Skip Navigation

Medical Expenditure Panel Survey
Medical Provider Component (MEPS-MPC)

Methodology Report 2018 Data Collection

Deliverable OP2-5
Version 0.2
May 2020

Prepared for:
Agency for Healthcare Research and Quality
Peter Tice, PhD
AHRQ Project Officer
AHRQ, Center for Financing, Access & Cost Trends
5600 Fishers Lane
Rockville, Maryland 20857

Prepared by:
RTI International
3040 Cornwallis Road
PO Box 12194
Research Triangle Park, NC 27709-2194

RTI Project Number 0215974.002
AHRQ Contract Number HHSA290201300007C


Table of Contents

1 Introduction
1.1 Changes from 2017 MPC to 2018 MPCC
2 Preparations for the 2018 MPC
2.1 Sample Preparations
2.1.1 Sample files in the 2018 MPC
2.1.2 MPC Sample Delivery from Household Component
2.2 Sample Maintenance
2.2.1 Contact Groups
2.2.2 Provider Type Classification
2.2.3 Priority Code
2.2.4 Fielding the 2018 MPC Sample
2.3 Integrated Data Collection System
2.3.1 Components of the Integrated Data Collection System
2.4 Enhanced Security Network
2.5 Recruiting and Training
3 Data Collection
3.1 Provider Recruitment and Data Collection Procedures
3.1.1 Hospitals
3.1.2 Institutions
3.1.3 Office Based Doctors (OBDs)
3.1.4 Home Health Providers
3.1.5 Pharmacy
3.1.6 Separately Billing Doctors (SBDs)
3.2 Data Abstraction
3.3 Coding Text Fields Collected in the 2018 MPC
3.4 Data Collection Schedule
3.5 Data Collection Results
3.5.1 Completion Rates
3.5.2 Refusal Conversion
3.5.3 Components of MPC Data Collection
3.5.4 Timing

Appendix A: Acronyms and Definitions
Appendix B: MPC Data Collection Summary Tables
Appendix C: Critical Items

Table 2-1 Summary of Design Factors in the Household Component, 2015-2018
Table 3-1 Percent of Participating Contact Groups that Provided Records, 2015-2018
Table 3-2 MPC Data Collection Schedule 2015-2018
Table 3-3 Provider-Level Completion Rates, MPC 2015-2018
Table 3-4 Pair-level Completion Rates, MPC 2015-2018
Table 3-5 SBD Node-Level Completion Rate, MPC 2015-2018
Table 3-6 Refusal Conversion Outcomes: MPC 2015-2018
Table 3-7 Hours per Completed Pair, 2015-2018 MPC

Figure 3-1 Hospital providers: Response factors over time
Figure 3-2 Office-Based providers: Response factors over time
Figure 3-3 SBD providers: Response factors over time
Figure 3-4 Pharmacy providers: Response factors over time

Table B-1 MPC Sample Sizes, Provider Level, 1996-2018
Table B-2 MPC Sample Sizes, Pair Level, 1996-2018
Table B-3 MPC Data Collection Results, Provider Level, 1996-2018
Table B-4 MPC Data Collection Results, Pair Level, 1996-2018


1. Introduction

The Medical Expenditure Panel Survey (MEPS) has been conducted by the Agency for Healthcare Research and Quality (AHRQ) each year since 1996. MEPS is a set of large-scale surveys of families and individuals, their medical providers, and their employers across the United States. MEPS collects data on specific health services, including frequency of use, costs, and sources of payment for services, and on the cost and scope of health insurance covering U.S. workers.

This report describes the methodology of the 2018 Cycle of the MEPS Medical Provider Component (MPC1) . The MEPS-MPC collects data from Hospitals, Office-Based Doctors, Home Health Agencies, Institutions (such as long-term care facilities) and Pharmacies reported by MEPS Household Component (HC) respondents as well as doctors who provide services for patients in Hospitals but bill separately from the Hospital (referred to as Separately Billing Doctors or SBDs). (See Section 2.1 for additional information about provider types.) The MEPS-HC is conducted by Westat, Inc. and the MEPS-MPC is conducted by RTI International and Social & Scientific Systems, Inc. (SSS).

Each cycle, providers for the MPC sample each year are identified in three rounds of HC data collection for two HC panels (see Table 2-1). Overall the HC panel design features five core rounds of interviewing over the course of two full calendar years. The HC collects data from a sample of families and individuals in selected communities across the United States, drawn from a nationally representative subsample of households that participated in the prior year’s National Health Interview Survey (conducted by the National Center for Health Statistics of the Centers for Disease Control and Prevention).

During the household interviews, the HC collects detailed information for each person in the household including demographic characteristics, health conditions, health status, use of medical services, charges and source of payments, access to care, satisfaction with care, health insurance coverage, income, and employment.

The 2018 MPC cycle was conducted by RTI International and SSS under the second option year of the 2017 - 2020 contract awarded by AHRQ to RTI in 2016. RTI completed data collection for Hospitals, Institutions, Office-Based Doctors, Home Health Agencies, and Separately Billing Doctors (SBDs) while SSS completed data collection for Pharmacies. This allocation was initially implemented in the 2013 Cycle to assure that data collection for each provider type was managed consistently within a single operations center.

Return to Table of Contents

1.1 Changes from 2017 MPC to 2018 MPC

Prior to data collection beginning a list of minor recommended Contact Guide and Event Form changes was submitted to AHRQ for review and approval. This included changing the reference year from 2017 to 2018, the removal of "retrievable" as a data entry option, the addition of a skip pattern for VA cases to avoid collecting SBDs, and the addition of one speciality to the SBD dropdown response option. The Pharmacy event form was moved from Hatteras software to Blaise software; the Pharmacy event form also had several edit checks added, including a check for duplicate NDC entries prior to validating the event form, changes to accept HCPCS for durable medical equipment (DME) as preference over NDC codes, as well as the addition of the drug name lookup feature within the Blaise programming. Detailed information about item wording and instrument flow was provided to AHRQ in Deliverable OP2-12, MPC 2018 Final Data Collection Instruments.

Return to Table of Contents

2. Preparations for the 2018 MPC

This chapter describes the 2018 MPC provider sample, preparations for data collection, including grouping patient-provider pairs by provider, grouping providers for the purpose of contacting facilities, and updating locating information.

Return to Table of Contents

2.1 Sample Preparations

Respondents in the HC are asked to identify all medical providers associated with health care services received by each member of the household for the reference period associated with the time period of the interview date. Thus, the basic sample unit in the MPC is a patient-provider pair (referred to as a "pair") where the patient is a member of a household participating in the HC and the provider is identified in the household survey as one associated with a medical event, that is, an office visit, a hospital stay, a prescription for medicine, or other health care event. To facilitate the MPC contacting medical providers household members are asked to sign an Authorization Form (AF) indicating their agreement to allow providers to release information about the event to the MPC. This form is compliant with the Health Insurance Portability and Accountability Act (HIPAA) implemented in 2003.

Within the HC, the term “medical provider” is intended to include any type of practitioner contacted by the household for what the household considers to be health care–hospitals, clinics, long-term care institutions, HMOs, medical doctors and doctors of osteopathy, dentists, home care providers, optometrists, podiatrists, chiropractors, psychologists, and other practitioners.

Eligibility for the MPC is restricted to services rendered in a hospital or by a medical doctor or doctor of osteopathy (MD or DO) or under the supervision of an MD or DO. The MPC excludes services provided by dentists, optometrists, psychologists, podiatrists, chiropractors, and other kinds of health care practitioners who do not provide care under the supervision of a MD or DO. Care provided by home care agencies is an exception to this criterion; the sample design includes all care provided through a home care agency. Pharmacies reported as sources of prescription medicines obtained by household respondents make up the fifith group of MPC pairs generated from the MEPS-HC. However, the MPC excludes pharmacies that provided durable medical equipment (DME) only, and no perscriptions. Finally, additional patient-provider pairs are identified during the MPC data collection as SBDs are identified in medical records obtained from Hospitals and Institutions.

In summary, provider types included in the MPC are:

Hospitals–Providers associated with an inpatient stay as well as hospital outpatient department or emergency room

Institutions–Long-term care providers

Pharmacies–Pharmacies (corporate and non-corporate) where household respondents obtained or purchased prescription medicines

Office Based Doctors (OBDs)–Physicians (MDs and DOs) associated with non-hospital care.

Home Health –Providers associated with care provided in the home of the household respondent, including either health care (Health Agencies) or other services excluding health care (Non-Health Agencies)

Separately Billing Doctors (SBDs)–Providers added to the MPC sample during abstraction of medical and patient account records of hospitals and institutions. Charges and payments for their services are not included in the hospital or institution financial records and must be obtained by contacting the offices of the SBDs.

Return to Table of Contents

2.1.1 Sample files in the 2018 MPC

The HC contractor prepared pair data from the computer assisted personal interview (CAPI) survey instrument used in the HC. For non-Pharmacy pairs, the file includes pairs with eligible dates of utilization (that is, calendar year 2018). In the file for Pharmacy pairs, the events (prescriptions) are not dated. Files for all provider types include the AFs signed by the household respondents. AHRQ subsampled OBDs at the HC Reporting Unit (RU) level, and delivered the extracted MPC sample files to RTI. The 2018 MPC OBD subsampling rate was 54%.

Table 2-1. Summary of Design Factors in the Household Component, 2015-2018

  2015 2016 2017 2018
  Panel 19,
Year 2
(Round 5)
Panel 20,
Year 1
(Round 3)
Panel 20,
Year 2
(Round 5)
Panel 21,
Year 1
(Round 3)
Panel 21,
Year 2
(Round 5)
Panel 22,
Year 1
(Round 3)
Panel 21,
Year 2
(Round 5)

Panel 23,
Year 1
(Round 3)

No. of PSUs for household sample 183 183 183 183 183 168 168 143
No. of household interviews 6,794 7,753 7,421 7,043 6778 6808 na 1 na 1
Subsampling of Office-Based Doctors in CAPI No No No No No No No No
Subsampling of Office-Based Doctors after CAPI Yes Yes Yes Yes Yes Yes Yes Yes

Sources: MEPS Household Component Annual Methodology Report ( July 15, 2019 ) Westat, Inc, Table 1.1 and Table 4.3.

1 The number of completed household interviews for these Panels/Rounds was not available in Table 4.3 of the July 15, 2019 Household Component Methodology Report.

Input to the MPC sample was provided in four separate files.

  1. Records in the main sample file were identified at the pair (PAIRID) level. All other files used to construct and load the sample were merged with this file. This file identified the MPC cases loaded into the IDCS Control System (CS) and tracked throughout the MPC data collection period. For the purposes of data collection in the MPC, the CS tracked at the event level, pair level, and provider level. During the matching process, the MPC data collected was linked back to the pairs from this original HC sample file.
  2. The person file contained identifying information for every household member associated with a pair in the main sample file. The file can be merged with the main sample using the person ID (PERSID).
  3. Provider contact information is contained in the NPI provider directory used by HC interviewers and the monthly non-matched files delivered by Westat containing providers not found in the NPI directory. For providers identified in the directory, the provider ID (PDDIRID) is the NPI ID (NPIPRVID) from the NPI directory. For providers not found in the directory, the provider ID (PDDIRID) is the PROVID assigned by Westat in the monthly files of non-matched providers. Both files contain provider name and contact information. For the non-matched providers, the contact information is the provider name and address that was provided by the HC respondent. The contact information was then loaded into the control system as part of the MPC case.
  4. The Pharmacy directory file can be merged with the main sample file using PHADIRID (same as PDDIRID) so that the name and contact information of the pharmacy can be loaded as part of the pharmacy case.
  5. Beginning with the 2017 HC, a Pharmacy NPI directory was used by the HC interviewers to assign IDs to pharmacies. If a match was found, a pharmacy NPI ID was assigned to the pharmacy reported by the HC respondent. The pharmacy NPI directory was delivered with the sample files and was merged with the main sample file using the Pharmacy NPI ID (NPIPHAID).
  6. Beginning with the 2018 cycle, RTI developed code for assigning pharmacy chain codes by searching for text strings in the pharmacy names.

Return to Table of Contents

2.1.2 MPC Sample Delivery from Household Component

For the 2018 MPC, Westat extracted the sample files used for inclusion in the MPC sample in three waves. Westat delivered the Pharmacy sample files directly to RTI. The non-Pharmacy files were first delivered to AHRQ for OBD subsampling and then forwarded to RTI for processing. Upon the completion of the OBD subsampling, AHRQ delivered the sample files to RTI. The waves of sample files were delivered to RTI in January (Wave 1), April (Wave 2), and July (Wave 3) of 2019. A total of 54,082 patient-provider pairs were in the 2018 MPC sample derived from the HC; 41,323 (76.4%) in Wave 1 of sample delivery; 7,253 (13.4%) in Wave 2, and 5,506 (10.2%) in Wave 3.

