Data Sources—Centers for Medicare & Medicaid Services (CMS)

Sponsor

U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services (CMS), Office of Clinical Standards and Quality.

Mode of Administration

Annually, each end stage renal disease (ESRD) network conducts a survey of ESRD facilities to validate the census of ESRD patients in the network at the end of the calendar year. Two data collection forms are used, one for hemodialysis patients and one for peritoneal dialysis patients. Clinical information contained in the medical records is abstracted for each patient in the adult hemodialysis sample and for all pediatric in-center hemodialysis patients who received in-center hemodialysis at any time during a selected target period. Clinical information contained in the medical records is also abstracted for each patient in the adult peritoneal dialysis sample who received peritoneal dialysis at any time during selected target periods.

Sample Design

The ESRD CPMs are collected on a national random sample of adult in-center hemodialysis patients, stratified by network area, and a random sample of 5% of adult peritoneal dialysis patients in the Nation.

Primary Content

Clinical records about the 16 CPM measures in the areas of hemodialysis and peritoneal dialysis adequacy, vascular access, and anemia management.

Population Targeted

ESRD hemodialysis patients and peritoneal dialysis patients.

Demographic Data

Age, gender, race, ethnicity.

Years Collected

Since 1999.

Schedule

Annual.

Geographic Estimates

State.

Contact Information

Agency home page: http://www.cms.hhs.gov.

Data system home page: http://www.cms.hhs.gov/CPMProject/.

Sponsor

U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services (CMS).

Mode of Administration

Completed by home health agency (HHA) personnel.

Primary Content

The OASIS is a group of data elements that represent core items of a comprehensive assessment for an adult home care patient. Because OASIS can measure changes in a patient's health status between two or more time points, it forms the basis for measuring patient outcomes for purposes of outcome-based quality improvement (OBQI).

The general categories of data and health status items in OASIS include demographics and patient history, living arrangements, supportive assistance, sensory status, integumentary (skin) status, respiratory status, elimination status, neuro/emotional/behavioral status, activities of daily living, medications, equipment management, and information collected at inpatient facility admission or agency discharge.

Demographic Data

Age, gender, race/ethnicity, State of residence, marital status, informal caregiver assistance, living arrangement, current payment sources for home care, and health status.

Population Targeted

U.S. adult, nonmaternity, home care patients receiving skilled services.

Years Collected

Medicare and Medicaid certified agencies have been required to collect data since 1999.

Data Collection Schedule

Most OASIS data items are designed to be collected at the start of care and every 2 months thereafter until and including time of discharge.

HHAs are required to electronically transmit OASIS data only for home care patients receiving Medicare and Medicaid skilled services to State survey and certification agencies. The State agencies have the overall responsibility for collecting OASIS data in accordance with CMS specifications. The State is also responsible for preparing OASIS data for retrieval by a national repository established by CMS.

Geographic Estimates

National, State, agency.

Contact Information

Agency home page: http://www.cms.hhs.gov.

Data system home page: http://www.cms.hhs.gov/oasis.

Notes

The Home Health Care tables are based on OASIS assessment data from January 1-December 31 each year. The data for the measures are based only on home health episodes that began and ended in the same calendar year. Approximately 22% of patients have more than one episode in the same year.

The data were downloaded from http://www.medicare.gov/download/downloaddb.asp on April 26, 2007, and reflect care delivered mainly in 2006. OASIS data are routinely updated with a delay of approximately 2 months to make sure CMS has all the data from the agencies to make the reports. The 2006 data were combined with the 2005 data from last year's NHQR tables.

Risk Adjustment for Measures

Each State average rate is risk adjusted using logistic regression models (details in http://www.cms.hhs.gov/HomeHealthQualityInits/Downloads/HHQIHomeHealthCompareRiskModels.pdf; Plugin Software Help). The national rate is the observed rate because the national rate already reflects national characteristics.

For OASIS improvement measures, a person is not included if he or she is at the highest level and cannot improve any more. Similarly, a person is excluded from stabilization measures if he or she is at the lowest level and cannot decline any more. Therefore, the denominator for each measure can vary widely. On average, for example, only 30% of all episodes are included in the urinary incontinence measure and about 40% in the confusion frequency measure, while 95% of episodes are included in bathing stabilization and about 83% of episodes in ambulation improvement.

Sponsor

U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services (CMS).

Mode of Administration

The data in the Hospital Compare database are obtained from two sources. Hospital identification and characteristics are derived from CMS's Online Survey, Certification, and Reporting (OSCAR) database; a summary of the latest survey results are used. Clinical quality measure data are extracted from the Medicare Quality Improvement Organization (QIO) Program Clinical Warehouse. Refer to the QIO entry in this document for further information.

Sample Design

Data in the Hospital Compare are provided voluntarily by short-term acute care hospitals and rural small, remote “critical access” hospitals. The facilities agree to report on an initial set of 10 quality performance measures and to have their data available to the public. The short-term acute care facilities receive an incentive payment for participating. Beginning with discharges in the second quarter of 2004, hospitals may elect to submit data on an additional seven measures.

Primary Content

The Hospital Compare contains data about 17 quality measures in 3 clinical areas: heart attack (acute myocardial infarction), heart failure, and pneumonia. Data are summarized at the hospital-level. State-level estimates are also calculated.

Population Targeted

Discharges from short-term acute care hospitals and rural small, remote “critical access” hospitals.

Demographic Data

None. Only hospital-level data are presented.

Years Collected

Since January 2004.

Schedule

Annual, updated quarterly.

