Summary of Motivation for Changes Made to The MEPS HC Instrument
In conjunction with re-engineering the MEPS HC CAPI instrument in Blaise COTS, AHRQ has updated the content of the instrument to 1) use features of Blaise, particularly grid navigation 2) reduce redundancy and reduce the burden of data collection for respondents by both simplifying tasks and eliminating the collection of some variables which were rarely or never used.
From a data quality perspective, this is particularly important for MEPS. We have noted that respondents are likely to underreport medical events in a variety of circumstances. Relieving the burden of reporting is therefore a critical step to improving the quality of MEPS use and expenditure instruments. The new design relieves the reporting burden on both the interviewer and the respondent.
While we have provided the full specifications for the instrument in other attachments, the purpose of this document is to provide a broad view of the changes to the instrument and to note any changes to variables that would be on the public use files, or other critical internal files.
While our overall goal in designing and developing the instrument was to improve data quality, we organized the updates around several different principles. Each principal supports improved data quality in general. In the event enumeration and detailed use and expenditure sections, we applied the following three principles with the goal of getting fuller health care use reporting:
Reduce the cognitive burden of responding, simplifying the response task so respondents report more fully;
Use methods that can reduce the errors that result in diminished reporting; and
Simplify interview administration tasks.
Throughout the instrument, we applied two additional principles in the design of the MEPS instrument that motivated the changes to the flow and structure of the instrument:
Enumerate first, then collect details; and
Reduce burden by focusing on items that support key policy and research needs and eliminate those that do not.
Comparison of Sections from 2017 Instrument to Current Instrument, by section
Reenumeration
Start/Restart (ST) section is new. It is administered at the beginning of each interview and when interviewing resumes following a breakoff. This section verifies the interviewer has selected the correct case, identifies the person to serve as primary respondent, and obtains consent for (audio) recording of the interview for quality control purposes. The questions in the ST section were previously included in RE-A.
The Reenumeration (RE-A and RE-B) section collects eligibility and demographic information on each person associated with the household participating in MEPS. Changes to the question wording in RE-A subsection more accurately identifies the keyness and eligibility status of household members and facilitates interviewer and respondent cognition, reducing the need to supplement respondent information and responses with detailed comments entered by the interviewer to clarify specific situations. While the mode of collecting the information has changed considerably, the use of grid-based administration allows the interviewer flexibility to navigate the instrument and collect information for multiple family members with increased efficiency, the content of this section has not changed.
Health Status
CAPI will only present the full Priority Condition Enumeration (PE) section in Rounds 1 and 3. At the time of the last OMB submission, CAPI asked the full PE section in Rounds 1, 3 and 5. Round 5 interviews are outside the reference period of the household so this information was of little utility.
Similarly, CAPI will no longer ask the Health Status (HE) section in all rounds of a panel. CAPI will omit the HE section in the fifth round of data collection for a panel. Again, Round 5 interviews are outside the reference period for the household so this information was of little utility.
Additionally, the HE section will only include the Department of Health and Human Services (DHHS) approved standard questions to capture limitations in daily living in Rounds 2 and 4, omitting the additional historical MEPS items that overlapped in content.
Use and Expenditures
Because of our ongoing concern regarding burden and accuracy of reporting, this section had significant changes, mostly to the method of reporting rather than the actual information being collected. Collecting use and expenditure information in MEPS consists of several tasks. These tasks include
Using a set of “provider probes” to elicit whether anyone has a specific type of event
Looking up address information for each named provider to facilitate the Medical Provider Component
Enumerating the dates of all of the visits/medical events
Collecting details about each visit/medical event including expenditure information
Changes to the use and expenditure section include the following:
MEPS will no longer include the possibility of asking the set of Provider Probes (PP) items for each person in the household individually to facilitate recall and reporting of health care utilization in the absence of records. Instead, the new CAPI instrument will only cycle through the provider probes set of questions a maximum of two cycles depending on the household composition. Only households with extended family members to the reference person will be asked the provider probes a second time, with the first cycle focusing only on health care received by the respondent and his/her spouse or partner and their dependent children, and the second cycle focusing on all other extended family members.