The following data elements were included in the MPC sample in order to identify each pair:

  • Unique person and Provider IDs used to link the data collected through the MPC back to the household-generated data for the matching process
  • Identifying information of the household member, such as name, address, gender, and date of birth, parent name if person under age 18, spouse name (if married), and policy holder name for insured persons
  • Identifying information about each provider, such as name, address, and telephone number
  • At the pair level, the number of each type of event identified for the person for that provider and any other HC variables necessary to assign priority flags (see section 2.2.4 below).

These data elements are necessary to define a pair, a key data collection unit of the MPC. The extracted file records were sorted so that all pairs for a provider were listed together, thereby creating provider-level records. (For more information about the data elements included in the extraction files, see the deliverable OP2-6 OP2-8 OP2-9 Sample Plan – 2018 MEPS.)

Return to Table of Contents

2.2 Sample Maintenance

In order to facilitate data collection, RTI sorted providers into contact groups, that is, groups where several providers share the same contact information (e.g., telephone number, practice name, street number, and provider name). Potential groups were carefully reviewed to confirm that grouping was appropriate. In the formation of contact groups, provider identification numbers and other detailed information from the HC were preserved to assure accurate linkages back to the initial sample files. During the MPC data collection, the IDCS enabled contact groups to change as facilities could be restructured, bought out by other entities, or change location of the medical and/or patient account records.

Return to Table of Contents

2.2.1 Contact Groups

All pairs were assigned to contact groups. A pair was assigned to a contact group first by checking whether the provider in the 2018 MPC sample was in a previous MPC sample. If so, the pair was assigned to the provider’s most recent contact group. Providers not found in a previous MPC sample were grouped to form a new contact group based on the provider’s contact information. An automated process was implemented in the 2011 MPC that grouped pairs by telephone number, address fields, and a SOUNDEX program in SAS to identify similar practice or provider names.

As in prior cycles, before delivery of sampled pairs, Westat checked for duplicate pairs based on unique identification numbers assigned to each person (PERSID) and provider (PROVID). The sample preparation process at RTI included further checking for duplicate pairs by searching the sample files for pairs that had the same PERSID and NPI identifier but a different PROVID. When duplicate pairs were identified, one pair was assigned a code that indicated the pair had been merged. This merged code was used to prevent the pair from being fielded. The other pair was fielded for data collection.

An additional check searched pairs within the same RU for instances where pairs had the same provider telephone number (reasoning that in these situations, providers with the same telephone numbers might be the same individual). Suspected duplicate providers were confirmed through manual review of provider names and addresses and, if associated with the same person, merged as above.

All Veterans Administration providers were grouped together because of their common organizational structure that makes them significantly different from the other providers in the sample. Once records were receipted, VA providers were assigned to a small group of Hospital abstractorsso that they could be worked consistently.

Similarly, HMO providers were grouped together and assigned to a small team to coordinate contacts with common corporate offices rather than with the individual providers. This grouping facilitated efficient contacts for recruiting HMO providers into the study and helped to make records abstraction more consistent and efficient.

Return to Table of Contents

2.2.2 Provider Type Classification

Provider type classification in the MPC is critically important operationally for several reasons. Because hospital events were likely to be associated with high expenditures, it was important to track provider type participation to assure that hospital providers are responsive to the survey. Hospitals are often complex environments, especially for data collection projects, and thus the MPC data collection instruments are designed to assist the data collection staff in dealing with multiple points of contact within the hospital and with potentially more complicated medical records and patient account information. The MPC Hospital data collection forms are also designed to facilitate the collection of SBD information associated with hospital events.

Provider type was assigned at both the pair level and the provider level. The initial provider type for the pair was assigned during the HC interview when the household respondent identifies the type of medical events associated with a medical provider. During sample processing, the household provider type is updated. First, labs and dialysis centers, Veterans Administration (VA) providers, imaging centers, and surgery centers are assigned a Hospital provider type. Second, providers will be assigned a Hospital provider type if they were in a Hospital contact group in the previous wave. In the initial processing of Wave 1 sample and contributing to the larger sample size, we identified 1,376 OBD pairs that were converted to the Hospital provider type during MPC sample processing. This anomaly in the sample only presented during Wave 1 and was not an issue for subsequent sample wave releases.

Note, that the provider type assigned during the HC could have been incorrect because of a household respondent's misunderstanding about a provider's status. Typically, this occurred when a household respondent confused hospital and office-based providers. Efforts were made to correct the classification during sample preparation and during the field period.

Following the sorting of provider pairs into contact groups, RTI reviewed the composition of contact groups to see if provider classification at the pair level was consistent within contact group. Inconsistencies, such as an OBD pair in a Hospital contact group, were resolved by creating a new contact group, so that all providers within a contact group were consistent.

In addition, during data collection, staff periodically learned that the provider type was incorrect, and the field was updated so that the appropriate event form could be administered. The most common change was to a Hospital provider from another provider type, typically an OBD provider. This provider type change was important so that the appropriate Hospital event form could be used to collect SBD information. Updating provider type was uncommon among other provider types.

As a result of such provider type changes during sample preparation and during data collection, in the 2018 MPC the count of Hospital pairs increased by 2,204 pairs, an increase of 20% between the count of Hospital pairs in the HC sample and the count at the close of the field period. Among changes to Hospital provider, 90% occurred during sample preparation and 10% during data collection. The overall count of Home Health pairs decreased by 19, a decrease of approximately 2%.The overall count of Institution pairs decreased by 4 (2%) and the overall count of OBD pairs decreased by 2184 (10.8%).

Return to Table of Contents

2.2.3 Priority Code

A priority code was attached to both providers and person/provider pairs. High priority cases include patients or providers expected to be associated with high costs. These priority cases were closely tracked and monitored during MPC data collection using production reports that track the progress of completing these priority cases. Priority flags were attached at the person level to ensure that contact groups with patients having priority flags were given priority by the data collection staff when working MPC cases. Priority flags set at the person-level were rolled up to the provider and contact group levels. A priority flag was set if the person meets one or more of the following criteria:

  • Hospital stay or Home Health event
  • Deceased
  • Institutionalized in a health care facility
  • Outpatient or office visit surgery.

If an SBD was identified in a high priority Hospital pair, the SBD pair was also coded as high priority.

Return to Table of Contents

2.2.4 Fielding the 2018 MPC Sample

The initial 2018 MPC sample (consisting of Hospital, Institution, OBD, Pharmacy, and Home Health pairs identified in the HC) was fielded in three waves following the receipt of each wave from Westat and AHRQ. Given the HC data collection procedures, it is possible for a pair to be included in more than one wave of the MPC sample. Before fielding the second and third wave, each was reviewed to identify pairs that had been included in an earlier wave. When a pair in the new wave matched a pair from an earlier wave and the same event types were reported in both (or all three) waves, the pair was not fielded in the later wave. If different event types are reported, the case is reviewed to determine whether additional data collection is necessary.
(Fielding the SBD sample is discussed in Section 3.1 below.)

Return to Table of Contents

2.3 Integrated Data Collection System

The Integrated Data Collection System (IDCS) supported the 2018 MPC data collection and tracking requirements. Its main purposes were to:

  • Manage and update the provider contact information
  • Collect updated information via telephone, or hardcopy form into one central database
  • Produce reports for project staff as well as AHRQ updating data collection progress at the event, pair, and provider level.
  • Provide a secure model to contain information with RTI’s Enhanced Security Network
  • Produce data files for the matching process.

The IDCS consisted of two main systems. A Web/Windows component was programmed in ASP.Net/Blaise to support the MEPS-MPC Contact Guides and Event Forms for data entry either during telephone calls or record abstraction. A Case Management System (CMS) facilitated call scheduling, contact information, appointment times, and event/status information. The components of the IDCS are described in the following paragraphs.

Return to Table of Contents

2.3.1 Components of the Integrated Data Collection System

Case Management System (CMS)

The CMS provided oversight and control over the MPC sample by tracking pending and final disposition for individual cases and for the aggregate sample. For individual cases, the CMS tracked the completion of data collection by individual medical events, patients, providers and provider practices (contact groups), providing call center supervisors and project staff a tool for measuring progress in completing the varied data collection units in the MPC. At the aggregate level, the CMS produced daily standard or customized reports to track performance of the data collection activity. The CMS was used to monitor production of cases completed via record abstraction as well as by telephone.

Contact Guides

Contact Guides were programmed for each of the major provider types as an aid to recruiting providers. Contact Guides were used to record contact information for several points of contact within a provider organization (e.g., a group practice or hospital) and results of each contact. The Contact Guides included the capability to generate packages of materials, including copies of the patient’s signed AF that were then either faxed or mailed to providers. Starting with the 2017 cycle, a secure portal was also used for sending AF packets to providers and receiving scanned medical records from them. The Guides interacted with the CMS to prompt follow-up contacts with providers after an appropriate time (24 hours for faxed material or or material sent via the webportal; 5 days for mailed material).

Event Forms

Event Forms were used for collecting information either during telephone calls with providers or by abstracting hardcopy medical or patient account records. In the 2009 MPC hardcopy Event Forms were replaced with electronic versions developed for each provider types. The Event Forms were adaptable to the particular format of medical and patient account records. The Event Forms featured edit checks on individual items and were also programmed to alert users to inconsistencies that may be resolved either with telephone respondents or by further investigation in hard copy records. As each Event Form was completed, it was checked for critical items and, if missing, the Form was flagged for follow-up.

Completion of Event Forms was tracked automatically in the CMS to record progress in completing information about medical events, patients, providers, and provider contact groups.

Control System

The Control System managed information flow among the CMS, Contact Guides, and Event Forms and triggered processes based on disposition codes. The Control System imported the provider sample files and arranged information about providers and patient into contact groups to facilitate provider recruiting efforts and data collection. Based on user-selected disposition codes or disposition codes generated automatically, the Control System updated the CMS with pending or final disposition codes. The Control System triggered the production of materials faxed, mailed, or sent via the mailportal to providers (including AFs). It notified data collection staff that these materials had been sent to providers and generated notices for follow-up.

Assignment Transfer

The Assignment Transfer System was used to re-assign cases among the data collection staff. Typically, this was used to reassign a reluctant provider to a more skilled negotiator on the data collection team or to balance workloads among staff. Results of all previous call attempts or entered data were accessible to the new user.

Automated Fax/Mail

Prior to data collection and using the contact information collected by the provider during initial contact, providers were sent (by fax, mail, or webportal) the following materials:

  • Cover sheet
  • Cover letter providing general information about the study from the U.S. Department of Health and Human Services
  • Brochure that addresses commonly asked questions about the MEPS-MPC study
  • Patient List of all MEPS-HC respondents who reported receiving services from the provider
  • AF for each patient on the Patient List
  • Return form used by the respondent when they preferred to fax, mail, or send via webportal their medical and patient account records for hardcopy abstraction. The fax and webportal return cover sheet contained pre-printed information for faxing/transmitting records. The mail return form includes a pre-printed mailing label for the provider to send via mail.

Return to Table of Contents

2.4 Enhanced Security Network

All files containing personally identifiable information (PII) or protected health information (PHI) were stored and managed within the FIPS-Moderate Enhanced Security Network (ESN), a network developed by RTI to meet the security requirements of NIST SP 800-53, Rev.4, Recommended Security Controls for Federal Information Systems and Organizations at the Moderate level (http://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-53r4.pdf). A key IDCS security feature provided access to the Web interface based on the login attributes assigned to individual users.

Return to Table of Contents

2.5 Recruiting and Training

Data collection specialists (DCSs) were the “front-line” staff charged with recruiting medical providers and abstracting medical event level from medical and payment records. Abstracting this information could be completed either over the telephone in interviews with provider staff or by abstracting hard-copy records sent in by providers. Separate training modules were administered to emphasize the different skills necessary to complete data collection in either mode. Although some DCSs developed expertise in either one or the other mode, many DCSs were cross-trained for either telephone or hard-copy abstraction methods.

Return to Table of Contents

3. Data Collection

In the 2018 MPC, the project team continued to follow a core protocol for collecting information from the provider types. The protocol was customized in the event forms to address the unique challenges of each provider type. Project procedures were designed to make data collection as efficient as possible for the providers and DCSs.

As noted in Section 2.1, the pairs in the sample files were sorted by provider. In addition, providers who appeared to work in the same practice were sorted into contact groups to minimize the number of contact attempts with individual providers.

As part of the initial communication with each contact group, the DCS identified appropriate individuals as points of contact (POCs) to facilitate data collection completion. The Contact Guide was designed to enable DCS staff to record the outcome of each contact attempt and allowing supervisors and project staff to review the provider group contact history prior to subsequent contact attempts. DCSs were assigned a set of provider contact groups so that they could establish rapport with contacts in each provider group. If any cooperation or staffing issues arose, cases were reassigned to refusal converters. During initial contacts, DCSs performed several tasks:

  • Introduce the study
  • Confirm the provider groupings in the initial assignment
  • Identify the provider staff who can fulfill our requests
  • Obtain fax numbers, addresses, or emails for sending project materials
  • Negotiate the manner in which data collection proceeds
  • Determine whether the facility charges a fee for providing records.

Depending on the size and complexity of the provider practice these tasks may have been completed in a single call or over several calls with different points of contact.

