Supporting Statement B for Request
The NHANES Longitudinal Study – Feasibility Component
OMB No. 0920-New
Contact Information:
Chia-Yih Wang
Team Lead, Planning Branch
Division of Health and Nutrition Examination Surveys
National Center for Health Statistics, CDC
3311 Toledo Road
Hyattsville, MD 20782
Email: [email protected]
Phone: 301 458-4697
FAX: 301-458- 4813
February 10, 2017
The NHANES Longitudinal Study – Feasibility Component
SUPPORTING STATEMENT B
TABLE OF CONTENTS
B. Collections of Information Employing Statistical Methods 3
1. Respondent Universe and Sampling Methods 3
2. Procedures for the Collection of Information 5
3. Methods to Maximize Response Rates and Deal with Nonresponse 13
4. Tests of Procedures or Methods to Be Undertaken 13
5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data 13
List of Attachments 15
Attachment 1 - Applicable Laws or Regulations (Excerpts)
Attachment 2a - Federal Register Notice
Attachment 2b - Federal Register Notice Comments and Response
Attachment 3a - Registration Form – Contact Confirmation and Scheduling Preference
Attachment 3b - Data Collection Forms for Presumed Living Participants – Home Visit
Attachment 3c - Data Collection Forms for Presumed Living Participants – Home Urine Collection
Attachment 3d - Data Collection Forms for Decedent Proxies
Attachment 4a - Advance Materials for Contacting Presumed Living Participants
Attachment 4b - Advance Materials for Contacting the Proxy of Deceased Participant
Attachment 5 - Consent Materials
Attachment 6 - Research Ethics Review Board Approval
Attachment 7 - Report of Test Results from Home Examination
Attachment 8a - Participant Information Sheet: “ABOUT THE REQUEST FOR HOSPITAL RECORDS”
Attachment 8b - Materials for Requesting Hospitalization Records
Attachment 9 - Targeted Methodological Studies
Supporting Statement B
The NHANES Longitudinal Study – Feasibility Component
This is a new request to conduct the NHANES Longitudinal Study – Feasibility Component. A two year clearance is requested.
B. Collections of Information Employing Statistical Methods
1. Respondent Universe and Sampling Methods
The sample design of NHANES (OMB No. 0920-0950, Exp. Date 12/31/19) is based on a continuous on-going annual survey of the non-institutionalized, civilian population of the U.S. Each single year and any combination of consecutive years comprise a nationally representative sample of the U.S. population. Annually, approximately 5,000 participants are examined in the NHANES; of these, about half are aged 20 or over. This design allows national estimates from NHANES for various subdomains every two years, including those defined by age, sex, and race and Hispanic origin.
The targeted sample size for each element in the NHANES Longitudinal Study – Feasibility Component is described in the table below:
Element |
Sample Size |
Field Feasibility Test |
|
Total |
800 |
Living Participant |
712 |
Decedent Proxy |
88 |
Hospitalization Record Retrieval |
|
30% of the 800 field feasibility test participants |
240 |
Targeted Methodological Studies |
|
Volunteers/Non- field feasibility test participants |
750 |
Sample Selection
Field feasibility test
The field feasibility test portion of the NHANES Longitudinal Study – Feasibility Component will include 800 adults, aged 20 or older, selected from the NHANES 2007-2014 survey cycle.
NHANES 2007-2014 included, 22,673 examined sample participants ages 20 years and older at baseline from 120 locations. To reduce the cost of the feasibility test, the 800 participants will be sampled within eight geographic clusters. These eight clusters were formed based on the geographic proximity of the original NHANES primary sampling units (PSUs), the number of PSUs in the cluster, the race and Hispanic origin characteristics of the original participants, years elapsed since baseline, and urbanization of the cluster; 7,118 of the 22,673 NHANES 2007-2014 adult participants are in these clusters. As noted above, this approach for the field feasibility test was selected as an optimal mechanism to obtain 800 participants meeting the desired parameters at a reasonable cost. The future longitudinal study of the entire 2007-2014 cohort will not include clustering, rather all eligible adults examined during these baseline years (possibly including the 800 persons sampled for the feasibility test) will be asked to participate.
