Supporting Statement B 02Jun2020

Validated Interview and Survey of Outpatient Clinicians on Antibiotic Stewardship Interventions

Request for OMB approval of a Revision Information Collection

June 2, 2020

Supporting Statement B

Contact:

Amy McMillen

Centers for Disease Control and Prevention

1600 Clifton Road NE

Atlanta, Georgia 30329-4027

Phone: (404) 639-1045

Email: [email protected]

Table of Contents

1. Respondent Universe and Sampling Methods 2

2. Procedures for the Collection of Information 2

3. Methods to maximize Response Rates and Deal with No Response 2

4. Tests of Procedures or Methods to be Undertaken 2

5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data 2

Both qualitative and quantitative data will be combined to assess knowledge, attitudes, and practices related to antibiotic prescribing among clinicians after implementation of a year-long Urgent Care stewardship initiative. We will assess if the interventions within the stewardship initiative were acceptable, appropriate, and feasible and will continue to benefit patients beyond the implementation period. We will use summary statistics (i.e., rates, averages, or proportions) and inferential statistical methods to analyze the data and make statistical inferences about the results that could be applied to other populations or settings.

1. Respondent Universe and Sampling Methods

Our source population will include over 400 urgent care clinicians whose contact information will be provided by the medical director of the urgent care service line.

First we will conduct semi-structured interviews with a sample of 40 clinicians. This sample will include 9 clinicians from our original qualitative study to determine changes in perceptions over time. In addition, we are proposing to sample 21 additional UC clinicians who participated in the year-long stewardship initiative and 10 control clinicians from the University of Utah. The reason we are proposing to sample at least 40 individuals from our urgent care service line is to ensure that we do not introduce unnecessary bias from a smaller sample size which would limit generalizability of the deep contextual information and would put our results at risk. Our sample will be selected to ensure diverse representation of geography of practice (rural vs urban), gender, clinical specialty, and current antibiotic prescribing practices to enhance the representativeness of practice and for comparative purposes. Using the contact information provided by the medical director of the urgent care service line, we will invite a select sample of clinicians to participate in a semi-structured interview using a script developed for e-mail use (Attachment 5). We will perform one follow-up communication with non-responders to maximize response from this selected sample group.

Second, we will disseminate a validated survey to all UC clinicians (400+ clinicians) employed by Intermountain. We expect roughly 250 clinicians to respond (50% response rate) to our survey. Using the contact information provided by the medical director of the urgent care service line, we will invite all of the service line clinicians to participate in a survey (Attachment 3) using the language included in our templates developed for e-mail use (Attachment 5).

2. Procedures for the Collection of Information

Researchers from the Healthcare Delivery Institute will conduct the proposed research activities. For semi-structured interviews, we will either travel by appointment to outpatient facilities or allow interviews to be conducted via virtual meetings or phone to make completion less burdensome for clinicians. Collection will occur using audio recording and personal field notes. Audio-recordings will be transcribed. Survey instruments will be disseminated electronically to all UC clinicians by the Healthcare Delivery Institute.

3. Methods to maximize Response Rates and Deal with No Response

We may have challenges recruiting busy healthcare workers to participate in an interview and/or to take a survey. To mitigate this risk, we will consider using (a) using short, personal email communications, (b) highlighting how the study results will be shared, (c) sending reminders through respected clinical leaders, and (d) giving deadlines to ensure diverse representation. We have provided potential clinicians with multi-modalities (ex. in-person interviews at their clinic, virtual meetings, or phone interviews) to participate in the semi-structured interviews. Follow-up invitations to complete the electronic survey will be sent out via other existing communication channels (regional meetings, newsletters) if there is a low initial response rate to the first invitation. We will use the follow-up template with language developed for e-mail use (Attachment 5). Intermountain Urgent Care leaders will also consider allow us to disseminate the survey and complete their responses during existing regional or clinic meetings to engage potential clinicians in the electronic survey to assess implementation.

4. Tests of Procedures or Methods to be undertaken

In attachment 3, we have included our semi-structured interview guide and the validated survey instrument that we will use to accomplish our purpose.

5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data

Intermountain Healthcare, Healthcare Delivery Institute, Collection and Analysis of Data:

Kim Brunisholz, 801-442-3912, [email protected]

Raj Srivastava, 801-507-8715, [email protected]

Nick Baker, 801-507-8712, [email protected]

Ishita Singh, 801-507-8711, [email protected]

Tom Belnap, 801-507-8717, [email protected]