Return to Table of Contents

3.1 Provider Recruitment and Data Collection Procedures

While the MPC includes data collection procedures core or common to all the provider types, each provider type also included unique features and specific procedures DCS are required to follow. The sections below describe the MPC data collection protocols and procedures for each provider type.

Return to Table of Contents

3.1.1 Hospitals

Data collection procedures were designed to be flexible in adapting to particular situations in provider facilities while maintaining consistency in the data obtained. DCSs typically contacted three hospital departments: medical records, patient accounts, and the administrative office. After the hospital received a provider information packet, the DCS re-contacted the medical records department. Because of the length and complexity of Hospital records and because Hospital providers were often associated with multiple pairs, sending records for abstraction by RTI was standard protocol. In a small percentage of cases (about 7% of medical records and 19% of patient accounts, see Table 3-1) was collected by telephone. This mode was also a preference so that records were available for quality assurance purposes.

Four key pieces of information were obtained from the hospital medical records:

  • Date(s) of service
  • Event type (ER, outpatient, inpatient)
  • Diagnoses (ICD-10 codes), and
  • Names and specialties of any health professionals who saw the patient during the hospital event and who charged for services separately from the hospital’s billing record (SBDs).

Concurrent with the request for this information, the DCS also contacted the patient accounts department to collect the services provided, charges, and sources and amounts of payment for each event identified. Finally, after records abstraction was completed, the DCS contacted the hospital’s administrative offices to obtain the billing status of each health professional identified by the medical records and contact information for confirmed SBDs.

Return to Table of Contents

3.1.2 Institutions

The procedures for Institutional care settings were similar to that for Hospitals. The institutional sample consisted of the long-term health care facilities, such as skilled nursing or rehabilitation facilities. Non-profit organizations are excluded.

Return to Table of Contents

3.1.3 Office Based Doctors (OBDs)

Compared with hospital providers, the information required from OBD practices was often less complicated. In addition, OBDs were typically associated with fewer pairs than hospital providers. For both reasons, OBD data collection was more amenable to telephone data collection and DCSs encouraged OBD providers to give information during the telephone contact when they had few patient records or only a few events to report (e.g., XX patient records or XX events). The Contact Guide was designed to factor in OBDs who use off-site billing services. DCSs were trained to collect information from off-site billing services during their contacts.

Return to Table of Contents

3.1.4 Home Health Providers

Data collection for home health providers followed the same basic protocol as the OBD sample. In certain cases, the DCSs contacted social service agencies or corporate offices in order to locate the necessary records. The home health event form was initially programmed for the 2009 MPC to conform to Medicare Home Health Prospective Payment System. The system allowed the option of collecting payment data in 2-month or 1-month time frame as appropriate.

Return to Table of Contents

3.1.5 Pharmacy

For small retail pharmacies unassociated with a chain, and for pharmacies associated with small chains, the DCS contacted the pharmacy to explain the study’s purpose and determine if patient profiles were available. If they were, the DCS verified that the profile contained required data elements. If patient profiles were not available or if the profiles did not contain all of the required data, the DCS collected the information by telephone or requested supplemental reports from the pharmacist. Pharmacy data was received in any format including hardcopy patient profiles, electronic files with patient profile data, and/or collecting or supplementing the profiles by telephone data collection.

For large retail pharmacy chains, individual pharmacies were grouped by chain using a unique code. Historical contact information from earlier data collection years was reviewed for each chain to develop a contact approach. Specially trained negotiators followed-up in one of two basic ways:

  • If the corporate office preferred to collect data from the local stores the data collection followed the small retail model. However, an endorsement from the corporate office was requested to be included with each contact packet.
  • If the pharmacy preferred the data request to be handled with a regional or central contact, the negotiator facilitated the most efficient method for data collection.

Return to Table of Contents

3.1.6 Separately Billing Doctors (SBDs)

Hospital, Institution, OBD, Home Health Agency, and Pharmacy providers were all identified by household respondents during the MEPS–HC. The balance of the MEPS-MPC sample consisted of physicians (reported by Hospitals and Institutions) who provide services during a Hospital or Institution-based event. These events often result in charges from providers who may or may not have direct patient contact (e.g., pathologists or radiologists) and whose fees may or may not be included in the hospital charge. These charges are a key part of hospital event costs, and this information can only be obtained from the MPC.

From all doctor names abstracted from the medical record, DCSs contacted the Hospital medical records or professional staffing department to confirm the SBD status. Either working with medical records personnel by telephone or from hard copy records, the DCS recorded each provider who provided any services and whose charge might not have been included in the hospital charge. The DCS then contacted the hospital’s administrative office to verify that the SBD billed separately. If there was any possibility of a separate charge, the DCS obtained complete contact information and created a link within IDCS to connect the Hospital provider, patient, event type, event date, and SBD. This link is referred to as a node, that is, a unique combination of hospital, patient, event type, event date, and SBD provider.

Similar to prior MPC cycles, fielded SBD nodes that were fielded were based on a priority status that was expected to yield nodes more likely to be eligible and to be associated with higher charges.  Beginning with the 2015 Cycle, the priority status definition used prior to then was revised based on a modelling exercise using 2013 and 2012 data. High, medium, and low priority nodes were defined using the following criteria:

  • High priority included nodes associated with a hospital stay or institutional care with a role code of active physician/providing direct care (excluding radiology and pathology specialty codes), or nodes with a specialty code of surgery or anesthesiology.
  • Medium priority included those not in the high group with a hospital stay or institutional care with a role code not active physician/ providing care, or specialty codes of radiology with active role, or specialty codes of pathology with active role excluding those with CPT codes only within the range of 80000-85999, or all other nodes with role code of active physician/providing direct care.
  • Low priority included all other nodes.

These criteria for assigning priority status were applied to the 2018 MPC. However, due to budget constraints, a smaller number of SBD nodes was fielded in 2018 compared to earlier years. Nodes were sampled for fielding such that the sample contained 25% high priority, 65% medium, and 10% low priority nodes. This was a slight oversample of high and medium priority nodes. The sample was constructed such that all nodes in a pair were fielded. Release of SBD pairs emphasized high priority nodes so that SBD providers and billing services would have ample time to respond. In the first wave of the SBD sample, a subsample of high and medium priority nodes, as well as low priority nodes that were included in contact groups along with high and medium priority nodes, were released and only these nodes were included in the requests to the providers and billing services. A subsample of all nodes was released in the second and third waves. Three waves were used in the 2018 SBD fielded, compared to four waves in the 2017 SBD sample. 

Prior to SDB sample release and data collection a computer algorithm was used to identify instances of overlapping OBD and SBD providers. The OBD and SBD provider identification numbers were required to be the same in order to be considered a match by the computer algorithm. Four situations were considered (node counts are from the set of nodes selected for data collection, that is, those that were held from data collection because they were low priority are not included in the counts reported in this section):

  1. Direct node match—As in recent previous cycles, nodes were filled using the overlap pair with an S-code event (that is, an inpatient, ER, or outpatient event) on the same date at the node. The following situations were also used to automatically link OBD and SBD nodes: Events where the OBD location of service is a physician's office and the SBD location is outpatient, dates of service are the same, and charges and payments are not the same; events where the SBD location is an inpatient and the OBD date of service is within the range of the inpatient stay (excluding fist and last day); and events where the SBD location of service is either outpatient or inpatient, the CPT4 codes for the OBD are associated with Hospital events and are not used in ambulatory settings, and the date of service is either the same for an outpatient event or within the date range of the inpatient event, including the first and last day of the stay. In the 2018 Cycle, 75 nodes were identified as a direct node match.
  2. Systematic coding of obvious disavowal nodes—For a large proportion of the nodes associated with an OBD pair with various types of specialty services with a date close to or the same as an OBD event, often the role of the SBD is “referring or copied doc.” Some examples of this situation are an office visit with an OB/GYN followed closely by a mammogram; an office visit with an internist preceded by a blood panel; and an office visit with an orthopedist followed closely by an x-ray.

    The specification used to identify the disavowal nodes were as follows:
    • If the OBD overlap pair does not have an S-code event within 2 weeks plus or minus of the SBD node, and
    • The node is either radiology or pathology (as defined by CPT4 codes that begin with a “7” or “8” or any BETOS code in categories 3-Imaging or 4-Tests, and
    • There is a regular OBD event (defined by CPT4 code that begins with a 99 or a BETOS code of M1A or M1B) within 2 weeks plus or minus of the SBD node (i.e., within 14 days before or 14 days after).

    The node was automatically coded as a referring/copied doc when all three of these conditions were met.

    If all OBD events have location of service as physician office, all OBD events have CPT 4 codes that are part of the evaluation/management series, and the SBD role is anything other than department head/followup, the SBD was coded as a referring/copied doc.

    If all OBD events have location of service as physician office, all OBD events have CPT 4 codes that are part of the evaluation/management series, and the SBD role is specified as department head/followup-doc, then the SBD node was coded as department head/followup doc.

    In the 2018 cycle, 253 nodes were coded as disavowals.

  3. If the overlap pair was a refusal during OBD data collection, the node was automatically coded as a refusal. In the 2018 cycle, 27 nodes were identified as refusals based on a match to a refusing OBD.
  4. Nodes were also reviewed to determine if any were abstracted in error. The logic for identifying these was when the OBD location of service is physician's office, the SBD location is outpatient, the dates of services are the same, and the charges and payments are identical. In the 2018 Cycle, 1 node was identified as abstracted in error.

Remaining nodes where the SBDs and OBDs were associated with different provider IDs were reviewed by senior project staff to determine whether to field the node or not and, if not fielded, code the node status. In the 2018 Cycle, 2,629 nodes were reviewed and, of these, 1,944 (74%) were not fielded and resolved as follows:

  • Included in an OBD, that is, a direct match that was not identified in the automated process (401 nodes)
  • Disavowal (1,516 nodes)
    • Type 2 Disavowal (14 nodes)
    • Referred or copied physician (1,467 nodes)
    • Department head of follow-up (35 nodes)
  • Abstracted in error (20 nodes)
  • Included in another SBD (6 nodes)
  • Included in Hospital bill (1 node)
  • Node is part of a global fee where charges were captured on another date, that is, node is a leaf, (0 nodes)

These procedures for identifying SBD-OBD overlap in the manual review were similar to those used for the automated review, except the manual review looked across the entire SBD contract group (instead of being restricted to OBD and SBD providers with the same provider identification number.  In addition to these rules, the SBD was coded as abstracted in error if the SBD should not have been recorded during the hospital stay because the specialty (such as "nurse") was included in the Hospital event charges. 

As a step in the preparation of the SBD sample, we attempted to match all SBD providers to a National Provider Identifier (NPI) in order to assign an identification number. In many instances, the provider's NPI was included in the records and was abstracted into the event form. If the NPI was not in the record, DCSs looked up the number in the NPI Registry. SBD providers that could not be associated with an NPI were assigned a unique identifier in the same format as the NPI. The NPI Registry includes both individual and organizational providers.

Prior to the 2013 MPC, if the NPI number was not found in the record, protocols for computer matching and manual look-up was followed to find the identification number in the NPI Registry. If a match could not be found in the NPI Registry, a number similar in format was assigned to the SBD record. In the 2012 Cycle, we assessed the value of the manual look-up and determined very few NPI numbers were identified from this process. Beginning with the 2013 Cycle, RTI has used computer algorithms to match SBD records to the NPI. Autocoding using a strict sequence of criteria was used to attempt to match to both individual and organizational NPI numbers. As in prior cycles, if a match could not be found in the NPI Registry, a number similar in format was assigned to the SBD record.

Return to Table of Contents

3.2 Data Abstraction

Once the provider acknowledged receipt of the authorization forms, the DCS either collected information over the telephone through electronic event forms specific to each provider type or made arrangements to receive hardcopy medical records and patient account information.

Table 3.1 displays the proportion of participating Hospital, OBD, and SBD contact groups2 that elected to participate by sending in medical records and patient account information for abstracting. Reflecting the preference for collecting Hospital records for abstraction, in the 2018 Cycle most Hospital contact groups 92.6%, participated by providing medical records for abstraction; 81% of Hospitals provided patient account records. In both OBD and SBD contact groups, protocols about collecting data over telephone were more flexible than in Hospitals. Close to half (47.3%) of OBD contact groups provided records and 30.5% of SBD contact groups provided records.

The distribution in 2018 Cycle continues to reflect emphasis on Hospital records abstraction and on telephone data collection for OBDs. Because Hospital records tend to be lengthy and because of the number of patients involved in the record requests, hospitals generally prefer to participate in the MPC by sending records rather than providing data over the telephone. This is also beneficial from a data quality perspective because the Hospital protocol can result in a great deal of information and availability of hard copy records for review is helpful to assuring comprehensive and accurate abstraction. Information obtained from OBD and SBD contact groups is more straightforward and more amenable to telephone data collection which can be less burdensome to providers as well as a more efficient mode for uncomplicated billing situations.

