of 10 patient cases and received a score between 0 and 3 points for each issue within each patient case. The final score for each participant was a proportion between 0 and 1. The proportion represented the sum of all points assigned to the participant, divided by the total number of points possible. Higher values on the scale represent greater accuracy of the written assessment and plan. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Identification of Planned Monitoring and Follow up Encounters in Assessment and Plan | 10 minutes | Each participant had 10 minutes maximum to review the patient case and write an Assessment and Plan. . The secondary outcome evaluated participants' recommendation about future monitoring of patient conditions. Participants reviewed a total of 10 patient cases and received a score of 0 or 1 point for each issue within each case. The final score for each participant was a proportion between 0 and 1. The proportion represented the sum of all points assigned to the participant, divided by the total number of points possible. Higher values on the scale represent a greater proportion of appropriate monitoring recommendations made. |
Countries
United States
Participant flow
Recruitment details
Recruitment 12/2010 to 3/2011 at the Salt Lake City VA and University of Utah health care systems. Simulations took place at either of these locations.
Participants by arm
| Arm | Count |
|---|---|
| Integration Medication Manager Experienced providers that participated in the EHR simulations. Half of the providers were assigned to use the new Integrated Medication Manager (intervention) during the simulation. The other half were assigned the VA's CPRS to use (standard EHR). Providers were randomly assigned which system to use. | 30 |
| Standard EHR Experienced providers that participated in the EHR simulations. Half of the providers were assigned to use the new Integrated Medication Manager (intervention) during the simulation. The other half were assigned the VA's CPRS to use (standard EHR). Providers were randomly assigned which system to use. | 28 |
| Total | 58 |
Baseline characteristics
| Characteristic | Integration Medication Manager | Standard EHR | Total |
|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical Between 18 and 65 years | 30 Participants | 28 Participants | 58 Participants |
| Region of Enrollment United States | 30 participants | 28 participants | 58 participants |
| Sex: Female, Male Female | 18 Participants | 15 Participants | 33 Participants |
| Sex: Female, Male Male | 12 Participants | 13 Participants | 25 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 30 | 0 / 28 |
| serious Total, serious adverse events | 0 / 30 | 0 / 28 |
Outcome results
Primary
Accuracy of Written Assessment and Plan in Terms of Control and Status
Each participant had 10 minutes maximum to review the patient case and write an Assessment and Plan. The primary outcome evaluated participants' recommendations for treatment of patient conditions. Participants reviewed a total of 10 patient cases and received a score between 0 and 3 points for each issue within each patient case. The final score for each participant was a proportion between 0 and 1. The proportion represented the sum of all points assigned to the participant, divided by the total number of points possible. Higher values on the scale represent greater accuracy of the written assessment and plan.
Time frame: 10 minutes
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Integrated Medication Manager | Accuracy of Written Assessment and Plan in Terms of Control and Status | 0.609 units on a scale |
| Standard EHR | Accuracy of Written Assessment and Plan in Terms of Control and Status | 0.569 units on a scale |
Comparison: Null hypothesis: Participants will receive the same scores for assessments and plans completed using either IMM or CPRS. Power calculation: With 2 replications per subject, and assuming an ICC of 0.15, a two-sided alpha 0.05 comparison adjusted for 5 multiple comparisons (adjusted alpha = 0.01), and power of 80%, an N of 32 clinicians was required for each group (32 using IMM, and 32 using CPRS).p-value: 0.1595% CI: [-0.01, 0.09]Mixed-effects linear model
Primary
Amount of Time to Complete Assessment and Plan
Each participant had 10 minutes maximum to review the patient case and write an Assessment and Plan.
Time frame: 10 minutes
Population: 58 providers were enrolled
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Integrated Medication Manager | Amount of Time to Complete Assessment and Plan | 8.5 minutes | Standard Deviation 1.9 |
| Standard EHR | Amount of Time to Complete Assessment and Plan | 8.7 minutes | Standard Deviation 1.6 |
Comparison: Null hypothesis: Participants will require the same amount of time to complete assessments and plans using either IMM or CPRS. Power calculation: With 2 replications per subject, and assuming an ICC of 0.15, a two-sided alpha 0.05 comparison adjusted for 5 multiple comparisons (adjusted alpha = 0.01), and power of 80%, an N of 32 clinicians was required for each group (32 using IMM, and 32 using CPRS).p-value: 0.04795% CI: [-35.24, -0.23]Mixed-effects linear model
Secondary
Identification of Planned Monitoring and Follow up Encounters in Assessment and Plan
Each participant had 10 minutes maximum to review the patient case and write an Assessment and Plan. . The secondary outcome evaluated participants' recommendation about future monitoring of patient conditions. Participants reviewed a total of 10 patient cases and received a score of 0 or 1 point for each issue within each case. The final score for each participant was a proportion between 0 and 1. The proportion represented the sum of all points assigned to the participant, divided by the total number of points possible. Higher values on the scale represent a greater proportion of appropriate monitoring recommendations made.
Time frame: 10 minutes
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Integrated Medication Manager | Identification of Planned Monitoring and Follow up Encounters in Assessment and Plan | 0.505 proportion |
| Standard EHR | Identification of Planned Monitoring and Follow up Encounters in Assessment and Plan | 0.477 proportion |
Comparison: Null hypothesis: Participants will receive the same proportion of acceptable scores for assessments and plans completed using either IMM or CPRS. Power calculation: With 2 replications per subject, and assuming an ICC of 0.15, a two-sided alpha 0.05 comparison adjusted for 5 multiple comparisons (adjusted alpha = 0.01), and power of 80%, an N of 32 clinicians was required for each group (32 using IMM, and 32 using CPRS).p-value: 0.00595% CI: [1.22, 2.98]Regression, Logistic