+1 vote
asked ago by (56.3k points)
From the AEA Committee on Economic Statistics:
The Agency for Healthcare Research and Quality (AHRQ), HHS, seeks OMB approval to renew the MEPS-IC and encourages comments to OMB regarding its request by December 2, 2019.  
Invitation to comment: https://www.federalregister.gov/documents/2019/11/01/2019-23872/agency-information-collection-activities-proposed-collection-comment-request    

MEPS-IC survey instruments and supporting docs on plans and methods https://reginfo.gov/public/do/PRAViewICR?ref_nbr=201910-0935-001
Employer-sponsored health insurance is the source of coverage for 84.4 million current and former workers, plus many of their family members, and is a cornerstone of the U.S. health care system. The Medical Expenditure Panel Survey—Insurance Component (MEPS-IC) measures the extent, cost, and coverage of employer-sponsored health insurance on an annual basis. These statistics are produced at the National, State, and sub-State (metropolitan area) level for private industry. Statistics are also produced for State and Local governments.

This research has the following goals:

(1) To provide data for Federal policymakers evaluating the effects of National and State health care reforms.

(2) to provide descriptive data on the current employer-sponsored health insurance system and data for modeling the differential impacts of proposed health policy initiatives.

(3) to supply critical State and National estimates of health insurance spending for the National Health Accounts and Gross Domestic Product.
The total budgeted sample size for the regular, annual collection is approximately 46,000 sample units before non-response and out-of-scope units.  The sample frame covers almost 100 percent of all organizations with at least one employee in the economy, excluding the Federal government. In order to meet the goals of the survey, the sample is developed in several steps.  These are:

•    the sample is allocated to the state and local government and private sectors
•    a small sample is set aside for certainty units in both the private sector and state and local governments
•    the remaining sample within the private sector is allocated by state and the government sample is allocated by Census Division
•    the sample is allocated to strata within each sector.


The private sector and state and local government samples are weighted separately.  Beginning with the inverse probability of selection as a base weight, the data are adjusted for non-response.    Among other characteristics, the number of plans, size of establishment, size of firm or government unit, industry, and whether the respondent received a mail survey must be considered.  Adjustments are made so that the sum of respondents in a cell equals the beginning weighted total for the cells.  It would be desirable to adjust for non-response in each state, but cell size may preclude that adjustment in all but the largest states.

Once adjustments for non-response are made, data are post stratified to a set of new control totals.  For the private sector, this would be new frame counts from the most recent BR for cells determined by state, industry, size of firm and size of establishment.  For governments, post stratification is done to counts provided by the Census of Governments Division.  Cells are determined by size of government and state.  

Estimation and Accuracy

Estimation is done using sampling weights and variances are calculated using standardized software, such as  SAS, using the Taylor series methodology to  account for the specialized sampling methods used for the survey.

Certain key variables, such as premiums, contributions and enrollments are imputed when item non-response occurs and values are missing.  Selection of donors is accomplished using a nearest-neighbor hot-deck process which chooses the best donor given a set of matching variables and their order of importance.  Variables used to match establishments to determine donors are chosen for their correlation with the variable to be imputed, with special care taken to select variables which are also correlated with non-response.   
Information collection instruments:

(1) Prescreener Questionnaire—The purpose of the Prescreener Questionnaire, which is collected via telephone, varies depending on the insurance status of the establishment contacted (establishment is defined as a single, physical location in the private sector and a governmental unit in state and local governments). For establishments that do not offer health insurance to their employees, the prescreener is used to collect basic information such as number of employees. Collection is completed for these establishments through this telephone call. For establishments that do offer health insurance, contact name and address information is collected that is used for the mailout of the establishment and plan questionnaires. Obtaining this contact information helps ensure that the questionnaires are directed to the person in the establishment best equipped to complete them.

(2) Establishment Questionnaire—The purpose of the mailed Establishment Questionnaire is to obtain general information from employers that provide health insurance to their employees. Information such as total active enrollment in health insurance, other employee benefits, demographic characteristics of employees, and retiree health insurance is collected through the establishment questionnaire.

(3) Plan Questionnaire—The purpose of the mailed Plan Questionnaire is to collect plan-specific information on each plan (up to four plans) offered by establishments that provide health insurance to their employees. This questionnaire obtains information on total premiums, employer and employee contributions to the premium, and plan enrollment for each type of coverage offered—single, employee-plus-one, and family—within a plan. It also asks for information on deductibles, copays, and other plan characteristics.

Please log in or register to answer this question.