Return to Table of Contents

Table 3-1. Percent of Participating Contact Groups that Provided Records 2015 - 2018

Provider Type Participating Contact Groups Groups Providing Records Percent
2015
Hospital—Medical Records 3,110 2,766 88.9%
Hospital—Patient Accounts 3,110 2,518 81.0%
Office-Based Doctors 8,369 3,697 44.2%
Separately Billing Doctors 5,087 1,502 29.5%
2016
Hospital—Medical Records 3,009 2,694 89.5%
Hospital—Patient Accounts 3,009 2,370 78.8%
Office-Based Doctors 8,824 3,929 44.5%
Separately Billing Doctors 5,100 1,736 34.0%
2017
Hospital—Medical Records 3,548 3,287 92.6%
Hospital—Patient Accounts 3,548 2,856 80.5%
Office-Based Doctors 10,624 4,801 45.2%
Separately Billing Doctors 3,719 1,136 30.5%
2018
Hospital—Medical Records 3,503 3,245 92.6%
Hospital—Patient Accounts 3,503 2,838 81.0%
Office-Based Doctors 9,256 4,374 47.3%
Separately Billing Doctors 3,634 1,126 31.0%

Return to Table of Contents

3.3 Coding Text Fields Collected in the 2018 MPC

Standard coding systems supported the coding of free text for the following types data:

  • Medical Conditions–verbatim text coded to the International Classification of Disease (ICD-10); additional classifications of these codes employed Clinical Software Coding (CCSMATCH) during final file preparations
  • Medical Procedures and Supplies–verbatim text coded to Berenson–Eggers Type of Service (BETOS) codes
  • Non-Pharmacy Sources of Payment–coded to AHRQ-supplied classification (SOP)
  • Pharmacy Sources of Payment–coded to AHRQ-supplied classification (RxSOP)
  • Prescribed Medicines–verbatim text coded to the General Product Identifier (GPI-9)
  • Separately Billing Doctors–verbatim text recording name, practice, and location information was used to assign an identifier from the National Provider Identifier Registry (NPI)
  • SBD Speciality–Specialties of SBD were coded to a specialty classification
  • Location of Service–coded.

Sources of payment (SOP) and SBD information were coded by RTI staff using coding schemes developed and used in previous MPC cycles; sources of payment data (RxSOP) for Pharmacy was coded by SSS staff. RTI also completed location of service and CCSMATCH coding as part of file preparations prior to matching. Coding for conditions (ICD-10), procedures and supplies (BETOS) was completed by Health Care Resolution Service (HCRS) a firm in Laurel, MD, with extensive medical coding experience. SSS was responsible for coding prescribed drugs. More detailed discussions may be found in Deliverable OP2-30 2018 Coding Plan and Deliverable OP2-28 2018 MPC: Plan for MPC to HC Events.

Return to Table of Contents

3.4 Data Collection Schedule

Table 3-2 summarizes the 2015-2018 MPC data collection schedules. Similar to recent cycles, the 2018 MPC sample was provided from the HC in three waves and fielded as such. Since the 2013 MPC cycle the SBD sample, developed during MPC data collection, has been fielded in four waves However, given the workflow of Hospital data collection during the 2018 cycle, only three SBD sample waves were fielded to ensure an adequate amount of sample available for processing at each wave..

Table 3-2. MPC Data Collection Schedule 2015-2018

Provider Type Start of first MPC wave Start of last MPC Wave End of MPC data collection Number of Waves Total Weeks
2015
Hospital 01/29/2016 07/22/2016 10/14/2016 3 37
Office-Based Doctors 01/29/2016 07/22/2016 10/14/2016 3 37
Institution 01/29/2016 08/05/2016 10/14/2016 3 37
Home Health Agencies 01/29/2016 08/05/2016 10/14/2016 3 37
Pharmacies 01/20/2016 07/15/2016 11/18/2016 3 42
SBDs 07/22/2016 11/18/2016 01/13/2017 4 23
2016
Hospital 02/01/2017 08/01/2017 10/13/2017 3 37
Office-Based Doctors 02/01/2017 08/01/2017 10/13/2017 3 37
Institution 03/08/2017 08/07/2017 10/13/2017 3 32
Home Health Agencies 03/08/2017 08/07/2017 10/13/2017 3 32
Pharmacies 02/01/2017 07/24/2017 11/03/2107 3 40
SBDs 08/01/2017 11/16/2017 01/12/2018 4 24
2017
Hospital 02/01/2018 07/30/2018 10/12/2018 3 37
Office-Based Doctors 02/06/2018 07/30/2018 10/12/2018 3 36
Institution 03/08/2018 07/25/2018 10/12/2018 3 32
Home Health Agencies 03/02/2018 07/25/2018 10/12/2018 3 33
Pharmacies 01/29/2018 07/17/2018 10/24/2018 3 39
SBDs 08/27/2018 11/20/2018 01/11/2019 4 20
2018
Hospital 02/01/2019 07/23/2019 10/11/2019 3 37
Office-Based Doctors 02/01/2019 07/23/2019 10/11/2019 3 36
Institution 03/04/2019 07/25/2019 10/11/2019 3 32
Home Health Agencies 2/27/2019 07/25/2019 10/11/2019 3 33
Pharmacies 01/28/2019 07/18/2019 10/23/2019 3 39
SBDs 08/22/2019 11/15/2019 01/10/2020 3 20

Following data collection, additional editing of the files preceded file preparation and matching tasks. These steps have been implemented to assure data quality and consistency in the data across survey years.

Return to Table of Contents

3.5 Data Collection Results

3.5.1 Completion Rates

The MPC applies the following criteria to assess or determine whether an event is complete or partially complete. (see Appendix C for a full discussion of critical items). The final event level codes determine the final pair disposition.

Criteria for non-Pharmacy Providers. In order for a pair to be considered partially complete, at least one event in that pair had to have a valid response for all critical items, that is no critical item in that event could contain a ”don’t know,“ ”refusal,“ or ”missing“ entry (see Appendix C for a full discussion of critical items). In the criteria under consideration, if one critical item in the event has a ”don’t know,“ ”refusal,“ or ”missing“ entry, the event would be assigned a new disposition code ”final critical item missing.“ If all the events in a pair have this new disposition, the pair is considered a partial record and becomes eligible for matching. As pairs roll up to the provider level, some providers that would have a final disposition of non-response under the former criteria would have a final disposition of partial complete under the revised criteria.

Criteria for Pharmacy Providers. As with other providers, for a pair to be considered partially completed, it must include an event where critical items contain valid data. Three additional categories take account of response to three data elements:  Patient Amount, Third Party Payment Source, and Third Party Payment Amount:

  • If Patient Amount was missing but at least one of the other two variables was complete, the event was assigned to Partial Category A.
  • If Patient Amount was complete, but either of the other two variables was missing, the event was assigned to Partial Category B.
  • If both Patient Amount and Third Party Payment Source were complete but Third Party Payment Amount was missing, the event was assigned to Partial Category C.

The 2018 MPC cycle target completion rates were the same as the 2017 goals, with pair target completion rates of 88% for Hospital, 80% for OBD, 90% for HMO, Home Health, and Institution, and 85% for Pharmacy providers. The SBD completion rate goal was 60% of fielded SBD nodes, which was estimated to be 12,000 completed nodes. Table 3-3 displays the provider-level results and Table 3-4 the pair-level results for the 2015 through 2018 MPC cycles. For Hospitals, OBDs, Home Care, Institutions, and SBDs, the provider-level and pair-level completion rates are about the same. However, for HMOs and Pharmacies, the provider-level completion rates are consistently higher.

In the 2018 MPC, the target pair-level completion rate was achieved among OBD, Institution and Pharmacy (overall) pairs. The pair level completion rate for SBD was above the target rate but did not yield the targeted 12,000 completed nodes. Among other provider types, the final pair completion rates were below targets (Table 3-4). Deliverable OP2-15 Evaluation of 2018 Data Collection Plans addresses several key factors that likely contributed to the actual 2018 cycle completion rates falling short of the targets on some provider types.

Return to Table of Contents

Table 3-3. Provider-Level Completion Rates, MPC 2015—2018

Provider Initial sample after subsampling Final eligible sample Completion rate Refusal rate Other nonresponse rate3
2015
Hospitals 6,719 6,323 0.811 0.053 0.136
Office-based providers 13,056 11,957 0.849 0.039 0.113
HMOs 358 343 0.813   0.187
Home care providers 890 728 0.794 0.008 0.198
Institutions 140 129 0.884 - 0.116
SBDs 33,351 19,786 0.591 0.000 0.408
Pharmacies 9,001 8,206 0.881 0.003 0.116
Total 63,515 47,472      
2016
Hospitals 6,609 6,170 0.861 0.024 0.116
Office-based providers 14,055 12,903 0.869 0.020 0.111
HMOs 375 323 0.833 0.000 0.167
Home care providers 908 763 0.847 0.007 0.147
Institutions 131 128 0.906 0.000 0.094
SBDs 34,627 22,573 0.549 0.036 0.415
Pharmacies 8,457 7,637 0.906 0.001 0.093
Total 65,162 50,497      
2017
Hospitals 7,026 6,551 0.879 0.006 0.115
Office-based providers 16,839 15,105 0.824 0.007 0.168
HMOs 369 323 0.910 0.000 0.090
Home care providers 858 713 0.851 0.000 0.149
Institutions 168 161 0.913 0.000 0.087
SBDs 20,936 12,825 0.670 0.000 0.330
Pharmacies 10,531 9,324 0.541 0.000 0.128
Total 56,727 45,002      
2018
Hospitals 7,970 7,321 0.881 0.005 0.114
Office-based providers 15,449 13,677 0.820 0.003 0.177
HMOs 331 299 0.890 0.000 0.110
Home care providers 952 838 0.850 0.001 0.149
Institutions 184 166 0.910 0.000 0.090
SBDs 20,002 11,827 0.682 0.001 0.317
Pharmacies 12,763 11,234 0.896 0.013 0.091
Total 57,651 45,362      

3 “Other nonresponse” includes unlocatable, type 1 disavowal, and other nonresponse.

Return to Table of Contents

Table 3-4. Pair-level Completion Rates, MPC 2015—2018

Patient-provider pair Initial sample after subsampling Final eligible sample Completion rate Refusal rate Other nonresponse rate5
2015
Hospitals 11,225 10,412 0.805 0.093 0.102
Office-based providers 16,727 15,338 0.845 0.082 0.073
HMOs 833 752 0.742   0.258
Home care providers 957 773 0.796 0.106 0.098
Institutions 147 134 0.888 0.052 0.060
SBDs 41,981 24,610 0.567 0.048 0.385
Pharmacies 20,826 18,415 0.832 0.023 0.145
Total 92,696 70,434      
2016
Hospitals 11,088 10,162 0.851 0.081 0.068
Office-based providers 18,445 16,927 0.861 0.070 0.069
HMOs 905 790 0.766 - 0.234
Home care providers 984 817 0.841 0.111 0.048
Institutions 134 131 0.908 0.046 0.046
SBDs 42,951 27,490 0.539 0.050 0.412
Pharmacies 20,218 17,366 0.850 0.067 0.083
Total 94,725 73,683      
2017
Hospitals 11,059 10,171 0.870 0.048 0.082
Office-based providers 19,382 17,370 0.820 0.036 0.144
HMOs 704 577 0.896 0.000 0.104
Home care providers 920 768 0.850 0.073 0.077
Institutions 173 166 0.916 0.018 0.066
SBDs 23,603 14,437 0.661 0.072 0.267
Pharmacies 19,262 16,735 0.858 0.025 0.117
Total 75,103 60,224      
2018
Hospitals 12,979 11,689 0.877 0.028 0.095
Office-based providers 18,256 16,166 0.824 0.036 0.140
HMOs 576 490 0.855 0.043 0.102
Home care providers 1,032 906 0.849 0.044 0.107
Institutions 191 169 0.905 0.018 0.077
SBDs 22,775 13,313 0.680 0.050 0.270
Pharmacies 20,872 17,744 0.878 0.050 0.072
Total 76,681 60,477      

5 “Other nonresponse” includes unlocatable, type 1 disavowal, and other nonresponse.

Table 3-5 presents SBD node-level results. A total of 354,994 nodes were released for data collection in the 2018 Cycle. Of these, 51.5% were confirmed as ineligible nodes (that is, no charges were recorded for that provider). Of the remaining 17,463 nodes (48.5% of the total), additional information was obtained for 10,713 nodes for a completion rate of 61.4%. Among eligible high priority nodes (n =2,776), the completion rate was 63.4%; among medium priority nodes, the completion rate was 62.1% (n =7,664); and among low priority nodes, 37.0% (n =273).

Return to Table of Contents

Table 3-5. SBD Node-Level Completion Rate, MPC 2015 – 2018

  2015 2016 2017 2018
Total nodes 64,581 66,614 34,990
35,994
Ineligible nodes 33,885 30,386 16,641 18,531
Eligible nodes 30,696 36,228 18,349 17,463
Completed nodes 16,093 17,381 10,982 10,713
Nonresponse7 14,603 18,847 7,367 6,750
Eligibility rate 47.53% 54.38% 52.44% 48.52%
Completion rate 52.43% 47.98% 59.85% 61.35%

7 In the reports for previous cycles, nodes with a pending disposition at the close of data collection (empty nodes) were reported separately. In this table, nodes with final dispositions of "pending" and "refusal" are combined into the "Nonresponse" row.

