SUPPORTING STATEMENT B
Assessing the Infrastructure for Public STD prevention Services
(OMB No. 0920-XXXX)
PART B: SPECIFIC INSTRUCTIONS – COLLECTIONS EMPLOYING
STATISTICAL METHODS
NEW SUBMISSION
April 26, 2017
Submitted by:
Division of Sexually Transmitted Disease Prevention
Centers for Disease Control and Prevention
Department of Health and Human Services
Refer questions to:
Jami Leichliter, PhD
Policy Science, Team Lead
Division of STD Prevention
Centers for Disease Control and Prevention
1600 Clifton Rd NE, MS E-02
Atlanta, Georgia 30329
(404) 639-1821
FAX (404) 639-8622
Email: [email protected]
B.1 Respondent Universe and Sampling Methods 3
B.2 Procedures for the Collection of Information 4
B.2.A. Sample Sizes and Expected Precision 4
B.2.D Estimation and Weighting 4
B.2.F Data Quality Control Procedures 4
B.3 Methods to Maximize Response Rates and Deal with Nonresponse 5
B.3.A Expected Response Rate 5
B.3.B Assessing Nonresponse and Analysis of Non-Response Bias 5
B.4 Test of Procedures or Methods to be Undertaken 5
B.5 Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data 6
2a. Public Comments
LHD Survey – Word Version
SHD Survey – Word Version
LHD Survey - Web Screen Shots
SHD Survey – Web Screen Shots
Definitions for Survey Terms
NCHHSTP Determination of Non-Research
Web Survey Invite – LHD and SHD
Web Survey Reminders – LHD and SHD
Follow-up phone call script
Extension email – LHD and SHD
This information collection will include two surveys: one for local health departments (LHDs, Attachment 3) and one for state health departments (SHDs, Attachment 4). The National Association of County and City Health Officials (NACCHO) will be responsible for collecting data for both surveys. The LHD survey will be distributed to 668 respondents and the SHD survey will be distributed to 50 respondents. Both surveys will occur twice in the three year period.
LHD Survey: For each data collection, the respondent universe for the LHD survey is 668. The size of the sample is based on the following factors: a desired confidence interval of 95% for estimates; a total population size for LHDs of 2,533, and an anticipated 50% response rate for the data collection (based on the response rate for the baseline data collection conducted in 2013). Given these factors, 334 completed surveys are needed. The information collection will be distributed to either the local health official or the STD program manager (if this information is known for the respondent) at each LHD in the respondent universe.
A two-step sampling method will be used to draw the sample of 668 LHDs. First, all LHDs that are in the highest 50 counties for primary and secondary syphilis, chlamydia, and gonorrhea by rate and number of cases will be included. For each STD (syphilis, chlamydia, and gonorrhea), counties will be ranked by cases and rates. Ranks will be summed across the STDs and the counties with the 50 lowest scores will be included in the sample. The data source to identify these 50 counties will be CDC’s 2015 STD surveillance data. Second, the remainder of the sample will be randomly selected from the rest of the LHD population that responded to the most recent NACCHO’s Forces of Change Survey. The total LHD population is 2,533. Thus, once the highest 50 for STDs are selected, 2,483 LHDs will remain, from which 618 LHDs will be selected (for a total of 668 LHDs in the sample). The sample will be stratified by jurisdiction population size (small = <50,000; medium = 50,000-499,999; and large = ≥500,000) and Census region (Northeast, Midwest, West, South). Since LHDs with a large population size represent a relatively small proportion of all LHDs, those LHDs will be over sampled to ensure a sufficient number of responses from large LHDs for the analysis.
SHD Survey: For each data collection, the survey will be distributed to STD program managers from all 50 state health departments. Centers for Disease Control and Prevention’s (CDC) Division of STD Prevention (DSTDP) maintains a list of program managers which will be used to gather contact information.
The supplemental table provides the number of respondents by information collection and annualized.
Type of Respondent |
Form name |
Attachment No. |
No. Respondents in IC #1 |
No. Respondents in IC #2 |
Annualized No. of Respondents |
STD program director, LHDs |
LHD survey |
3 |
334 |
334 |
223 |
STD program director, SHDs |
SHD survey |
3 |
44 |
44 |
29 |
For the LHD survey at each information collection, the target sample size of 668 LHDs and the sample allocation to strata were developed to ensure the required precision levels for key subgroup estimates (i.e. jurisdiction size and Census region). The expected sample size of at least 334 completed LHD surveys will obtain a 95% confidence level with ±5% margin of error for the total population of 2,533 LHDs.
For the SHD survey at each information collection, the sample size is 50 SHDs.
For each information collection, the LHD survey sample will be selected with stratified random sampling from NACCHO’s LHD database. The sample sizes will be allocated to strata with the approaches discussed in Section B.1.
For each information collection, the SHD survey sample includes the state health department in each of the 50 U.S. states.
For the LHD survey, respondent contact information (name, LHD name, state, email address, and phone number) will be obtained from NACCHO’s LHD member database.
For the SHD survey, respondent contact information (name, SHD name, state, email address, and phone number) will be obtained from DSTDP’s list of STD program managers. DSTDP maintains this list as it provides funding to STD programs in all 50 states.
For LHD data subsequent to each information collection, weights will be developed based on U.S. Census region, jurisdiction population size, and non-response to represent the total population of 2,533 LHDs.
Clearance is being sought for a cross-sectional survey that will be administered over a 2-month period. This survey will be repeated every 3 years to evaluate changes in the infrastructure for STD services over time.