From the eight clusters, the 800 field feasibility test participants will be selected via a probability sample using nine sampling domains defined by race and Hispanic origin (30% Hispanic, 30% non-Hispanic black, and 40% non-Hispanic other) and age at baseline (30% for ages 20-39, 30% for 40-59, and 40% for 60 years and over). Due to the small sample size, age distribution targets will not be required by race/Hispanic origin. Another evaluation factor for the field feasibility test is to determine the success of locating, interviewing, examining, and verifying participants’ vital status relative to the number of years since baseline. While geographic cluster and time since interview are not sampling domains in order to meet targets by race, Hispanic origin and age, it is expected that the sample will be relatively balanced by these factors due to criteria (listed above) used to form clusters. This expectation is supported by preliminary assessment.
Based on preliminary analyses of the NHANES linked mortality files, with 3-10 years between baseline and follow-up, we anticipate that in 2017, about 11% of the sample will be deceased. This should lead to about 88 potential decedent proxy respondents for assessing data collection procedures and obtaining an indication about response propensity for this subset, though detailed information about proxy respondents will not be feasible.
Hospitalization record retrieval
Permission of obtaining hospitalization records will be sought from all field feasibility test participants. Among those who provided permission, we will select a 30% convenience sample, which is approximately 240 participants, and request their hospitalization records. An average of 3 hospital stays per person is anticipated among this cohort, therefore, an estimated 720 requests (240 persons * 3 stays) will be made. Location of hospitalization and reported reason for hospitalization will be considered in selecting which participants will have hospital records requested to ensure a variety of situations will be encountered to inform decisions prior to incorporating hospital record retrieval into the full longitudinal study.
Additional targeted methodological tests
As part of the preparation efforts for a longitudinal study of all examined adults from NHANES 2007-2014, up to 750 participants may be asked to participate in targeted tests of proposed methods and procedures such as bio-specimen collections, cognitive testing for questions, or protocol tests for additional exam components (Attachment 9). These targeted tests will only occur if resources permit and if tracing and participation in the field feasibility test is successful. These targeted methodological studies will be conducted with non-NHANES volunteers or past NHANES participants who are not part of the potential NHANES Longitudinal Study sample (for example, past NHANES participants from the 1999-2006 cycle). Change Request will be submitted for clearance for each targeted methodological test with specific sample and protocol described.
2. Procedures for the Collection of Information
Data Collection Procedures
Prior to data collection, a review of the NHANES linked mortality files will be conducted to assist in determining the vital status of sampled participants. A contact profile will be prepared based on presumed vital status, baseline contact information, and change of address information.
For presumed living participants (no known vital records indicating death)
Data collection for the field feasibility test will proceed in two phases. In Phase 1, data collection will be carried out at one of the eight sample clusters to test all the training protocols, the data collection procedures, and the case management process (such as mailing schedule, case load distribution, and appointment scheduling). This cluster will be the one closest to NCHS allowing the maximum observation of field activities. Phase 2 of the field feasibility test involves the data collection for the reminding 7 clusters, including follow up on participants who moved out of the locations they were originally sampled from. There will be an interval of approximately one month between the end of the Phase 1 and the start of the Phase 2 to provide an opportunity to fine-tune all procedures and materials for training, and modify contact procedures and case management process as needed. Changes to the questionnaires, exam components, and consent documents will not be made. All planned aspects of the field feasibility test will take place during the Phase 1 and the results from the first cluster will be included in the final response rates.
A “Welcome back to the NHANES” letter (Attachment 4a) will be sent to the participant’s last known address to make the initial contact. An infographic with race and Hispanic origin specific data on diabetes will be embedded in the letter to help emphasize the longitudinal study’s chronic diseases focus, and to help connect to participants with different race and Hispanic origin backgrounds. This letter will provide instructions for the participant to confirm contact information (phone number and address) and to communicate their scheduling preferences for the in-home visit (see Attachment 3a for the registration module). Participants will be given the options to provide their contact confirmation and appointment preferences on-line or via telephone. In addition, the mailing will also include an NHANES magnet (the same magnet participants received at the baseline) to jog the participant’s memory of previous participation, and $2 as a token of appreciation to encourage continued participation.