Return to Table of Contents

3.5.2 Refusal Conversion

Table 3-6 provides additional information about refusal conversion for the 2015-2018 MPC cycles. The analytic unit in this table is contact group, an operational unit which may consist of several providers who share facilities for medicals records and billing (e.g. a medical group practice with several physicians or a health care system with several Hospitals). The final column in this table displays the percent of initial refusals that were converted to a participating or partially participating contact group (i.e., provided all or some of the requested information). The 2018 MPC cycle refusal conversion rates by provider type were: 65.2% for Hospital, 33.35% for OBD, 24.4% for Pharmacy, 26.9% for Home Health, and 22.2% for SBD.

Return to Table of Contents

Table 3-6. Refusal Conversion Outcomes: MPC 2015 - 2018

Contact Group Provider Type Initial Sample8 Ever coded Refusal Ineligible Final Refusal Other Nonresponse Complete
  N N Pct of Initial Sample Pct of Ever Coded Refusal N Pct of Ever Coded Refusal N Pct of Ever Coded Refusal N Pct of Ever Coded Refusal N Pct of Ever Coded Refusal
2015
Hospital 3,756 350 9.3% 100.0% 6 1.7% 37 10.6% 122 34.9% 185 52.9%
Office-based 10,320 886 8.6% 100.0% 30 3.4% 210 23.7% 333 37.6% 313 35.3%
Pharmacy 2,520 184 7.3% 100.0% 7 3.8% 8 4.3% 101 54.9% 68 37.0%
Home Health 10,320 1099 10.6% 100.0% 213 19.4% 7 0.6% 620 56.4% 259 23.6%
SBDs 827 75 9.1% 100.0% 9 12.0% 2 2.7% 54 72.0% 10 13.3%
2016
Hospital 3,446 421 12.2% 100.0% 9 2.1% 54 12.8% 83 19.7% 275 65.3%
Office-based 10,567 1019 9.6% 100.0% 36 3.5% 179 17.6% 363 35.6% 441 43.3%
Pharmacy 2,262 108 4.8% 100.0% 6 5.6% 1 0.9% 59 54.6% 42 38.9%
Home Health 10,567 960 9.1% 100.0% 61 6.4% 329 34.3% 357 37.2% 213 22.2%
SBDs 842 83 9.9% 100.0% 10 12.0% 2 2.4% 53 63.9% 18 21.7%
2017
Hospital 4,085 377 9.2% 100% 11 2.9% 4 1.0% 106 28.1% 256 67.9%
Office-based 13,500 1009 7.4% 100% 26 2.6% 55 5.4% 612 60.6% 316 31.3%
Pharmacy 2,437 91 3.7% 100% 9 9.9%     61 67.0% 21 23.1%
Home Health 800 76 9.5% 100% 20 2.6%     39 51.3% 17 22.4%
SBDs 9,663 497 5.1% 100% 5 1.0%     369 74.2% 93 18.7%
2018
Hospital  4,090 423 10.3% 100.0% 16 3.8% 3 0.7% 128 30.3% 276 65.2%
Office-based 12,331 970 7.9% 100.0% 80 8.2% 11 1.1% 554 57.1% 325 33.5%
Pharmacy 2,361 127 5.4% 100.0% 20 15.7% 41 32.3% 35 27.6% 31 24.4%
Home Health 10,258 524 5.1% 100.0% 42 8.0% 7 1.3% 334 63.7% 141 26.9%
SBDs 913 54 5.9% 100.0% 9 16.7% 0 0.0% 33 61.1% 12 22.2%

8 Note counts in this table are of contact groups, not individual providers.

Return to Table of Contents

3.5.3 Components of MPC Data Collection

Figures 3-1 through 3-4 display historical MPC data collection information at the provider level for Hospitals, OBDs, SBDs, and Pharmacies (corporate and non-corporate). Each graph displays:

  • Provider sample size (eligible providers), as a proportion of the eligible sample in 2002
  • Provider ineligibility rate, expressed as the complement of the eligibility rate (1- (Eligibility Rate)) for presentation purposes,
  • Final provider completion rate, and
  • Final provider refusal rate.

For Hospitals, (Figure 3-1), the sample size increased from the previous year, the ineligibility rate went up slightly, and the completion rate and refusal rate were about the same.

For Office-Based Doctors (Figure 3-2), the total sample increased from the previous year, the ineligibility rate went up slightly, while the completion rate and refusal rate decreased slightly.

For Separately-Billing Doctors (Figure 3-3), the sample size was smaller than the previous year, the ineligibility rate and completion rate increased, while the refusal rate was about the same.

For Pharmacies (Figure 3-4), the sample size was much larger than the previous year while the ineligibility rate, completion rate, and refusal rate increased slightly.

Return to Table of Contents

Figure 3-1: Hospital providers - Response factors over time

Image displaying response factors over time, from 1996 through 2018 for Hospital providers

Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Sample Rel to 2002 0.526 0.658 0.513 0.519 0.548 0.822 1.000 0.882 0.897 0.885 0.867 0.842 0.755 1.018 0.802 0.859 0.932 0.915 0.954 1.000 0.975 1.035
1.157
Ineligibility Rate 0.023 -0.024 0.064 0.068 0.078 0.074 0.067 0.074 0.069 0.076 0.068 0.067 0.068 0.129 0.088 0.099 0.050 0.054 0.064 0.059 0.066 0.068 0.081
Completion Rate 0.951 0.894 0.939 0.926 0.910 0.912 0.900 0.898 0.920 0.931 0.941 0.944 0.946 0.890 0.846 0.900 0.870 0.877 0.848 0.811 0.861 0.878 0.881
Final Refusal Rate 0.021 0.058 0.025 0.036 0.037 0.038 0.048 0.047 0.027 0.026 0.022 0.023 0.022 0.012 0.034 0.016 0.015 0.036 0.001 0.053 0.024 0.006 0.005

Return to Table of Contents

Figure 3-2: Office-Based providers - Response factors over time

Image displaying response factors over time, from 1996 through 2018 for office based providers

Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Sample Rel to 2002 0.568 0.516 0.539 0.592 0.818 1.324 1.000 1.011 1.324 1.238 0.884 0.988 0.698 0.670 0.765 0.745 1.030 0.970 1.165 0.876 0.945 1.106
1.002
Ineligibility Rate 0.256 0.271 0.125 0.122 0.138 0.125 0.103 0.101 0.106 0.107 0.105 0.117 0.114 0.106 0.118 0.117 0.110 0.110 0.112 0.084 0.082 0.103 0.115
Completion Rate 0.881 0.871 0.861 0.888 0.864 0.850 0.837 0.835 0.864 0.859 0.869 0.875 0.891 0.801 0.806 0.889 0.876 0.890 0.865 0.849 0.869 0.824 0.820
Final Refusal Rate 0.069 0.053 0.043 0.053 0.071 0.069 0.097 0.095 0.076 0.086 0.074 0.077 0.067 0.003 0.062 0.023 0.028 0.036 0.001 0.039 0.020 0.007 0.00271

Return to Table of Contents

Figure 3-3: SBD providers - Response factors over time

Image displaying response factors over time, from 1996 through 2018 for SBD providers

Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Sample Rel to 2002 0.623 0.379 0.551 0.521 0.503 0.922 1.000 0.870 0.946 0.928 0.931 0.888 0.813 1.422 1.493 1.518 1.437 1.572 1.562 1.416 1.615 0.917 0.846
Ineligibility Rate 0.300 0.659 0.280 0.318 0.370 0.376 0.346 0.347 0.342 0.345 0.384 0.361 0.410 0.179 0.200 0.298 0.376 0.365 0.340 0.407 0.348 0.387 0.409
Completion Rate 0.949 0.885 0.862 0.842 0.840 0.795 0.773 0.828 0.840 0.846 0.823 0.874 0.860 0.683 0.565 0.443 0.598 0.578 0.539 0.591 0.549 0.670 0.682
Final Refusal Rate 0.042 0.104 0.063 0.061 0.065 0.094 0.121 0.104 0.076 0.075 0.111 0.072 0.097 0.081 0.101 0.000 0.000 0.008 0.001 0.000 0.036 0.000 0.001

Return to Table of Contents

Figure 3-4: Pharmacy providers - Response factors over time

Image displaying response factors over time, from 1996 through 2018 for pharmacy providers

Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Sample Rel to 2002 0.574 0.791 0.558 0.546 0.556 0.878 1.000 0.874 0.827 0.817 0.808 0.837 0.758 0.858 0.768 0.801 0.914 0.913 0.872 0.885 0.943 1.006 1.212
Ineligibility Rate 0.129 0.145 0.099 0.113 0.106 0.107 0.091 0.088 0.110 0.099 0.116 0.100 0.099 0.110 0.106 0.103 0.233 0.085 0.083 0.088 0.097 0.115 0.120
Completion Rate 0.722 0.700 0.838 0.822 0.820 0.761 0.790 0.729 0.794 0.787 0.799 0.797 0.756 0.689 0.610 0.749 0.805 0.846 0.852 0.881 0.906 0.872 0.896
Final Refusal Rate 0.061 0.068 0.084 0.079 0.078 0.113 0.122 0.200 0.159 0.167 0.149 0.165 0.271 0.050 0.015 0.015 0.016 0.013 0.011 0.003 0.001 0.000 0.013

Return to Table of Contents

3.5.4 Timing

Table 3-7 presents the hours per completed pair by provider type for the 2015-2018 MPC cycles. These timings include telephone and hard copy record abstraction as well as recruiting efforts.

Table 3-7. Hours per Completed Pair/Node, 2015 - 2018 MPC

  Provider Type
Year Hospital Office-Based Doctor Home Health Institution Pharmacy Separately Billing Doctor (nodes)
2015 7.9 3.9 5.0 3.1 0.7 3.1
2016 8.5 3.4 4.1 3.9 0.8 2.9
2017 7.9 2.9 4.3 1.4 0.8 2.6
2018 7.4 3.3 3.3 2.9 0.8 2.2

Return to Table of Contents

Appendix A: Acronyms and Definitions

AF: Authorization Form
AHRQ: Agency for Healthcare Research and Quality
BETOS: Berenson-Eggers Type of Service Codes
CMS: Case Management System
Contact Guide: Forms used to collect and manage information about contacts at provider facilities
CS: Control System
CPT: Current Procedural Terminiology Codes
DCS: Data Collection Specialist
ESN: Enhanced Security Network, developed by RTI to meet requirements of NIST Moderate Security
Event Forms: Forms used to record information about medical events identified in the HC
GPI: General Product Identifier
HC: Household Component of the MEPS
HIPAA: Health Insurance Portability and Accountability Act
ICD: International Classification of Diseases
IDCS: Integrated Data Collection System
MEPS: Medical Expenditure Panel Survey
MEPS-HC (HC): Household Component of the MEPS
MEPS-MPC (MPC): Medical Provider Component of the MEPS
NPI: National Provider Identifier
OBD: Office-Based Doctor
PHI: Protected Health Information
PII: Personally Identifiable Information
POC: Point of Contact in the provider facility
RU: Reporting Unit
SOP: Source of Payment
SBD: Separately-Billing Doctor

Return to Table of Contents

Appendix B: MPC Data Collection Summary Tables

Table B-1. MPC Sample Sizes, Provider Level, 1996-2018

  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Hospital
   Initial Sample 3,301 6,045 4,844 3,520 3,760 6,801 8,811 7,806 7,567 7,461
Sample after subsampling n/a 4,065 3,468 n/a 3,760 5,616 6,780 6,023 6,094 6,059
Final in-scope sample 3,330 4,163 3,247 3,284 3,467 5,201 6,325 5,580 5,671 5,600
HMO
Initial Sample 296 396 228 247 118 476 559 607 420 422
Sample after subsampling n/a 350 171 n/a 118 334 290 280 300 301
Final in-scope sample 628 467 155 225 113 287 256 218 250 241
Institution
Initial Sample 59 81 63 52 63 83 114 81 92 121
Sample after subsampling n/a 80 69 n/a 63 82 110 81 92 116
Final in-scope sample 50 75 65 45 60 76 103 73 89 108
Home Health
Initial Sample 415 674 456 393 319 520 631 588 568 606
Sample after subsampling n/a 653 420 n/a 319 509 611 586 556 593
Final in-scope sample 375 579 384 293 281 436 537 527 509 539
Office-based physician
Initial Sample 10,118 14,646 10,483 9,202 12,962 26,344 32,889 28,946 27,617 26,972
Sample after subsampling n/a 9,663 8,403   12,962 20,651 15,222 15,361 20,212 18,933
Final in-scope sample 7,758 7,047 7,356 8,076 11,167 18,078 13,652 13,808 18,069 16,898
SBD
Initial Sample 10,323 14,730 10,711 10,680 11,144 20,644 21,385 18,613 20,094 19,810
Sample after subsampling n/a 7,365 10,711 n/a 11,144 20,644 21,385 18,613 20,094 19,810
Final in-scope sample 8,705 5,297 7,704 7,288 7,026 12,891 13,976 12,154 13,225 12,971
Pharmacy
Initial Sample 6,109 8,547 5,734 5,703 5,762 9,118 10,200 8,882 8,608 8,404
Sample after subsampling n/a 8,547 5,734 n/a 5,762 9,118 10,200 8,882 8,608 8,404
Final in-scope sample 5,321 7,335 5,168 5,058 5,152 8,141 9,268 8,101 7,663 7,568

