0379 - HCPS Supporting Statement B 06.21.19

Supporting Statement B

Health Center Patient Survey: Cognitive Pre-Testing to Support the Development of Survey Questions

OMB Control No. 0915-0379

2. Collection of Information Employing Statistical Methods

If statistical methods will not be used to select respondents and the corresponding item is checked “No”, this section does not have to be completed.

For the pretest (cognitive interviews), the Health Resources and Services Administration (HRSA) will not use statistical methods to select respondents. Instead, HRSA will utilize the convenient sampling approach. HRSA contracted with RTI International to conduct the pretest for the 2019 Health Center Patient Survey. Under convenient sampling, HRSA will select participants based on their convenient accessibility and proximity to RTI offices, which will ensure the appropriate distribution of participants in terms of age and gender to test the survey in each of the five languages (English, Spanish, Tagalog, Chinese, and Vietnamese). HRSA will recruit English- and Spanish-speaking participants at urban and rural health centers, which will allow for pretesting of on-site recruitment procedures and pretesting the questionnaire. Tagalog-, Chinese-, and Vietnamese-speaking participants will be recruited through traditional ads, including posting flyers and online advertising.

1. Respondent Universe and Sampling Methods

The respondent universe for the pretest includes parents of children 12 and younger, adolescents 13-17, and adults 18 and over that have received services at a health center in the past 12 months. HRSA will not use statistical methods to select patients during the pretest.

The pretest (cognitive interviews) will involve two rounds of testing. In round 1 (Table 1), 34 interviews will be conducted using computer-assisted personal interviewing (CAPI). HRSA will recruit one rural and one urban health center within 150 miles of Durham, NC, to conduct on-site interviews. HRSA may recruit a health center within 100 miles of the Chicago RTI office because RTI has staff geographically positioned to conduct these interviews. HRSA will recruit and interview 16 participants (eight English and eight Spanish) at participating health centers and interview an additional 18 participants (six Tagalog, six Vietnamese, and six Chinese [Mandarin and Cantonese]) who recently visited a health center for a health-related service. To effectively recruit 18 respondents speaking languages other than English and Spanish, RTI will post advertisements in community forums and post flyers at local health centers, community centers, and libraries in Raleigh-Durham, NC, and Chicago, IL. Other potential locations include Washington, D.C. and San Francisco, CA. HRSA chose these locations due to a higher concentration of populations speaking Chinese, Tagalog, or Vietnamese and close proximity to RTI offices and staff.

Each sampled respondent will receive approximately one-half of the questionnaire to ensure that the respondent is not overly burdened. The whole questionnaire, along with the cognitive probes, would take about two hours to complete. As a result, the interviewer will administer a sample of modules to each respondent. HRSA pre-selected modules for respondents to receive based on age and gender to ensure that, collectively, there are enough data from each module to accurately assess the questionnaire. Each respondent will receive all the questions in the selected module. Cognitive testing will use scripted and spontaneous probing techniques, along with “think-aloud” methods to identify potential problems during administration.

Table 1: Round 1 of Cognitive Interviews

Cohort	Test	Recruited and Conducted at Clinic	Recruited through Local Ads and Conducted at RTI / Other Location	Total
English	1/2 of the questionnaire per participant	8	0	8
Spanish	NA	8	0	8
Tagalog	NA	0	6	6
Chinese	NA	0	6	6
Vietnamese	NA	0	6	6
TOTAL	NA	16	18	34

Following the first round of testing, HRSA will make edits to the questionnaire in each of the languages and administer the revised version in the second round of cognitive testing. HRSA will conduct a total of six English and six Spanish interviews on site at the health center (Table 2). In addition to the 12 on-site interviews, HRSA will recruit and conduct four interviews in Tagalog, four in Chinese (Mandarin and Cantonese), and four in Vietnamese through advertisements posted in community forums at Raleigh-Durham, NC, Chicago, IL, and potentially at Washington, D.C. and San Francisco, CA. Round 2 of the pretest will include a total of 24 cognitive interviews. All efforts will be made to recruit a diverse sample in terms of respondent age, gender, and race/ethnicity. HRSA will continue to recruit until the pretest achieves the targeted number of participants in each demographic category in each location via advertisements and flyers. Experience from the 2014 Health Center Patient Survey suggests that HRSA will receive a sufficient number of inquiries to participate in the interview to ensure a diverse selection.

Each participant in round 2 will be administered the entire questionnaire utilizing skip logic so that participants will only receive questions relevant to them. Cognitive testing in round 2 will use debriefing probes administered after the survey is completed. These probes will allow the interviewer to assess the flow of the questions from domain to domain and will serve as dry runs to evaluate questionnaire timing. HRSA will also test the skip logic while assessing comprehension issues.

Table 2: Round 2 of Cognitive Interviews

Cohort	Test	Recruited and Conducted at Clinic	Recruited through Local Ads and Conducted at RTI/ Other Location	Total
English	All	6	0	6
Spanish	All	6	0	6
Tagalog	All	0	4	4
Chinese (Mandarin and Cantonese)	All	0	4	4
Vietnamese	All	0	4	4
TOTAL	All	12	12	24

2. Procedures for the Collection of Information

Interviews recruited on-site at a local clinic: In round 1, HRSA will recruit patients from one urban and one rural clinic. This will allow HRSA to test the questionnaire, on-site recruitment procedures, and assess burden on clinic staff. Interviewers will receive training on the procedures and will train the clinic staff once the facility has agreed to participate in the pretest.

