Agency Information Collection Activities: Proposed Collection; Comment Request

SUPPLEMENTARY INFORMATION:

Title of Information Collection: Evaluating the Implementation of PCOR to Increase Referral, Enrollment, and Retention through Automatic Referral to Cardiac Rehabilitation (CR) with Care Coordinator.

OMB No.: 0935-0252.

Type of Request: Reinstatement without change to discontinue the collection.

The aim of this project, known as TAKEheart, was to (a) raise awareness about the benefits of cardiac rehabilitation (CR) after myocardial infarction or coronary revascularization, then to (b) spread knowledge about the best practices to increase referrals to CR, and, finally, (c) to increase CR uptake.

AHRQ evaluated TAKEheart to assess:

• the extent and effectiveness of the dissemination and implementation efforts
• the uptake and usage of Automatic Referral with Care Coordination and
• levels of referral to CR at the end of the intervention.

Evaluation results were used to improve the intervention and to provide guidance for future AHRQ dissemination and implementation projects. Two cohorts of “Partner Hospitals,” up to 125 hospitals in total, engaged in efforts to implement Automatic Referral with Care Coordination over twelve-month periods. The evaluation ascertained the diversity of hospitals engaged in the activities that contributed to (or hindered) their efforts, and the types of support which they reported having been most (and least) useful. This information was used to improve recruitment, technical assistance, and tools for the second cohort.

In addition, hospitals—including those involved in the implementation—were invited to attend Affinity Group virtual meetings organized around specific topics of interest which are not intrinsic to Automatic Referral with Care Coordination. Hospital staff engaged in Affinity Groups created a vibrant Learning Community. The evaluation determined which Affinity Groups engaged the most participants of the Learning Community, and which resources participants determined the most useful. This information was used to develop resources which were available on a new, permanent website dedicated to improving CR.

This study was conducted by AHRQ through its contractor, Abt Associates Inc., pursuant to AHRQ's statutory authority to disseminate government-funded research relevant to comparative clinical effectiveness research. 42 U.S.C. 299b-37(a).

Method of Collection

To collect data on the many facets of the intervention, the collection implemented multiple data collection tools, each of which had a specific purpose and set of respondents.

1. Partner Hospital Champion Survey. Each Partner Hospital designated a “Champion” who coordinated activities associated with implementing Automatic Referral with Care Coordination at the hospital and provide the Champion's name and email address. Champions could have had any role in the hospital, although they were expected to be in relevant positions, such as cardiologists or quality improvement managers. We conducted online surveys of 125 Champions (one Champion per hospital). We used the email addresses to send the Champion a survey at two points: seven months after the start of implementation and at the end of the 12-month implementation period. The first survey focused on four constructs. First, it captured data about the hospital context, such as whether it had prior experience customizing an EMR or is a safety net hospital. Second, it addressed the hospital's decision to participate in TAKEheart. Third, it captured data on the CR programs the hospital refers to, whether the number or type has changed, and why. Fourth, it collected feedback on the training and technical assistance received. The second survey focused on three constructs. The first construct collected feedback on the TAKEheart components, including training, technical assistance, and use of the website. The second construct asked about the hospitals' response to participating in TAKEheart, such as changes to referral workflow or CR programs. The third construct asked those Partner Hospitals that had not completed the process of implementing Automatic Referral with Care Coordination whether they anticipated continuing to work towards that goal and their confidence in succeeding.

2. Partner Hospital Interviews.

a. Interviews with Partner Hospital Champions. We selected, from each cohort, eight Partner Hospitals which demonstrated a strong interest in addressing underserved populations or reducing disparities in participation in cardiac rehabilitation. We conducted a key informant interview with the Champion of each selected Partner Hospital to delve into how they were addressing the needs of underserved populations by implementing Automatic Referral with Care Coordination.

b. Interviews with Partner Hospital cardiologists. We selected, from each cohort, eight hospitals based on criteria selected in conversation with AHRQ, such as hospitals which serve specific populations, or have the same EMRs, which informed their experience customizing the EMR. We conducted semi-structured interviews with one cardiologist at each of the selected hospitals twice. In the second month of the cohort implementation, we asked about their needs, concerns, and expectations of the program. In the 11th month of the cohort implementation, we determined whether their concerns were addressed appropriately and adequately.

c. Interviews with Partner Hospitals that withdraw. We expected that a small number of Partner Hospitals would withdraw from the cohort. We identified these hospitals by their lack of participation in training and technical assistance events; Technical Assistance (TA) Providers confirmed their withdrawal. We interviewed up to nine withdrawing hospitals to better understand the reason for withdrawal ( e.g., a merger resulted in a loss of support for the intervention, Champion left), as well as facilitators and barriers of each hospitals' approach to implementing Automatic Referral with Care Coordination. If more than nine hospitals withdrew, we ceased interviewing.

3. Learning Community Participant Survey. We conducted online surveys of 250 currently active Learning Community participants at two points in time, in months 18 and 31 of the project. We administered the survey by sending a link to an online survey to email addresses entered by virtual meeting participants during registration. The email described the purpose of the survey.

4. Learning Community Follow-up Survey. We conducted a brief online survey with up to 15 Learning Community participants following the final virtual meeting for each of 10 Affinity Group, to ascertain whether the hospitals were able to act on what they learned during the session. The total sample was 150 Learning Community participants.

Estimated Annual Respondent Burden

Exhibit 1 presents estimates of the reporting burden hours for the data collection efforts. Time estimates were based on prior experiences and what could reasonably be requested of participating health care organizations. The number of respondents listed in column A, Exhibit 1 reflects a projected 90% response rate for data collection effort 1, and an 80% response rate for efforts 3 and 4 below.

1. Partner Hospital Champion Survey. We assumed 113 hospital champions would complete the survey based on a 90% response rate. It was expected to take up to 45 minutes to complete for a total of 169.5 hours to complete.

2. Partner Hospital Interviews. In-depth interviews occured with select Partner Hospital staff.

a. Interviews with Partner Hospital Champions. We had a single, 90 minute interview with eight Partner Hospital Champions, in each cohort, from Partner Hospital which have a common characteristic of particular interest, for a total of 24 hours.

b. Interviews with Partner Hospital cardiologists. We held individual, up-to-30 minute interviews with eight cardiologists, twice in each cohort, for a total of 16 hours.

c. Interviews with Partner Hospitals that withdraw. We interviewed up to nine withdrawing hospitals for no more than 20 minutes to better understand the reason for withdrawal as well as facilitators and barriers, for a total of 2.7 hours.

3. Learning Community Participant Survey. We assumed 200 Learning Community participants would complete the survey based on an 80% response rate. It was expected to take up to 15 minutes to complete each survey for a total of 100 hours.

Learning Community Follow-up Survey. We conducted a brief, up to 10 minute, online survey of participants of each of just ten selected Affinity Groups at two months after the virtual meeting. We assumed 120 Learning Community participants would complete the survey based on an 80% response rate. It was expected to take up to 15 minutes to complete each survey for a total of 20.4 hours.