Return to Table of Contents

Table B-1. MPC Sample Sizes, Provider Level, 1996-2018 (continued)

  2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Hospital
Initial Sample 7,447 7,110 6,470 n/a n/a n/a n/a n/a n/a n/a 6,609 n/a n/a
Sample after subsampling 5,884 5,708 5,126 7,391 5,564 6,034 6,207 6,119 6,442 6,719 6,170 7,026 7,970
Final in-scope sample 5,484 5,328 4,776 6,436 5,072 5,435 5,896 5,788 6,031 6,323 n/a 6,551 7,321
HMO
Initial Sample 333 501 517 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
Sample after subsampling 284 316 243 249 378 327 412 336 410 358 375 369 331
Final in-scope sample 238 247 198 249 309 275 380 300 366 343 323 323 299
Institution
Initial Sample 80 76 81 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
Sample after subsampling 80 75 77 105 106 93 157 136 143 140 131 168 184
Final in-scope sample 78 72 72 101 92 88 151 128 132 129 128 161 166
Home Health
Initial Sample 655 534 505 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
Sample after subsampling 648 516 498 664 511 568 655 760 794 890 908 858 952
Final in-scope sample 602 464 446 603 454 487 573 646 677 728 763 713 838
Office-based physician
Initial Sample 27,620 25,052 25,537 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
Sample after subsampling 13,473 15,273 10,762 10,234 11,841 11,522 15,797 14,608 17,906 13,056 14,055 16,839 15,449
Final in-scope sample 12,062 13,492 9,533 9,148 10,441 10,169 14,065 13,236 15,904 11,957 12,903 15,105 13,677
SBD
Initial Sample 21,126 19,435 19,262 24,208 26,093 30,235 42,756 34,590 33,092 33,351 n/a n/a n/a
Sample after subsampling 21,126 19,435 19,262 24,208 26,093 30,235 29,168 34,590 33,092 33,351 34,627 20,936 20,002
Final in-scope sample 13,013 12,410 11,364 19,874 20,868 21,222 20,080 21,968 21,829 19,786 22,573 12,825 11,827
Pharmacy
Initial Sample 8,471 8,619 7,799 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
Sample after subsampling 8,471 8,619 7,799 8,935 7,960 8,270 9,250 9,246 8,812 9,001 8,457 10,531 12,763
Final in-scope sample 7,489 7,760 7,026 7,949 7,118 7,420 8,472 8,463 8,085 8,206 7,637 9,324 11,234

Return to Table of Contents

Table B-2. MPC Sample Sizes, Pair Level, 1996-2018

  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Hospital
Initial Sample 6,729 11,694 7,922 6,712 7,849 11,798 16,481 13,876 13,175 12,933
Sample after subsampling n/a 8,192 6,434 n/a 7,849 11,377 14,477 13,094 12,772 12,601
Final in-scope sample 6,570 7,938 5,825 6,163 7,016 10,155 12,805 11,532 11,589 11,279
HMO
Initial Sample 534 809 436 555 382 965 1,134 939 791 804
Sample after subsampling n/a n/a n/a n/a 382 791 567 625 665 685
Final in-scope sample 924 911 346 472 324 637 477 466 514 514
Institution
Initial Sample 63 85 64 53 66 86 116 86 94 123
Sample after subsampling n/a 85 70 n/a 66 86 115 85 94 123
Final in-scope sample 53 80 70 45 63 79 107 77 90 113
Homecare
Initial Sample 461 750 520 394 367 607 713 652 610 689
Sample after subsampling n/a 750 491 n/a 367 601 682 641 610 689
Final in-scope sample 385 662 445 340 317 471 606 579 555 619
Office-based physician
Initial Sample 13,681 19,157 12,641 11,974 17,407 33,518 42,327 36,804 34,611 33,854
Sample after subsampling n/a 12,635 10,747 n/a 17,407 26,886 19,309 19,731 26,392 24,517
Final in-scope sample 10,251 9,632 9,334 10,409 14,935 23,376 17,198 17,692 23,446 21,821
SBD
Initial Sample 12,488 17,394 13,658 14,906 15,955 28,905 30,780 26,965 29,271 28,930
Sample after subsampling n/a 8,697 13,658 n/a 15,955 28,930 30,780 26,965 29,271 28,930
Final in-scope sample 9,187 6,301 9,691 10,100 9,893 17,529 19,977 17,566 18,694 18,720
Pharmacy
Initial Sample 14,531 20,248 12,321 13,183 14,847 22,165 26,046 22,438 21,720 21,077
Sample after subsampling n/a n/a n/a n/a 14,847 22,165 26,046 22,438 21,720 21,077
Final in-scope sample 12,146 16,241 10,386 11,317 12,728 19,256 23,057 19,649 18,571 18,159

Return to Table of Contents

Table B-2. MPC Sample Sizes, Pair Level, 1996-2018 (continued)

  2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Hospital
Initial Sample 13,071 11,220 11,374               n/a n/a n/a
Sample after subsampling 11,911 10,646 10,672 14,199 9,960 10,404 11,361 11,017 10,909 11,225 11,088 11,059 12,979
Final in-scope sample 10,830 9,611 9,600 12,262 8,664 8,978 10,534 10,314 10,048 10,412 10,162 10,171 11,689
HMO
Initial Sample 694 852 968               n/a n/a n/a
Sample after subsampling 594 621 572 601 624 595 764 610 794 833 905 704 576
Final in-scope sample 476 459 449 601 478 458 702 541 667 752 790 577 490
Institution
Initial Sample 80 78 81               n/a n/a n/a
Sample after subsampling 80 78 80 113 108 95 159 140 148 147 134 173 191
Final in-scope sample 78 75 75 109 92 90 152 132 136 134 131 166 169
Home Health
Initial Sample 719 574 566               n/a n/a n/a
Sample after subsampling 719 572 564 728 512 609 712 820 842 957 984 920 1,032
Final in-scope sample 661 513 502 656 454 505 615 694 710 773 817 768 906
Office-based physician
Initial Sample 37,576 30,812 32,546               n/a n/a n/a
Sample after subsampling 17,139 19,201 16,713 13,386 14,256 14,583 19,945 16,921 21,280 16,727 18,445 19,382 18,256
Final in-scope sample 15,274 16,713 12,281 11,954 12,378 12,663 17,639 15,279 18,879 15,338 16,927 17,370 16,166
SBD
Initial Sample 31,058 26,407 27,496 27,480 30,584 38,873 49,782 43,568 41,670   n/a n/a n/a
Sample after subsampling 31,058 26,407 27,496 27,480 30,584 38,873 35,182 43,568 41,670 41,981 42,951 23,603 22,775
Final in-scope sample 18,699 16,660 16,144 22,417 23,958 26,802 23,406 27,346 27,064 24,610 27,490 14,437 13,313
Pharmacy
Initial Sample 20,990 19,052 19,678 22,587 18,761 19,807 22,731       n/a n/a n/a
Sample after subsampling 20,990 19,052 19,678 22,587 18,761 19,807 22,731 22,192 20,405 20,826 20,218 19,262 20,872
Final in-scope sample 17,418 16,313 17,038 19,683 16,261 17,414 20,510 20,028 18,424 18,415 17,366 16,735 17,744