Date picker – This CAPI feature uses a calendar-based screen format to enhance respondent recall of health care event dates and make the date entry process more efficient. New recurrence options enable quick identification of health care event dates for care that repeats on one or more days per week, weekly, or monthly. The recurrence options enable much quicker entry of frequent health care visits, such as dialysis, which previously required separate entry for each visit.
Provider Search tool – The new provider lookup tool has a user interface similar to an Internet Search Engine. The database is customized to only include the providers with National Provider Identifier records (NPI) within 100 miles of each household’s zip code. This approach allows for faster identification of providers.
“Switch” feature – This feature allows interviewer to switch from person to person, as directed by the respondent, in any person-level sections. Choosing which household member to ask questions about next facilitates increased self-reporting by household members when they are available to assist the primary respondent. This respondent-centered feature also customizes the flow of the interview when asking about persons in a particular order reflects the respondent’s preference or the organization of the household’s records.
“Off-path” feature – This is another respondent-centered tool that allows interviewers to accept responses provided after the relevant section, accounting for variation in the ability to recall and report all requested information at the exact point defined by CAPI flow. Adding data such as missed health care events or household members using the off-path feature ensures that the interview administers all of the relevant CAPI questions for the newly reported information. It also reduces the added data processing costs necessary when data is added in a comment.
The Event Driver (ED) section uses a new grid format to structure the verification and modification of reported events. Event detail is collected for only the first event in a repeat visit group. Also, by reducing the number of linked events from three to two as a requirement to create a repeat visit group, the burden for households with high rates of health care utilization is greatly reduced.
Calendar section (CA) – This redesigned section eliminates different paths through the sections of the questionnaire that collect information on medical events based on the interviewer’s assessment of the completeness of the household’s record. The CA section links with the Event Roster (EV), Provider Roster (PV), and Event Driver (ED) sections of the CAPI interview.
Reduce the amount of “event detail”
Hospital Stays (HS) and Emergency Room Visits (ER) – no changes in event detail
Outpatient Department Visits (OP), Medical Provider Visits (MV) –If there are multiple visits to the same provider, CAPI now collects certain information about the provider only once.
Dental Visits (DN) - Changes to this section simplify the collection of dental services by reorganizing seven dental services groupings into five major groupings and combining two periodontic categories. The reduced set of categories may make it easier for respondents to identify the specific service that applies to them. Additionally, the section added a response that makes it easier for respondents to identify a flat fee situation for orthodontia.
Home Health care (HH) – As in the past, this section collects information on the kind of provider administering care in the home, the reason for the care and the frequency of visits. The revisions include simplifying the response categories for provider type and encouraging the respondent to group visits for reporting.
The new Institutional Care (IC) section captures condition data and provider type (e.g. inpatient rehabilitation facility, residential addiction treatment center) for institutional care stays in an institution providing 24-hour skilled nursing care.
Prescribed medications (PM) – As in the past, this section collects detail about medications reported as part of other medical events as well as those reported outside of a particular event. A new lookup function reduces the need for manual pharmacy address input.
Other medical expenses (OME) – Expenses are asked each interview round to reduce the recall period to the same reference period as the rest of the interview. Previously, expenses were asked in Rounds 3 and 5 with respondents asked to report on all applicable expenses for the entire, prior calendar year.
Charge Payment (CP) - As in the past, this section tracks total charges and sources of payment for medical events reported in earlier sections. In addition to repeat visit groups, events for a person can be linked together as a flat fee (FF) for those types of medical payment arrangements that charge a grouped amount for multiple visits are services. The revisions to this section include the deletion of questions about reimbursements, or the amount of additional expected payments, updated question wording to improve respondent understanding of concepts, and simplifying the cognitive task when verifying information about repeat co-pays in Rounds 2-5.
Event follow up (EF)– This new section reduces underreporting errors by following up on certain events such as a hospital stay for a women who has given birth to a baby to ensure hospital stay events were collected for both mother and child. This section is also administered if a household member reports having received lab tests, to determine whether the tests were performed at the medical provider’s office or during a separate health care event at a lab.