The welcome letter will be sent with the “Address Service Requested” option. If the welcome letter is not successfully delivered, further tracking and tracing activities will occur. Tracking and tracing activities may include contacting the two individuals named by the NHANES respondent at baseline as future contacts, and/or utilizing other linkage resources such as State Departments of Motor Vehicles (DMV), Centers for Medicare & Medicaid Services (CMS), and LexisNexis databases. These tracking and tracing activities will be conducted by contract staff who completed the required NCHS confidentiality training and signed the Designated Agent Agreement in compliance with the NHANES Security Plan.
If the welcome-back letter is delivered successfully, but the participant does not provide a response to the contact request, another letter will be sent 14 days later to encourage the participant to respond. A third mailing in the form of a reminder card will be sent to the respondent 7 days later, if the participant still has not made contact. Please refer to Attachment 4a for all the advance materials for presumed living participants, and Attachment 3a for the registration module that the participants can use to confirm their contact information and scheduling preferences.
Once the participant provides his/her contact confirmation, a Health Representative will call and schedule a home visit. Participants who cannot be contacted by phone or who have not responded to the advance mailings will have a non-scheduled home visit by a Health Representative. If no one is at home or no one comes to the door after two attempts, the Health Representative will try to contact neighbors to get information such as whether the house is occupied and when someone is likely to be home. The two attempts at the dwelling unit have to occur on two different dates, and at a different time of the day (morning, afternoon, or evening) or different type of day (weekday vs weekend). Health Representatives will identify themselves as working on a health survey for the National Center for Health Statistics, and will not reveal the name of the survey or the name of the sampled person to the neighbor, or leave any material that may identify the sampled person as a survey participant. A neighbor refers not only to the person who lives next door but also to anyone in the immediate vicinity (e.g., a building manager, a mailman passing by, someone in the yard across the street from the assigned address) who may be able to provide the information.
During the home visit, a set of questions will be asked first to verify the identity of the participant. Please see Attachment 3b for the suggested introductory script and the verification questionnaire. Once identified, the Health Representative will present the informational consent brochure (Attachment 5) to the participant and have the participant read and sign the study consent form (Attachment 5) prior to the interview and the home examination. The consent form will be read to the participant if he/she has difficulty reading. If the respondent is unable to sign the form, a witness signature is obtained to indicate that informed consent was obtained from the participant verbally. Permission to audiotape the interview will also be obtained verbally and digitally recorded at the start of the interview.
Proxy interviews will be conducted for mentally or cognitively impaired participants, following the same protocol used in the current NHANES. Consent will be sought from guardians or family member caregivers. Mentally or cognitively impaired individuals capable of understanding the basics of the research are given the opportunity to consent to participation. Assessments of this capability are made by Health Representatives and/or supervisors in collaboration with the guardian or caregivers. Interviews or examinations will not be conducted for those who are capable of giving consent but decline to participate.
The core module currently planned for the NHANES Longitudinal Study, including an in-person interview and a health examination will be administered following the consent. The majority of the questions included in the interview are the same as the ones used in the baseline survey. Some minor modifications such as probing for a different time frame were made with input from NCHS’ Collaborating Center for Questionnaire Design and Evaluation Research (CCQDER) to adjust for the follow-up study setting. Additional questions were developed or adapted from the current NHANES (OMB No. 0920-0950, Exp. Date 12/31/19), the National Health Interview Survey (NHIS; OMB No. 0920-0214, Exp. Date 12/31/19), or other well-established surveys such as NIH’s REasons for Geographic and Racial Differences in Stroke (REGARDS) project (under a cooperative agreement that does not require OMB approval), Atherosclerosis Risk in Communities Study (ARIC; OMB No. 0925-0281, Exp. Date 5/31/17), or Michigan Neuropathy Screening Instrument (MNSI) to accommodate for specific topics. Please refer to Attachment 3b for the proposed questionnaire and more detailed information on the source of individual questionnaire sections. Data collected during the interview includes the following 4 subject domains:
Sociodemographic Information
Sociodemographic information is collected to characterize the survey respondent, including the sampled participant, and the proxy and/or interpreter, if used. Data collected may be used as covariates in health related analyses or for future contact purpose. Specific components included in this section are:
Respondent information (RIQ)
Demographic information of the sampled participant (DMQ)
Participant’s contact information (MAQ)
Tracking and tracing information (TTQ)
Health Status and Medical Conditions
Health status at follow-up and the development of chronic medical conditions are the main outcomes of interest for the NHANES longitudinal study. The majority of the data collected in this section are planned to be used to estimate the incidence of specific conditions or combination of conditions since baseline. In addition, several condition specific data items (such as currently taking prescription medication for hypertension or diabetes) are included to provide covariates for condition related analyses. Specific components included in this section are:
Medical conditions (MCQ)
Disability (DLQ)
Diabetes (DIQ)
Vision (VIQ)
Neuropathy (PNQ)
Blood pressure and cholesterol (BPQ)
Cardiovascular disease (CDQ)
Kidney conditions (KIQ)
Health Care Services
Access to health care and health insurance have been considered important factors that relate to the prevention and control of chronic medical conditions. Data collected in this section will enable related analysis in this topic and provide covariates for condition related analyses. In addition, we will collect information that can be used to facilitate linkage with Medicare and Medicaid data, and to obtain hospitalization records. Components included in this section are:
Hospital utilization and access to care (HUQ)
Health insurance (HIQ)
Hospitalizations (HVQ)
Health Behaviors
These behavior factors both at baseline and follow-up are important covariates in evaluating the health outcomes of interest in the NHANES Longitudinal Study. Components included in this section are:
Physical activity (PAQ)
Sleep disorders (SLQ)
Weight history (WHQ)
Cigarette smoking (SMQ)
Alcohol use (ALQ)
Aspirin and prescription medication use (RXQ)
Following the interview, a health examination will be conducted as part of the same home visit. The participant’s weight, waist circumference, and sitting blood pressure will be measured, and a peripheral neuropathy assessment will be conducted using a monofilament. A capillary blood sample will be collected from the participant’s finger to test for hemoglobin A1C. In addition, instructions and a urine collection kit will be given to the participant to collect and return a urine sample for urine albumin and creatinine testing. This proposed project will assess the feasibility of conducting these tests and procedures in the home examination setting. Please refer to Attachment 3b and Attachment 3c for more details of each individual examination component.
Upon completion of each task (i.e., registration, interview and health examination, home urine collection), each participant will receive incentives as detailed in Section A.9. Test results from the weight, waist circumference, and blood pressure measurements will be given to the participant at the end of the home examination. The full set of test results from the examination will be mailed to the participant several weeks after the home visit. A summary of the examination test results given to each participant is included in Attachment 7.
For participants deceased prior to the re-contact
A review of the NHANES linked mortality files will be conducted prior to data collection to assist in determining the vital status of sampled participants. A death certificate will be requested from individual states for those identified as deceased. A proxy will be identified and contacted for an interview via telephone. Procedures used to identify the proxy for a deceased participant have been reviewed by the Office of the General Counsel at CDC for confidentiality protection and CIPSEA compliance and are described in Attachment 4b. All sampled participants not identified as deceased will be assumed to be alive.
Once identified, information collected at baseline and the participant’s death certificate will be used in conjunction with the National Change of Address database to compile the potential proxy’s latest contact address and phone number. A letter (Attachment 4b) will be sent to the identified proxy’s last known address to make the initial contact. The letter will provide instructions for the proxy to confirm contact information and to communicate his/her scheduling preferences for a telephone interview. The respondents will be given the options to provide their contact confirmation and appointment preferences on-line or via telephone. A NCHS Confidentiality Brochure will be included in the mailing to inform the potential proxy how his/her privacy will be protected (Attachment 4b).
Similar to the protocol used for contacting presumed living participants, the letter will be sent with the “Address Service Requested” option. If the letter is not successfully delivered, further tracking and tracing activities including utilizing other linkage resources such as State Departments of Motor Vehicles (DMV), Centers for Medicare & Medicaid Services (CMS), and LexisNexis databases will take place. If the letter was delivered successfully, but the proxy did not provide a response to the contact request, a second letter will be sent 14 days later to encourage the person to respond. Please refer to Attachment 4b for all the advance materials planned for contacting the proxy of deceased participants, and Attachment 3a for the registration module that the proxies can use to confirm their contact information and scheduling preferences.