To test recruitment procedures, HRSA will train on-site clinic staff on patient tracking and recruitment procedures. During the time when an interviewer is at the site, the receptionist will be asked to screen the patient’s eligibility when a patient enters a site and registers for service. The receptionist will ask the patient’s age, race/ethnicity, and veteran status to determine eligibility. The receptionist will keep track of the number of patients who enter the site, the number of patients who are eligible, and number of patients referred while the interviewer is at the site to conduct cognitive interviews.

Interviews recruited through traditional advertising: In both rounds, HRSA will recruit participants using traditional advertising, including posted flyers and advertisements posted on free websites and on-line forums. HRSA will screen each interested participant to determine his or her eligibility and will then draw a sample, ensuring the diversity of participants in terms of age, gender, and race/ethnicity. Recruitment will continue until a diverse sample, based on selected demographics, is achieved.

Once patients agree to participate, interviewers will administer the pretest questionnaire. Interviewers will not contact the participants following the cognitive interview.

3. Methods to Maximize Response Rates and Deal with Nonresponse

HRSA will provide eligible participants a $25 incentive for completion of the interview. Response rates for the pretest will be a function of success in two basic activities: (1) identifying eligible patients and (2) obtaining those patients’ cooperation and time to complete the interview. During the consent process, patients are informed that their participation is voluntary. Choosing not to participate will not affect any services they receive at the health center or from any other programs. They can opt to not answer some of the questions, not finish the interview, or take breaks at any time during the interview. If interviewers cannot complete the interview at the scheduled time, they will reschedule interviews as necessary, make multiple attempts to reach the participant, and reschedule participants who miss appointments. HRSA does not anticipate nonresponse to be a concern during the cognitive pretest.

4. Tests of Procedures or Methods to be Undertaken

Efforts will be used to improve information technology to reduce burden. HRSA is targeting 58 completed interviews over two rounds of testing using CAPI on a laptop computer. The CAPI approach offers several advantages that keep respondent burden at a minimum while ensuring high-quality data collection. First, the questionnaire is complex and involves numerous skip patterns and screening questions. These are easily and quickly performed by the computer upon completion of CAPI programming. In addition, the CAPI instrument detects erroneous and inconsistent responses, increasing data accuracy and validity. The use of CAPI will enable the interviews to be completed in less time, with reduced burden, and with more accuracy than a paper and pencil interview. HRSA will then test cognitive understanding of the questions along with the usability of the computerized instrument.

HRSA has determined that a total of 58 interviews are necessary to adequately test all new and revised questions, ensuring the congruency of questions and changes the skip patterns of questions or routing, in each of the five languages (English, Spanish, Vietnamese, Tagalog, and Chinese – Mandarin / Chinese – Cantonese). The questionnaire varies by respondent type (Child-proxy, Adolescent, Adult). Having a larger pretest sample will ensure proper testing of the questionnaire among these various sub-groups.

A majority of the questions in the 2019 questionnaire (about 80%) come from the 2014 Health Center Patient Survey. HRSA is adding new topic areas and adapting questions from existing surveys to allow for comparability between questionnaires. The topics and the study source of the questions added in 2019 are listed in Table 3.

Table 3. New Topical Areas and Source of Questions in the 2019 HCPS

Topic	Source
Activities of daily living	2000 Longitudinal Study of Aging (OMB No. 0920-0411); 2018 BRFSS (OMB No, 0920-1061)
Falls	2018 BRFSS (OMB No, 0920-1061)
Undiagnosed diabetes	2017 NHIS (OMB No, 0920-0124)
Exercise	2018 NHIS (OMB No, 0920-0124)
Food security	2012 USDA Six-Item Module
E-cigarette use	2017 BRFSS (OMB No, 0920-1061)
Medications	2018 Home Health Care CAHPS Survey (OMB No. 0938-1066)
Medication management	Krousel-Wood M, Islam T, Webber LS, Re R, Morisky DE, Muntner P. New medication adherence scale versus pharmacy fill rates in hypertensive seniors. The American journal of managed care. 2009;15(1):59-66
Mental health (suicide)	2018 NSDUH (OMB No, 0930-0110)
Mental health (autism/ADHD)	2018 NHIS (OMB No, 0920-0124)
Prenatal care and breastfeeding	2016 PRAMS Phase 8 Core Questionnaire
Barriers to obtaining medication	2018 NHIS (OMB No, 0920-0124)
Public housing	2018 NHIS (OMB No, 0920-0124)
Telehealth	Created by RTI methodologist
Neighborhood characteristics	2012 Community Advantage Panel Study

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

Although statistical analysis is not anticipated for the pretest, the following individuals will be consulted on data collection and qualitative analysis for the pretest.

Alissa Harvey

Contracting Officer’s Representative

Health Resources and Services Administration

301-443-5979

[email protected]

Josh Bolton

Statistician

Health Resources and Services Administration

301-443-0547

[email protected]

Kathleen Considine

Project Director

RTI International

(919) 541-6612

[email protected]

Azucena D. Hammill

Assistant Project Director/Data Collection Task Leader

RTI International

(919) 541-7231

[email protected]

Tim Flanigan

Instrumentation Task Leader

RTI International

(919) 485-7743

[email protected]

Patrick Chen

Sampling Task Leader

RTI International

(919) 541-6309

[email protected]

Cynthia Bland

Analysis Task Leader

RTI International

(919) 541-6154

[email protected]

6