A new section, Event Enumeration (EE), uses a navigation tool to prompt the respondent for any additional health care in specific event categories including emergency room visits that lead to a hospital stay or visits to any type of residential rehabilitation facility or treatment facility. The goal of this section is to simplify the recall task for respondents while enhancing respondent reporting particularly in specific event categories identified as being erroneously omitted in historical CAPI.
The Additional Healthcare Question (AH) section includes a subset of questions previously included in the eliminated sections of Disability Days (DD), Accidents/Injuries and Conditions (CN). It assesses the impact of any physical illness, injury, or mental/emotional on household members’ attendance at work. Two new items, administered in Rounds 3 and 5, ask about smoking and engaging in vigorous exercise.
Employment and Health Insurance
Employment – Changes to this section (EM-A) and subsections RJ (Review of Employment) and EW (Employment Wage) include omitting questions with a very small applicable universe and that had very little variation in response. Changing the question order and reducing the detail collected for last job outside the reference period and any retirement jobs, are designed to improve interview flow. A list of the omitted questions could be provided on request.
Health Insurance – Changes include a new set of items (verification series), that is administered if at least one current household member is without any source of comprehensive public or private health insurance during the entire reference period. The question flow for these items is modeled after similar questions in the Current Population Survey (CPS)*. Other structural changes include more grids and person-level administration.
Health Insurance Detail (HP) – Minor changes streamlined this section such as automatically skipping a question about whether any dependents living in the RU are covered by this insurance plan if the respondent lives alone.
Time Period Covered Detail (HQ) – A single-screen grid format was developed to help interviewers and respondents more easily identify similar coverage periods for different household members.
Managed Care (MC) – No substantive changes.
Old Employment and Private Related (OE) – The use of a single grid screen is designed to reduce cognitive burden and simplify reporting by requiring the respondent to answer “yes” and “no” for each person individually.
Public Insurance (PR) – Rewording of some questions to specifically mention RU members covered to eliminate confusion. As with the OE section, the use of a single grid screen forces a “yes” or “no” for each person individually when reviewing coverage from a previous round.
CAPI Supplements
Quality Supplement (QS) – The redesigned QS section (formerly PC) presents an overview of all hard copy supplements requested from any household member, distributes each supplement and collects an initial status for each item. Previously the introduction and distribution of hard copy supplements were part of several CAPI sections, for example the Diabetes Care Supplement was distributed in the former PC section while other self-administered questionnaires were introduced, distributed, and collected in the Closing section.
Preventive Care (AP) – This section is deleted. The same or similar information is collected in the hard copy Adult SAQ (SAQ), Your Health and Health Opinions.
Access to Care (AC) – The AC section is administered in Rounds 2 and 4. Changes to structure and question wording allow for better benchmarking to other federal surveys regarding the measurement of unmet/delayed health care needs by using items corresponding to the National Health Interview Survey (NHIS).
Income (IN) – This section is administered in Rounds 3 and 5. The re-designed section uses previously provided responses about marital status, employment status, and the ages of household members to skip questions that are not applicable about alimony, unemployment compensation, and child support.
Assets (AS) – The Assets section is asked only in Round 5. This section has been revised to reduce burden by significantly reducing the number of topics covered, now focusing only on the two main assets applicable to the national sample.