Return to Table of Contents

Table B-3. MPC Data Collection Results, Provider Level, 1996-2018

  Initial Sample Sub-sample Eligible Sample Completion Rate Refusal Rate Other Nonresponse Rate
1996 Providers
Hospitals 3,301 3,301 3,224 0.951 0.021 0.028
Office-based providers 10,118 10,118 7,530 0.881 0.069 0.051
HMOs 296 296 601 0.805 0.085 0.110
Home care providers 415 415 353 0.875 0.062 0.062
Institutions 59 59 50 0.960 0.040 -
SBDs 10,323 10,323 7,223 0.949 0.042 0.009
Pharmacies 6,109 6,109 5,321 0.722 0.061 0.217
Total 30,621 30,621 24,302      
1997 Providers
Hospitals 4,768 4,065 4,163 0.894 0.058 0.048
Office-based providers 10,095 9,666 7,047 0.871 0.053 0.069
HMOs 350 350 467 0.717 0.090 0.193
Home care providers 653 653 579 0.834 0.090 0.076
Institutions 80 80 75 0.827 0.107 0.067
SBDs 14,730 14,730 5,026 0.885 0.104 0.012
Pharmacies 8,574 8,574 7,335 0.700 0.068 0.232
Total 39,250 38,118 24,692      
1998 Providers
Hospitals 3,468 3,468 3,247 0.939 0.025 0.037
Office-based providers 10,483 8,403 7,356 0.861 0.043 0.096
HMOs 228 171 155 0.871 0.103 0.026
Home care providers 456 420 384 0.820 0.089 0.091
Institutions 63 69 65 0.754 0.169 0.077
SBDs 10,711 10,711 7,707 0.862 0.063 0.075
Pharmacies 5,734 5,734 5,167 0.838 0.084 0.079
Total 31,143 28,976 24,081      
1999 Providers
Hospitals 3,520 3,520 3,282 0.926 0.036 0.037
Office-based providers 9,202 9,202 8,075 0.888 0.053 0.058
HMOs 247 247 225 0.876 0.080 0.044
Home care providers 338 338 293 0.840 0.082 0.078
Institutions 52 52 44 0.773 0.182 0.045
SBDs 10,680 10,680 7,289 0.842 0.061 0.097
Pharmacies 5,703 5,703 5,058 0.822 0.079 0.099
Total 29,742 29,742 24,266      
2000 Providers
Hospitals 3,760 3,760 3,467 0.910 0.037 0.054
Office-based providers 12,962 12,962 11,167 0.864 0.071 0.065
HMOs 118 118 113 0.929 0.035 0.035
Home care providers 319 319 281 0.858 0.068 0.075
Institutions 63 63 60 0.850 0.067 0.083
SBDs 11,144 11,144 7,026 0.840 0.065 0.094
Pharmacies 5,762 5,762 5,152 0.820 0.078 0.102
Total 34,128 34,128 27,266      
2001 Providers
Hospitals 6,801 5,616 5,201 0.912 0.038 0.050
Office-based providers 26,344 20,651 18,078 0.850 0.069 0.081
HMOs 476 334 287 0.899 0.021 0.066
Home care providers 520 509 436 0.851 0.060 0.046
Institutions 83 82 76 0.934 0.079 -
SBDs 20,644 20,644 12,891 0.795 0.094 0.111
Pharmacies 9,118 9,118 8,141 0.761 0.113 0.126
Total 63,986 56,954 45,110      
2002 Providers
Hospitals 8,811 6,780 6,325 0.900 0.048 0.045
Office-based providers 32,889 15,222 13,652 0.837 0.097 0.066
HMOs 559 290 256 0.899 0.055 0.047
Home care providers 631 611 537 0.823 0.093 0.084
Institutions 114 110 103 0.913 0.058 0.029
SBDs 21,385 21,385 13,976 0.773 0.121 0.106
Pharmacies 10,200 10,200 9,268 0.790 0.122 0.088
Total 74,589 54,598 44,117      
2003 Providers
Hospitals 7,806 6,023 5,580 0.898 0.047 0.055
Office-based providers 28,946 15,361 13,808 0.835 0.095 0.070
HMOs 506 280 218 0.876 0.032 0.092
Home care providers 607 586 527 0.850 0.068 0.082
Institutions 83 81 73 0.945 0.027 0.027
SBDs 18,613 18,613 12,154 0.828 0.104 0.068
Pharmacies 8,882 8,882 8,101 0.729 0.200 0.106
Total 65,443 49,826 40,461      
2004 Providers
Hospitals 7,567 6,094 5,671 0.920 0.027 0.053
Office-based providers 27,617 20,202 18,069 0.864 0.076 0.060
HMOs 420 300 250 0.892 0.056 0.052
Home care providers 568 556 509 0.809 0.108 0.083
Institutions 93 92 89 0.910 0.056 0.034
SBDs 20,094 20,094 13,225 0.840 0.076 0.084
Pharmacies 8,608 8,608 7,663 0.794 0.159 0.047
Total 64,967 55,946 45,476      
2005 Providers
Hospitals 7,461 6,059 5,600 0.931 0.026 0.043
Office-based providers 26,972 18,933 16,898 0.859 0.086 0.055
HMOs 422 301 241 0.963 0.012 0.025
Home care providers 606 593 539 0.810 0.111 0.080
Institutions 121 116 108 0.963 0.009 0.028
SBDs 19,810 19,810 12,971 0.846 0.075 0.077
Pharmacies 8,404 8,404 7,568 0.787 0.167 0.046
Total 63,796 54,216 43,925      
2006 Providers
Hospitals 7,447 5,884 5,484 0.941 0.022 0.037
Office-based providers 27,620 13,473 12,062 0.869 0.074 0.057
HMOs 333 284 238 0.920 0.042 0.038
Home care providers 655 648 602 0.856 0.080 0.065
Institutions 80 80 78 0.808 0.115 0.077
SBDs 21,126 21,126 13,013 0.823 0.111 0.066
Pharmacies 8,471 8,471 7,489 0.799 0.149 0.052
Total 65,732 49,966 38,966      
2007 Providers
Hospitals 7,110 5,708 5,328 0.944 0.023 0.033
Office-based providers 25,052 15,273 13,492 0.875 0.077 0.048
HMOs 501 316 247 0.923 0.036 0.041
Home care providers 534 516 464 0.883 0.060 0.057
Institutions 76 76 72 0.930 0.042 0.028
SBDs 19,435 19,435 12,410 0.874 0.072 0.054
Pharmacies 8,619 8,619 7,760 0.797 0.165 0.038
Total 61,327 49,943 39,773      
2008 Providers
Hospitals 6,470 5,126 4,776 0.946 0.022 0.035
Office-based providers 25,537 10,762 9,533 0.891 0.067 0.054
HMOs 517 243 198 0.970 - 0.031
Home care providers 505 498 446 0.901 0.077 0.032
Institutions 81 77 72 0.944 0.044 0.015
SBDs 19,262 19,262 11,364 0.860 0.097 0.066
Pharmacies 7,799 7,799 7,026 0.756 0.271 0.050
Total 60,171 43,767 33,415      
2009 Providers
Hospitals n/a 7,391 6,436 0.890 0.012 0.098
Office-based providers n/a 10,234 9,148 0.801 0.003 0.227
HMOs n/a 249 249 - - -
Home care providers n/a 664 603 0.861 0.053 0.086
Institutions n/a 105 101 0.921 0.030 0.050
SBDs n/a 24,208 19,874 0.683 0.081 0.236
Pharmacies n/a 8,935 7,949 0.689 0.050 0.262
Total n/a 51,786 44,366      
2010 Providers
Hospitals n/a 5,564 5,072 0.846 0.034 0.119
Office-based providers n/a 11,841 10,441 0.806 0.062 0.132
HMOs n/a 378 309 0.832 - 0.168
Home care providers n/a 511 454 0.775 0.097 0.128
Institutions n/a 106 92 0.880 0.054 0.065
SBDs n/a 26,093 20,868 0.565 0.101 0.335
Pharmacies n/a 7,960 7,118 0.610 0.015 0.283
Total n/a 52,453 44,354      
2011 Providers
Hospitals n/a 6,034 5,435 0.919 0.016 0.065
Office-based providers n/a 11,522 10,169 0.890 0.023 0.086
HMOs n/a 327 275 0.869 - 0.131
Home care providers n/a 568 487 0.893 0.035 0.072
Institutions n/a 93 88 0.920 0.023 0.057
SBDs n/a 30,235 21,222 0.447 0.000 0.553
Pharmacies n/a 8,270 7,420 0.749 0.015 0.237
Total n/a 57,049 45,096      
2012 Providers
Hospitals n/a 6,207 5,896 0.870 0.015 0.115
Office-based providers n/a 15,797 14,065 0.876 0.028 0.096
HMOs n/a 412 380 0.776 0.042 0.182
Home care providers n/a 655 573 0.843 0.019 0.080
Institutions n/a 157 151 0.894 0.053 0.053
SBDs 42,756 29,168 20,080 0.598 0.000 0.402
Pharmacies n/a 9,250 8,472 0.805 0.016 0.230
Total n/a 64,676 49,617      
2013 Providers
Hospitals n/a 6,119 5,788 0.877 0.036 0.087
Office-based providers n/a 14,608 13,236 0.890 0.036 0.073
HMOs n/a 336 300 0.687 - 0.313
Home care providers n/a 760 646 0.862 0.025 0.113
Institutions n/a 136 128 0.914 0.023 7.586
SBDs n/a 34,590 21,968 0.578 0.008 0.414
Pharmacies n/a 9,246 8,463 0.846 0.013 0.138
Total   65,795 50,529      
2014 Providers
Hospitals n/a 6,442 6,031 0.848 0.001 0.151
Office-based providers n/a 17,906 15,904 0.865 0.001 0.134
HMOs n/a 410 366 0.719 - 0.281
Home care providers n/a 794 677 0.861 - 0.139
Institutions n/a 143 132 0.924 - 0.076
SBDs n/a 33,092 21,829 0.539 0.001 0.460
Pharmacies n/a 8,812 8,085 0.852 0.011 0.137
Total   67,599 53,024      
2015 Providers
Hospitals n/a 6,719 6,323 0.811 0.053 0.136
Office-based providers n/a 13,056 11,957 0.849 0.039 0.113
HMOs n/a 358 343 0.813 - 0.187
Home care providers n/a 890 728 0.794 0.008 0.198
Institutions n/a 140 129 0.884 - 0.116
SBDs n/a 33,351 19,786 0.591 0.000 0.408
Pharmacies n/a 9,001 8,206 0.881 0.003 0.116
Total   63,515 47,472      
2016 Providers
Hospitals n/a 6,609 6,170 0.861 0.024 0.116
Office-based providers n/a 14,055 12,903 0.869 0.020 0.111
HMOs n/a 375 323 0.833 0.000 0.167
Home care providers n/a 908 763 0.847 0.007 0.147
Institutions n/a 131 128 0.906 0.000 0.094
SBDs n/a 34,627 22,573 0.549 0.036 0.415
Pharmacies n/a 8,457 7,637 0.906 0.001 0.093
Total   65,162 50,497      
2017 Providers
Hospitals n/a 7,026 6,551 0.879 0.006 0.115
Office-based providers n/a 16,839 15,105 0.824 0.007 0.168
HMOs n/a 369 323 0.910 0.000 0.090
Home care providers n/a 858 713 0.851 0.000 0.149
Institutions n/a 168 161 0.913 0.000 0.087
SBDs n/a 20,936 12,825 0.670 0.000 0.330
Pharmacies n/a 10,531 9,324 0.872 0.000 0.128
Total   56,727 45,002      
2018 Providers
Hospitals n/a 7,970 7,321 0.881 0.005 0.114
Office-based providers n/a 15,449 13,677 0.820 0.003 0.177
HMOs n/a 331 299 0.890 0.000 0.110
Home care providers n/a 952 838 0.850 0.001 0.149
Institutions n/a 184 166 0.910 0.000 0.090
SBDs n/a 20,002 11,827 0.682 0.001 0.317
Pharmacies n/a 12,763 11,234 0.896 0.013 0.091
Total n/a 57,651 45,362      

Return to Table of Contents

Table B-4. MPC Data Collection Results, Pair Level, 1996-2018

  Initial Sample Sub-sample Eligible Sample Completion Rate Refusal Rate Other Nonresponse Rate
1996 Pairs
Hospitals 6,729 6,729 6,570 0.932 0.038 0.030
Office-based providers 13,681 13,681 10,251 0.865 0.079 0.056
HMOs 534 534 924 0.803 0.105 0.092
Home care providers 461 461 385 0.875 0.057 0.068
Institutions 63 63 53 0.943 0.057 0.000
SBDs 12,488 12,488 8,689 0.937 0.056 0.007
Pharmacies 14,531 14,531 12,146 0.671    
Total 48,487 48,487 39,018      
1997 Pairs
Hospitals 11,694 8,192 7,938 0.874 0.070 0.056
Office-based providers 19,157 12,635 10,062 0.862 0.062 0.076
HMOs 809 809 911 0.626 0.156 0.218
Home care providers 750 750 662 0.823 0.095 0.082
Institutions 85 85 80 0.825 0.113 0.063
SBDs 17,397 8,697 5,964 0.865 0.123 0.013
Pharmacies 20,248 20,248 16,241 0.672 0.075 0.253
Total 70,140 51,416 41,858      
1998 Pairs
Hospitals 7,922 6,434 5,824 0.925 0.031 0.044
Office-based providers 12,641 10,747 9,334 0.852 0.050 0.098
HMOs 436 436 346 0.832 0.133 0.035
Home care providers 520 491 445 0.825 0.085 0.090
Institutions 64 70 65 0.754 0.169 0.077
SBDs 13,658 13,658 9,687 0.836 0.084 0.080
Pharmacies 12,321 12,321 10,388 0.793 0.116 0.091
Total 47,562 44,157 36,089      
1999 Pairs
Hospitals 6,712 6,712 6,160 0.909 0.053 0.039
Office-based providers 11,974 11,974 10,409 0.879 0.061 0.060
HMOs 555 555 472 0.886 0.068 0.047
Home care providers 394 394 340 0.818 0.088 0.094
Institutions 53 53 45 0.756 0.200 0.044
SBDs 14,907 14,907 10,101 0.808 0.091 0.100
Pharmacies 13,183 13,183 11,317 0.788 0.099 0.113
Total 47,778 47,778 38,844      
2000 Pairs
Hospitals 7,849 7,849 7,016 0.891 0.056 0.053
Office-based providers 17,407 17,407 14,935 0.854 0.079 0.067
HMOs 382 382 324 0.873 0.059 0.068
Home care providers 367 367 317 0.864 0.063 0.073
Institutions 66 66 63 0.825 0.095 0.079
SBDs 15,955 15,955 9,893 0.823 0.094 0.084
Pharmacies 14,847 14,847 12,728 0.768 0.105 0.127
Total 56,873 56,873 45,276      
2001 Pairs
Hospitals 11,798 11,377 10,155 0.899 0.023 0.051
Office-based providers 33,518 26,886 23,376 0.843 0.077 0.081
HMOs 965 791 637 0.878 0.028 0.094
Home care providers 607 601 471 0.847 0.064 0.089
Institutions 86 86 79 0.937 0.051 0.013
SBDs 28,905 28,905 17,529 0.778 0.127 0.095
Pharmacies 22,165 22,165 19,256 0.703 0.144 0.153
Total 98,044 90,811 71,503      
2002 Pairs
Hospitals 16,481 14,477 12,805 0.895 0.061 0.045
Office-based providers 42,327 19,309 17,198 0.832 0.104 0.065
HMOs 1,134 567 477 0.870 0.052 0.078
Home care providers 713 682 606 0.820 0.100 0.081
Institutions 116 115 107 0.907 0.056 0.037
SBDs 30,780 30,780 19,977 0.745 0.160 0.095
Pharmacies 26,046 26,046 23,057 0.734 0.156 0.110
Total 117,597 91,976 74,227      
2003 Pairs
Hospitals 13,876 13,094 11,532 0.895 0.052 0.054
Office-based providers 36,804 19,731 17,692 0.828 0.103 0.070
HMOs 939 625 466 0.852 0.054 0.094
Home care providers 652 641 579 0.853 0.067 0.079
Institutions 86 85 77 0.948 0.026 0.026
SBDs 26,965 26,965 17,566 0.804 0.152 0.045
Pharmacies 22,438 22,438 19,649 0.671 0.251 0.078
Total 101,760 83,579 67,561      
2004 Pairs
Hospitals 13,175 12,772 11,589 0.922 0.028 0.050
Office-based providers 34,611 26,392 23,446 0.858 0.084 0.058
HMOs 791 665 514 0.813 0.088 0.099
Home care providers 610 610 555 0.805 0.115 0.080
Institutions 94 94 90 0.911 0.056 0.033
SBDs 29,271 29,271 18,694 0.827 0.103 0.070
Pharmacies 21,720 21,720 18,571 0.715 0.214 0.071
Total 100,272 91,524 73,459      
2005 Pairs
Hospitals 12,933 12,601 11,279 0.923 0.036 0.041
Office-based providers 33,854 24,517 21,821 0.852 0.094 0.054
HMOs 804 685 514 0.955 0.014 0.031
Home care providers 689 689 619 0.816 0.113 0.071
Institutions 123 123 113 0.965 0.009 0.027
SBDs 28,930 28,930 18,720 0.824 0.114 0.063
Pharmacies 21,077 21,077 18,159 0.711 0.214 0.075
Total 98,410 88,622 71,225      
2006 Pairs
Hospitals 13,071 11,911 10,830 0.934 0.031 0.035
Office-based providers 37,576 17,139 15,274 0.861 0.082 0.056
HMOs 694 594 476 0.903 0.059 0.038
Home care providers 719 719 661 0.847 0.082 0.071
Institutions 80 80 78 0.808 0.115 0.077
SBDs 31,058 31,058 18,699 0.807 0.144 0.049
Pharmacies 20,990 20,990 17,418 0.734 0.196 0.070
Total 104,188 82,491 63,436      
2007 Pairs
Hospitals 11,220 10,646 9,611 0.929 0.032 0.039
Office-based providers 30,812 19,021 16,713 0.870 0.083 0.047
HMOs 852 621 459 0.919 0.046 0.035
Home care providers 574 572 513 0.887 0.057 0.056
Institutions 78 78 75 0.933 0.040 0.027
SBDs 26,407 26,407 16,660 0.864 0.046 0.090
Pharmacies 19,052 19,052 16,313 0.737 0.217 0.046
Total 88,995 76,397 60,344      
2008 Pairs
Hospitals 11,374 10,672 9,600 0.943 0.026 0.034
Office-based providers 32,546 13,917 12,281 0.884 0.077 0.054
HMOs 968 572 449 0.958 0.002 0.042
Home care providers 566 564 502 0.902 0.077 0.031
Institutions 81 80 75 0.947 0.042 0.014
SBDs 27,496 27,496 16,144 0.846 0.133 0.049
Pharmacies 19,678 19,678 17,038 0.706 0.356 0.060
Total 92,709 72,979 56,089      
2009 Pairs
Hospitals n/a 14,199 12,262 0.877 0.014 0.109
Office-based providers n/a 13,386 11,954 0.798 0.055 0.136
HMOs n/a 601 601 - - -
Home care providers n/a 728 656 0.854 0.055 0.087
Institutions n/a 113 109 0.927 0.028 0.046
SBDs n/a 27,480 22,417 0.683 0.084 0.233
Pharmacies n/a 22,587 19,683 0.632 0.260 0.108
Total n/a 79,094 67,682      
2010 Pairs
Hospitals n/a 9,960 8,664 0.825 0.055 0.120
Office-based providers n/a 14,256 12,378 0.801 0.073 0.126
HMOs n/a 624 478 0.791 - 0.209
Home care providers n/a 512 454 0.773 0.106 0.121
Institutions n/a 108 92 0.880 0.054 0.065
SBDs n/a 30,584 23,958 0.552 0.112 0.336
Pharmacies n/a 18,761 16,261 0.661 0.020 0.319
Total n/a 74,805 62,285      
2011 Pairs
Hospitals n/a 10,404 8,978 0.909 0.043 0.047
Office-based providers n/a 14,583 12,663 0.887 0.057 0.056
HMOs n/a 595 458 0.856 - 0.144
Home care providers n/a 609 505 0.889 0.036 0.075
Institutions n/a 95 90 0.900 0.056 0.044
SBDs n/a 38,873 26,802 0.441 0.033 0.525
Pharmacies n/a 19,807 17,414 0.730 0.022 0.248
Total n/a 84,966 66,910      
2012 Pairs
Hospitals n/a 11,361 10,534 0.846 0.032 0.122
Office-based providers n/a 19,945 17,639 0.868 0.056 0.076
HMOs n/a 764 702 0.715 0.056 0.229
Home care providers n/a 712 615 0.849 0.080 0.072
Institutions n/a 159 152 0.895 0.053 0.053
SBDs 49,782 35,182 23,406 0.576 0.019 0.405
Pharmacies n/a 22,731 20,510 0.743 0.030 0.226
Total n/a 90,854 73,558      
2013 Pairs
Hospitals n/a 11,017 10,314 0.865 0.074 0.061
Office-based providers n/a 16,921 15,279 0.886 0.060 0.054
HMOs n/a 610 541 0.643 0.331 0.023
Home care providers n/a 820 694 0.846 0.097 0.058
Institutions n/a 140 132 0.902 0.045 0.053
SBDs n/a 43,568 27,346 0.555 0.035 0.410
Pharmacies n/a 22,192 20,028 0.763 0.072 0.165
Total   95,268 74,334      
2014 Pairs
Hospitals n/a 10,909 10,048 0.835 0.045 0.120
Office-based providers n/a 21,280 18,879 0.863 0.051 0.000
HMOs n/a 794 667 0.705 - 0.295
Home care providers n/a 842 710 0.856 0.075 0.069
Institutions n/a 148 136 0.919 0.037 0.044
SBDs n/a 41,670 27,064 0.509 0.034 0.457
Pharmacies n/a 20,405 18,424 0.792 0.029 0.179
Total   96,048 75,928      
2015 Pairs
Hospitals n/a 11,225 10,412 0.805 0.093 0.102
Office-based providers n/a 16,727 15,338 0.845 0.082 0.073
HMOs n/a 833 752 0.742 - 0.258
Home care providers n/a 957 773 0.796 0.106 0.098
Institutions n/a 147 134 0.888 0.052 0.060
SBDs n/a 41,981 24,610 0.567 0.048 0.385
Pharmacies n/a 20,826 18,415 0.832 0.023 0.145
Total   92,696 70,434      
2016 Pairs
Hospitals n/a 11,088 10,162 0.851 0.081 0.068
Office-based providers n/a 18,445 16,927 0.861 0.070 0.069
HMOs n/a 905 790 0.766 - 0.234
Home care providers n/a 984 817 0.841 0.111 0.048
Institutions n/a 134 131 0.908 0.046 0.046
SBDs n/a 42,951 27,490 0.539 0.050 0.412
Pharmacies n/a 20,218 17,366 0.850 0.067 0.083
Total n/a 94,725 73,683      
2017 Pairs
Hospitals n/a 11,059 10,171 0.870 0.048 0.082
Office-based providers n/a 19,382 17,370 0.820 0.036 0.144
HMOs n/a 704 577 0.896 - 0.104
Home care providers n/a 920 768 0.850 0.073 0.077
Institutions n/a 173 166 0.916 0.018 0.066
SBDs n/a 23,063 14,437 0.661 0.072 0.267
Pharmacies n/a 19,262 16,735 0.858 0.025 0.117
Total n/a 75,103 60,224      
2018 Pairs
Hospitals n/a 12,979 11,689 0.877 0.028 0.095
Office-based providers n/a 18,256 16,166 0.824 0.036 0.140
HMOs n/a 576 490 0.855 0.043 0.102
Home care providers n/a 1,032 906 0.849 0.044 0.107
Institutions n/a 191 169 0.905 0.018 0.077
SBDs n/a 22,775 13,313 0.680 0.050 0.270
Pharmacies n/a 20,872 17,744 0.878 0.050 0.072
Total n/a 76,681 60,477      