Once the identified proxy provides contact confirmation and preferred appointment time, an interviewer will call and conduct a telephone interview. Proxies who have not been successfully contacted in advance will receive a non-scheduled phone call to request an interview. For those who do not have access to a phone, a home visit will be conducted to collect the data. Please refer to Attachment 3d for suggested introductory telephone script for contacting decedent proxies, and the verification procedure that will be used to confirm the identity of the decedent. Designated telephone interviewers administering the decedent proxy interview will receive additional training on working with topics of a sensitive nature to help them communicate with potentially grieving respondents.
After the completion of verification, the interviewer will invite the proxy to continue with a 20-minute interview on the decedent’s health history and medical conditions prior to the death. Permission to record the interview will be obtained verbally and digitally recorded at the start of the interview. If the respondent declines the request to have the interview recorded, no recording will be done. A waiver of signed consent from decedent proxies has been requested from the NCHS Research Ethics Review Board (ERB) because it’s not feasible for the telephone interview. The only information we will collect on the proxy is his/her name, contact information, and relationship to the decedent. Otherwise, the subject for the interview will be the decedent. Even though the proxy may not be considered as a human subject in the proposed study under title 45 CFR part 46, informed consent will be obtained verbally. At the beginning of the interview, the proxy will be told (Attachment 3d, question RIQ.245, page 20):
“The information collected in this study is protected by the Public Health Service Act and the Confidential Information Protection and Statistical Efficiency Act. We are required by these laws to use the information you provided for statistical research only and to keep it confidential. In addition, the Federal Cybersecurity Enhancement Act of 2015 requires the federal government to protect federal computer networks by using computer security programs to identify cybersecurity risks like hacking, internet attacks, and other security weaknesses. The Act allows software programs to scan information that is sent, stored on, or processed through government networks in order to protect the networks. If any cybersecurity risk is detected, the information system may be reviewed for specific threats by computer network experts working for the government (or contractors or agents who have governmental authority to do so). The brochure we sent to you with the initial letter of invitation for the study has more details on how your information will be kept confidential. I’m happy to send you another copy if you would like one.
Your taking part in this study is voluntary. If you choose to take part, you don’t have to answer every question and you can stop the interview at any time.”
The verbal permission for the interview will be digitally recorded.
Data collected during the proxy interview include the following topics:
Sociodemographic Information
Sociodemographic information is collected to characterize the survey respondent, including the sampled participant and the proxy, and interpreter (if used). Specific components included in this section are:
Respondent information (RIQ)
Demographic information of the sampled participant (DMQ)
Respondent’s contact information (MAQ)
Health and Medical Histories
Health status at follow-up and the development of chronic medical conditions are the main outcomes of interest for the NHANES longitudinal study. The majority of the data collected in this section are planned to be used to estimate the incidence of specific conditions or combination of conditions since baseline. Several data items (such as taking prescription medication for hypertension or diabetes prior to death, and cigarette use in the year before the death) are included to provide covariates for health outcome and medical condition related analyses. In addition, we will collect information that can be used to facilitate hospitalization record retrieval. Specific components included in this section are:
Medical conditions (MCQ)
Diabetes (DIQ)
Blood pressure and cholesterol (BPQ)
Kidney conditions (KIQ)
Hospitalizations (HVQ)
Cigarette smoking (SMQ)
Similar to the questionnaire designed for living participants, most questions included in the decedent proxy interview are the same as the ones used in the baseline survey. Some minor modifications are made to adjust for the follow-up study setting. Additional questions are developed or adapted from current NHANES or other well-established surveys to accommodate for specific topics. Please refer to Attachment 3d for the questionnaire proposed for the deceased participant proxy interview and details on the source of the questions.
If, during the re-contacting process, it is learned that a participant is recently deceased, the field staff will verify that the deceased was the baseline NHANES participant based on procedure outlined in Attachment 3d. Once verified, the home office will obtain the death certificate and identify a proper proxy as described in Attachment 4b. The proxy interviewer will follow the same protocol as for pre-identified decedents and conduct the interview via telephone.