Child Health Supplement (CS) – This section, asked in Rounds two and four, includes four different scales or measures. CAPI will only ask two measures in a year of data collection, alternating across the two years of the life of a panel to collect all four measures. The four scales include:
Living with Illness Measures (LWIM) – Identifies chronic medical, emotional, and behavioral illnesses
Columbia Impairment Scale (CIS) – Rates degree of functioning in 13 behavioral areas
Consumer Assessment of Healthcare Providers and Systems (CAHPS) – Collects information on unmet needs and provider satisfaction
Prevention – Reduced number of preventive health questions from the historical CAPI instrument by removing:
if and when blood pressure has been measured
if and when a health care provider gave advise about regular dental checkups
collection of body weight detail (ounces) for children weighing less than 20 lbs
access to care, tests, or treatments believed necessary by a physician and/or parent/guardian
Self-Administered Questionnaires
Your Health and Health Opinions – Adult SAQ (SAQ) - A brief self-administered questionnaire (SAQ) collects self-reported (rather than through household proxy) information on health status, health opinions and satisfaction with health care for adults 18 and older. The satisfaction with health care items are a subset of items from the Consumer Assessment of Healthcare Providers and Systems (CAHPS). The health status items are from the Veterans Rand 12 item health survey (VR-12), a generic instrument developed with the support of the Department of Veterans Affairs and the Centers for Medicare and Medicaid Services. The SAQ is administered once a year, either in Round 2 or Round 4, with follow up collection of outstanding forms in Rounds 3 or 5. The SAQ will be administered in 2019 (for Panel 24 Round 2 and Panel 23 Round 4) and 2021 (for Panel 26 Round 2 and Panel 25 Round 4). Administration of the SAQ will rotate each year with administration of the Male/Female SAQ (PSAQ) (see next item below).
Your Health and Health Choices – Male/Female Adult SAQ (PSAQ) – A self-administered questionnaire administered once in a year in either Round 2 or Round 4, with follow up collection of outstanding forms in Rounds 3 and 5. The PSAQ will be administered in 2020 for Panel 25 Round 2 and Panel 24 Round 4. There are two versions, one for male and one for female respondents. The PSAQ collects information about preventive care. Many of the questions in this SAQ were previously part of the Preventive Care (AP) section of the CAPI interview that have been deleted. Administration of the PSAQ will rotate each year with the administration of the SAQ.
A Survey About Your Diabetes Care – This is a brief self-administered paper-and-pencil questionnaire (DCS) on the quality of diabetes care is administered once a year (during Round 3 and 5) to persons identified as having diabetes. Included are questions about the number of times the respondent reported having certain blood tests, foot and eye exams, and whether the diabetes has caused kidney or eye problems. Respondents are also asked if their diabetes is being treated with diet, oral medications, or insulin. This questionnaire is unchanged from the previous OMB submission.
Understanding Veterans’ Health Care Needs – A self-administered questionnaire (VSAQ) to collect health and health care related information for approximately 1,500 veterans selected from the in-scope sample based on being a veteran of the U.S. military identified in the MEPS Round 1 interview. This SAQ is new and will be administered only one time in Rounds 1, 3, and 5 in spring 2019.
Respondent Forms
This new section, part of the CAPI interview in all five rounds, is administered to all households that are asked to complete at least one hard copy supplement or to sign at least one authorization form (medical provider or pharmacy). Signed authorization forms and completed hard copy supplements (Self-Administered Questionnaires) are collected and an updated status is recorded for each item.
This section used the historical CAPI section Closing (CL) as a base for development. The changes to the new Respondent Forms section allows the interviewer to switch the order in which the instrument asks about each household member’s forms and supplements to be more responsive to the availability of each household member to ask questions and complete forms.
Closing
This section has been streamlined for the collection of information that facilitates contacting households in subsequent rounds. Additionally, CAPI can now collect up to 10 telephone numbers and questions are asked to obtain permission to send text messages to cell phones while identifying any rules or restrictions for using each phone number for contacting the household in subsequent rounds.
*Question text modeled after CPS:
HX210: {I have recorded that {you are/PERSON} is} covered by {Indian Health Service} {and} {health insurance that may not include hospital and physician benefits}. I have recorded that {you are/PERSON} is} not currently covered by any kind of health plan or health coverage.} {Were you/Was {PERSON} covered at any time {since {PERSON’S START DATE}/between {PERSON’S START DATE} and {PERSON’S END DATE}}, even if just for one day, by any {other} kind of health insurance plan or health coverage that included hospital and physician benefits?
HX215: For that coverage, {do you/does {PERSON}} get it through a job, the government or state, is it privately purchased, for example through an insurance company, HMO, or {do you/does {he/she}} get it some other way?
HX225: Is that plan related to military service in any way?
HX30: From which of the government or state sources on card HX-4 {were you/was {PERSON}} covered by?
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | DHHS |
| File Modified | 0000-00-00 |
| File Created | 2021-01-13 |