To ensure the integrity of the web-based survey, both the LHD and SHD survey instruments will undergo a quality control process during beta testing, which will be conducted by NACCHO. Before the survey is programmed into Qualtrics, the online data collection platform used by NACCHO, it was written and formatted in Microsoft Word. The Word document contains all survey content, programming logic, and programming notes. From the Word document, the surveys were programmed into Qualtrics by a member of NACCHO’s Research and Evaluation team (web-versions of the surveys are in Attachments 5 and 6). After programming, another NACCHO staff member went through the surveys to ensure that all survey content from the Word documents is included and that the programming logic and notes are consistent with the functioning in Qualtrics. Test links were shared with key CDC personnel for the purpose of beta testing. After beta testing, NACCHO pilot tested the surveys. See Section B.4.
Unique data collection links will be used for respondents to help ensure that there will be no respondents outside the intended universe. Unique links will also enable NACCHO to add jurisdiction-level demographic information maintained in NACCHO LHD database to respondents’ data after they complete their survey, so that respondents are not asked to provide such information (e.g., jurisdiction population size and method of governance).
Data from the web-based survey instrument will be downloaded and cleaned by NACCHO. All open-ended and “other/specify” fields will be reviewed and back-coded, if applicable.
For each information collection, the expected response rate for the LHD survey is 50%. This expectation is based on the response rate achieved during the 2013 baseline data collection. Additionally, this rate is similar to what has been achieved via similar NACCHO data collection efforts with LHDs.
For each information collection, the expected response rate for the SHD survey is 80%. Although the SHD response rate for the 2013 assessment was 61%, we have added telephone follow-up in an attempt to increase this response rate. We are hopeful that we will obtain this higher response rate for SHDs given that DSTDP collaborates with all 50 states for STD prevention activities.
To ensure sufficient time to complete the information collection, the deadline for survey completion will be four weeks from the date of notification giving respondents 20 full business days to complete the survey. If necessary, the deadline will be extended by two weeks.
Although participation in the assessment is voluntary, NACCHO will make every effort to maximize the rate of response. Efforts include sending up to two reminder emails to complete the survey. Nonresponses will be constantly monitored and assessed during the data collection process by examining a list of non-respondents periodically and checking the response rate of participants in total and by jurisdiction size and census region. If the target response rate has not been reached after the two e-mail reminders, NACCHO will attempt to increase the response rate by calling non-respondents, and, if necessary, extending the period of time for health departments to respond to the survey. Reminder emails will be sent at the beginning of the third and fourth weeks to non-respondents to urge them to complete the assessment (see Attachment 10). Calls will be made to non-respondents beginning at the end of the third week and extending into the fourth week (see Attachment 11). If necessary, based on the response rate achieved to-date and the feedback received during follow-up calls, the survey deadline will be extended by one week (see Attachment 12).
The information collection is 100% electronic, and is designed to facilitate respondent survey completion with minimal burden. Features for doing so include clearly displaying survey instructions and questions, and including additional instruction when needed, such as at the beginning of survey sections; making it easy to navigate from one webpage to the next, as well as the ability to go back and save and continue at a later time; and adding user assistance tools, such as hover text to provide definitions for key terms within survey questions (Attachment 7). Non-response bias will be assessed by comparing the main characteristics (e.g., jurisdiction size, region) of respondents and non-respondents. Bivariate analysis will be used to check the proportion of 1) LHDs in each category of main characteristics and 2) SHDs by U.S. Census region among respondents and non-respondents. The results of significance tests will provide evidence on the extent of non-response bias.
The information collection instruments were pilot tested by 8 health departments – 5 LHDs and 3 SHDs – in March 2017. The feedback provided from the pilot test was used to refine questions, ensure accurate programming and skip patterns, and establish the estimated time required to complete the information collection instrument. Slight modifications were made to a few survey items to clarify the timeframe covered by these items. Additionally, after reviewing responses, we found that SHDs were trying to respond for their own activities as well as those of the LHDs within their jurisdiction despite instructions stating to only respond for SHD activities. We have revised the survey to emphasize these instructions throughout which should significantly decrease the amount of time SHDs spend completing the survey and will improve the quality of the data that we receive. Finally, the pilot also led us to delete a new response option that had been added but was too vague to be of use.
The following individuals have reviewed technical and statistical aspects of procedures that will be used to pilot and implement the Survey of STD Provider Policies and Practices:
NACCHO will be responsible for collecting and cleaning the data, provided review and consultation for the development of the survey instruments, and will assist in data analysis. NACCHO staff include:
Gretchen Weiss, MPH
Director, HIV, STI, and Viral Hepatitis
NACCHO
1100 17th St. NW, 7th Floor
Washington, DC 20036
(202) 783-5550
Samantha Ritter, MPH
Senior Program Analyst, Adolescent Sexual Health
NACCHO
1100 17th St. NW, 7th Floor
Washington, DC 20036
(202) 783-5550
Kat Kelley, BS
Program Analyst, HIV, STI & Viral Hepatitis
NACCHO
1100 17th St. NW, 7th Floor
Washington, DC 20036
(202) 783-5550
Scientists from CDC/DSTDP will lead the analysis of data from the survey. The lead scientists from CDC are:
Jami Leichliter, PhD
Team Lead (Lead Health Scientist)
CDC
1600 Clifton Rd NE, MS E-02
Atlanta, GA 30329
(404) 639-1821
Kendra Cuffe, MPH
Health Scientist
CDC
1600 Clifton Rd NE, MS E-02
Atlanta, GA 30329
(404) 639-1847
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Thomas, Matthew |
| File Modified | 0000-00-00 |
| File Created | 2021-01-22 |