Obtaining hospitalization records
Permission to collect hospital discharge data, including diagnoses at discharge and procedures performed during hospitalization will be obtained during the interview for both living (Attachment 3b, page 115) and deceased participants (Attachment 3d, page 51). Among those who provided permission, we will select 240 participants and have their hospitalization records requested to evaluate the record retrieval protocol for the study cohort among different medical facilities. Two different types of record requests will be made to help us assess the feasibility of the data collection procedures and the analytic usability of the content of requested hospitalization records. Specifically, for 160 of these 240 participants, we will contact the hospitals and ask them to send the hospital face sheet with diagnostic codes (ICD-9 and/or ICD-10 CM codes), and the discharge summary to us. If the discharge summary is not available, progress note from the last day of hospitalization will be requested. For the other 80 participants, full medical records will be requested for all their hospitalizations, which will include the following: hospital face sheet, discharge summaries, progress notes for the entire hospitalization, emergency room reports, admission notes, consultations, operative reports, pathology reports, radiology reports, EKGs, laboratory reports, and cardiac catheterization reports.
Living participants will be asked to sign HIPAA compliant authorizations (Attachment 8b) during the home visit to allow the identified hospitals to disclose the participant’s protected health information to NHANES. To fully inform participants of the nature of the hospitalization record request, an information sheet (Attachment 8a) will be reviewed prior to the request for HIPAA authorization. Please refer to Attachment 3b, page 123, questions HVQ.150 for detailed probes and talking points.
For participants who can only be reached by telephone, the interview portion of the home visit will be conducted through the phone. The permission to collect hospitalization data will be sought as part of the interview (Attachment 3b, page 115). However, rather than meeting HIPAA requirements by obtaining a written authorization over the phone, we have requested a HIPAA Waiver from the NCHS Research Ethics Review Board (ERB). This request is compliant with HIPPA privacy rules because the use and disclosure of protected health information involves no more than a minimal risk to the privacy of the participant, which is assured by adherence to the Section 308(d) of the Public Health Service Act (42 United States Code 242m) and the Confidential Information Protection and Statistical Efficiency Act (CIPSEA, Title 5 of Public Law 107-347). No identifiable information will be released without consent. In addition, obtaining a written Authorization is not feasible during a telephone interview thus the research cannot practicably be conducted without the requested waiver, and the research cannot practicably be conducted without the access to and use of this protected health information.
For participants identified as deceased, we have also requested a HIPAA Waiver of obtaining a written authorization from the proxy. This request is based on HIPPA privacy rules specified in Title 45, Section 164.512 for research on decedent's information, since the use or disclosure of the information being sought is solely for research on the protected health information of decedents, and the protected health information being sought is necessary for the research.
A package will be sent to hospitals to request the hospitalization records for the participant since baseline. Please refer to Attachment 8b for all the materials included in the package. Hospitals will be asked to return the records via FedEx with pre-printed return airbills (no option to FAX or email will be provided). Three weeks after mailing the initial request, hospitals that have not yet responded will be contacted via telephone. If requested, the original package will be mailed to the hospitals again and/or the request will be sent by FAX. Instances in which information returned from providers is unclear, incomplete, or contains inconsistencies may require further contact with the hospitals to retrieve correct information.
All proposed procedures and materials used for obtaining hospitalization records have been reviewed by the Office of the General Counsel at CDC.
A contractor is responsible for data collection procedures. The responsibilities of the contractor are to:
Translate all questionnaires as required
Hire home office and field staff
Create manuals and training programs for all field procedures (including training in NCHS confidentiality guidelines and regulations)
Conduct trainings for field staff members
Develop publicity/outreach methods and materials
Select the sample
Locate sampled participants and identify their vital status
Obtain death certificates for deceased participants
Contact and schedule respondents
Conduct the interviews and examinations
Obtain hospitalization records
Design and carry out quality control procedures
Transmit data to NCHS
Quality Control
The majority of the field feasibility test data will be collected from respondents electronically. Built-in quality control checks will be programmed into the automated data collection system to maximize the accuracy and consistency of the data.
All field staff will complete training requirements for safety, privacy and confidentiality. In addition, all staff will complete extensive component-specific training to learn the standardized protocol for each of their assigned tasks. Equipment calibration and on-site inspections will be performed and documented regularly.