Return to Table of Contents

1Following convention, the 2018 MPC refers to the data collected about calendar year 2018 which are matched with data from the 2018 Household Component (HC) of MEPS. Data collection for 2018 MPC began in February 2019 and continued through January 2020 (see Section 3.4).

2 Note that these counts and percentages are based on participation at the contact group level, not individual providers. As noted in section 2, contact groups may consist of multiple providers as, for example, a group practice that employs a number of physicians or a health care system that may contain several hospitals. Note as well that contact group is a different metric than the concept of "provider wave" reported in the MPC prior to 2009. In a provider wave, a provider is counted one for each wave of the sample in which it is represented. Table 3.1 reports the percentage of contact groups that provided medical and patient account records.

Appendix C: Critical Items

Event level

Answers are required for the following in order to be a full complete event:

  • Event month and year for outpatient
  • Event days, months, year for inpatient or “somewhere else”
  • Global fee months and years
  • At least one CPT code
  • Surgical codes
  • Was it FFS or Capitated
  • If FFS- At least one payment ($0 counts as a payment, but should only be used when we are sure the SOP did not pay)
  • If Capitated- insurance type

An event can still be a full complete if we have “don’t know” in any of the following:

  • If outpatient event DK to the day part of the event date is ok
  • Location of service (however, if we can’t determine location of service, we typically default to outpatient for hospital events)
  • Diagnosis
  • SBD info
  • Global fee days (only month and year are required)
  • Charges for each CPT
  • FFS- Some payments can be “don’t know“ if we know at least one payment ($0 counts as a payment, but should only be used when we are sure the SOP did not pay)
  • Reasons payments less than or greater than charges
  • Expecting additional payments
  • If capitated:
    • Copayment
    • Who paid copayment
    • Other payments

Pair-level

  • If all events in the pair are full complete events, the pair is finalized as a completed pair
  • If at least one event in the pair is full complete, the pair is finalized as a partial complete pair
  • If all the events in a pair have some data but all are missing critical items, the pair is a special partial pair. 
  • If the pair contains no events that contain critical items
  • We also created a new “special partial”, which is an event that has any data at all.  These special partials show up as final others in our main production report, but show up as partials in an alternate production report.  We want to minimize the special partials during the field period, but this means that all pairs that have any records at all should at least be data entered a special partial (and not coded out as a refusal).

Critical Items

Table C-1. Critical Items

Item Item is complete if: Hospital OBD Home Health Agency
HCH-Health
HCN-Non-Health
Institution SBD
  1. Admit and discharge dates for inpatient stays
Valid dates
Don’t Know
Refusal
A2a     A1  
  1. Date of visit for outpatient visits
Valid date
Don’t Know
Refusal
A2c B1      
  1. Dates of service
Valid dates
Don’t Know
Refusal
    E1 (HCH)
D1 (HCN)
  B2b
  1. Diagnosis
Verbatim description or ICD-9 code
Don’t Know
Refusal
    E2    
  1. Home health care personnel type and hours:
    • Home health aide
    • Homemaker
    • IV/Infusion Therapist
    • Nurse/Nurse Practitioner
    • Nurse’s aide
    • Occupational therapist
    • Personal care attendant
    • Physical therapist
    • Respiratory therapist
    • Social worker
    • Speech therapist
    • Yard worker
    • Driver
    • Babysitter
    • Other
Number of hours for each type (includes 0)
Don’t Know
Refusal
    E3(HCH)
D2(HCN)
   
  1. (IF GLOBAL FEE) Dates of other services covered by fee
Valid dates
Don’t Know
Refusal
A5d B2b      
  1. Location of service
    • Physician office
    • Hospital, Inpatient
    • Hospital, Outpatient
    • Hospital, Emergency Room
    • Somewhere else
(For each location)
Yes
No
Don’t Know
Refusal
  B3      
  1. Services Provided
Description or CPT code
Don’t Know
Refusal
A6a B5a E4    
  1. DRG
Valid DRG
None
Don’t Know
Refusal
A8        
  1. Surgical procedures
Description or CPT code
Don’t Know
Refusal
A10a       B5a
  1. Fee-For-Service or Capitated
Fee or capitated C3 C3   Q5 C5
  1. Total charge
Dollar value
Don’t Know
Refusal
      Q6  
  1. Dollar payment by payer:
    • Patient or patient’s family
    • Medicare
    • Medicaid
    • Private insurance
    • VA/CHAMPVA
    • Tricare
    • Worker’s compensation
(For each source)
Dollar value (includes 0)
Don’t Know
Refusal
C4 C4 C4a Q7
Q11a
Q13
Q16
C4
  1. Other payment source and amount
Dollar value (includes 0)
Don’t Know
Refusal
C4
Other Loop
C4
Other Loop
C4
Other Loop
C7, Q11a, Q13, Q16
Other Loop
C4
Other Loop
  1. What kind of insurance plan covered the patient for (this visit/these visits/this stay)?
    • Medicare
    • Medicaid
    • Private insurance
    • VA/CHAMPVA
    • Tricare
    • Worker’s compensation
(For each source)
Yes
No
Don’t Know
Refusal
C7a C7a      
  1. Payment source for ancillary charges
    • Patient or patient’s family
    • Medicare
    • Medicaid
    • Private insurance
    • VA/CHAMPVA
    • Tricare
    • Worker’s compensation
Dollar value (includes 0)
Don’t Know
Refusal
      Q20  
  1. Other payment source for ancillary charges
Dollar value (includes 0)
Don’t Know
Refusal
      Q20 Other Loop  
  1. Who paid co-payment?
    • Patient or patient’s family
    • Medicare
    • Medicaid
    • Private insurance
Yes
No
Don’t Know
Refusal
      Q21f  

Non-Pharmacy Providers. For hospital, OBD, HMO, Home Health, Institution, and SBD providers, the definition of partially complete events was expanded. In the 2010 MPC data collection and earlier, for a pair to be considered partially complete at least one event had to have a valid response for all critical items (no “don’t know,” “refusal,” or missing entries). At the event level, if one critical item has a “don’t know,” “refusal,” or missing entry, the event is coded as “final critical item missing.” Because of a modification in the procedures for matching MPC events to HC events in the 2010 MPC, events coded as “final critical item missing” are included as events that could be matched. For this reason, beginning with the 2011 data collection and in subsequent cycles, criteria for partially complete events were revised to include events with at least one critical item answered.

Pharmacy Critical Items

Item Item is complete if: Item Number
  1. NDC or Drug Name
NDC: 11 DIGITS
Don’t Know
Refusal
Drug Name: Text
Don’t Know
Refusal
Q2a / Q2b
  1. If Drug Name:
    • Strength
Numeric value
Don’t Know
Refusal
Q2c / Q2c1
  1. If Drug Name:
    • Strength Unit
Range of Units & Other Specify
Don’t Know
Refusal
Q2d / Q2d2
  1. If Drug Name:
    • Dosage Form
Range of Forms & Other Specify Q2e
  1. Quantity
Numeric value up to 3 decimal points
Don’t Know
Refusal
Q3a
  1. Patient Payment
Dollar Value
$0 – $500
Don’t Know
Refusal
Q5
  1. Third party payer type
Range of Types & Other Specify
Don’t Know
Refusal
Q6
  1. Third party payment
Dollar value
$0 – $5000
Don’t Know
Refusal
Q7

Return to Table of Contents