Additional quality control monitoring methods will be implemented to ensure that high quality data were collected during the survey. All interviews will be audio recorded and the recorded interviews will be reviewed by NCHS staff and interviewer supervisors. The recordings will be kept for 3 years after the completion of field feasibility test data collection. Feedback from the review will be communicated to the interviewers and their supervisors on an on-going basis. Retraining will be carried out when necessary. Field observations will be done throughout the survey to monitor staff performance and case management.
Data collected will be electronically transmitted to NCHS. Extensive review will be conducted regularly to verify the completeness, consistency, and analytic usefulness of the data. Extreme values will be cross-checked with other available data.
Estimation Procedures
Similar assessments on response rate will be conducted for deceased participants as appropriate. Estimates include the percentage of participants with death confirmed by a death certificate or a proxy interview, and the completion rates of proxy interview will be examined overall and by demographic subgroups, the number of years since baseline, and selected health conditions.
Non-response bias assessments will be conducted using baseline NHANES health and demographic data with model based approaches. While detailed assessments within specific gender, age, race and Hispanic origin, and survey year cells will not be possible in the field feasibility test, conditional evaluations should provide provisional information about the differences between respondents and non-respondents.
3. Methods to Maximize Response Rates and Deal with Nonresponse
Interviewers have access to a variety of materials to assist them in nonresponse conversion. Other methods to maximize response include:
Personalized examples of NHANES data embedded in the advance letter to remind previous participation and to relate the respondent to the survey
Providing test results from the examination
Monetary incentives (see details in Section A.9)
Bilingual staff (English and Spanish)
Interpreters for languages other than Spanish
Flexible interview and home examination schedules including evenings and weekends
Telephone reminders before scheduled appointments
Intensive follow-up efforts including proactively calling participants to reschedule broken appointments
Endorsement letters from local and national organizations
4. Tests of Procedures or Methods to Be Undertaken
Almost all of the proposed content for the field feasibility test were part of previous NHANES such as sections of the questionnaires and components of the examination. The study procedures of the NHANES I Epidemiologic Follow-up Study which included four rounds of re-contact from 1982-1992 of the 1971-1975 NHANES cohort were highly successful. These procedures will be adapted to current technology and confidentiality standards for the proposed data collection.
5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data
a. Individuals Consulted on Statistical Aspects
Jennifer D. Parker, Ph.D.
Mathematical Statistician
Division of Health and Nutrition Examination Surveys
National Center for Health Statistics
Centers for Disease Control and Prevention
Phone: 301-458-4419
b. Individual Responsible for Data Collection Activities
Kathryn S. Porter, MD, MS, FACPM
Captain, U.S. Public Health Service
Director, Division of Health and Nutrition Examination Surveys
National Center for Health Statistics, Centers for Disease Control and Prevention
Phone: 301-458-4441
c. Individual Responsible for Data Analysis
Kathryn S. Porter, MD, MS, FACPM
Captain, U.S. Public Health Service
Director, Division of Health and Nutrition Examination Surveys
National Center for Health Statistics, Centers for Disease Control and Prevention
Phone: 301-458-4441
List of Attachments
Attachment 1 - Applicable Laws or Regulations (Excerpts)
Attachment 2a - Federal Register Notice
Attachment 2b - Federal Register Notice Comments and Response
Attachment 3a - Registration Form – Contact Confirmation and Scheduling Preference
Attachment 3b - Data Collection Forms for Presumed Living Participants – Home Visit
Attachment 3c - Data Collection Forms for Presumed Living Participants – Home Urine Collection
Attachment 3d - Data Collection Forms for Decedent Proxies
Attachment 4a - Advance Materials for Contacting Presumed Living Participants
Attachment 4b - Advance Materials for Contacting the Proxy of Deceased Participant
Attachment 5 - Consent Materials
Attachment 6 - Research Ethics Review Board Approval
Attachment 7 - Report of Test Results from Home Examination
Attachment 8a - Participant Information Sheet: “ABOUT THE REQUEST FOR HOSPITAL RECORDS”
Attachment 8b - Materials for Requesting Hospitalization Records
Attachment 9 - Targeted Methodological Studies
Page
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Vicki Burt |
| File Modified | 0000-00-00 |
| File Created | 2021-01-22 |