Geographic Estimates

National, State. Some States may not have enough reporting facilities to calculate reliable measure estimates.

Notes

The State average presented in the tables is calculated by CMS and provided as part of the Hospital Compare data set.

Contact Information

Agency home page: http://www.cms.hhs.gov/.

Data system home page: http://www.hospitalcompare.hhs.gov/.

Sponsor

U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services (CMS), Office of Strategic Planning.

Mode of Administration

MCBS is a computer-assisted personal survey of Medicare beneficiaries residing in the United States and Puerto Rico. The data in this report are from repeat interviews of Medicare beneficiaries who reside in community settings. Community residents receive a baseline questionnaire and community core questionnaires. The baseline questionnaire collects information on health insurance, household composition, health status, access to and satisfaction with medical care, and demographic and socioeconomic characteristics. The core questionnaires collect information on health insurance, medical care use, charges and payments for medical care, and specific topics such as income and assets. These data are supplemented with administrative data on Medicare and Medicaid eligibility, enrollment in Medicare managed care, and claims for Medicare-covered services.

Survey Sample Design

The MCBS sample consists of older and disabled beneficiaries enrolled in Medicare Part A (hospital insurance) or Part B (medical insurance) or both. Beneficiaries are selected for the survey by using a stratified, multistage area probability design. A key feature of the survey is that sample people are followed over time as part of a 4-year rotating panel design. Each person is interviewed 3 times per year over 4 years, regardless of whether he or she resides in the community or a long-term care facility.

The data are used to produce calendar-year public use files on access to care and cost and use. All four panels are included in the access to care file. Only three panels are included in the cost and use file, since the panel that is being retired during a calendar year is not asked about medical utilization for that year.

Primary Survey Content

Medical expenditure data as well as detailed data on health conditions, health status, use of medical care services, charges and payments, access to care, satisfaction with care, health insurance coverage, income, and employment.

Population Targeted

The MCBS is a nationally representative survey of all Medicare beneficiaries, regardless of age or place of residence.

Demographic Data

Age, race, ethnicity, region, occupation, employment status, household composition.

Years Collected

1991 to present.

Schedule

Annual releases of access to care and cost and use files. The access to care file for a calendar year is released about 1 year after the end of data collection. The cost and use file for a calendar year is released about 2 years after the end of data collection.

Geographic Estimates

National. The data also can be shown for the four U.S. Census Bureau regions.

Notes

The MCBS contains four overlapping panels of Medicare beneficiaries. Each year, one panel is dropped from the survey and a new one is added. This design produces 3 calendar years of medical utilization data for each sample person. The data are collected over a 4-year period in which sample people are interviewed 12 times. The first interview collects baseline information on the beneficiary. The next 11 interviews are used to collect 3 complete years of utilization data.

Contact Information

Agency home page: http://www.cms.hhs.gov.

Data system home page: http://www.cms.hhs.gov/MCBS/.

References

Adler GS. A profile of the Medicare Current Beneficiary Survey. Health Care Financ Rev 1994;15(4):153-63.

Sponsor

U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services (CMS), Patient Safety Task Force of the Department of Health and Human Services.

Mode of Administration

Chart abstraction using electronic data collection tool. Once medical records are selected from the sample, abstractors use explicit clinical criteria to examine the process of care, ascertain the occurrence of specific adverse events, determine that the event is an adverse event, and assess patient risk factors.

Sample Design

Cross-sectional. The MPSMS data sources are a national random sample of medical records for all Medicare hospital discharges stratified by State and Medicare administrative data.

Primary Content

MPSMS inpatient medical records: a randomly selected, nationwide subset of inpatient medical records from the Medicare Hospital Monitoring Program (HPMP) and Medicare administrative data.

The purpose of the system is to identify rates of adverse events in the Medicare population as expressed in medical records and administrative data. Adverse event categories include the following:

• Adverse events associated with the use of central venous catheters.
• Adverse events associated with surgical procedures.
• Adverse events associated with joint revisions and replacements.
• Hospital-acquired, ventilator-associated pneumonia.
• Hospital-acquired bloodstream infections.
• Adverse drug events that are hospital associated or the cause of hospital admission.

The system also collects and examines patient risk factor information, such as demographics, principal diagnosis and its severity, and comorbid conditions and treatments.

Population Targeted

Medicare beneficiaries as required for the relevant measure.

Demographic Data

Age, gender, race.

Years Collected

Since 2002.

Schedule

Data are collected monthly and aggregated annually.

Geographic Estimates

All States, Puerto Rico.

Contact Information

Agency home page: http://www.cms.hhs.gov.

Data system home page: http://www.cms.hhs.gov.

References

Hunt DR, Verzier N, Abend SL, et al. Fundamentals of Medicare patient safety surveillance: intent, relevance, and transparency. In: Advances in patient safety: from research to implementation. Vol. 2. Rockville, MD: Agency for Healthcare Research and Quality; February 2005. p. 105-17. AHRQ Publication No. 050021-2. Available at: https://www.ahrq.gov/qual/advances/.

Sponsor

U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services (CMS).

Description

Under the direction of CMS, the QIO Program consists of a national network of 53 QIOs responsible for each U.S. State, territory, and the District of Columbia. QIOs work with consumers, physicians, hospitals, and other caregivers to refine care delivery systems to make sure patients get the right care at the right time, particularly among underserved populations. The program also safeguards the integrity of the Medicare trust fund by ensuring that payment is made only for medically necessary services and investigates beneficiary complaints about quality of care.

Beginning with the QIO 2002-2005 contract cycle, CMS expanded performance expectations to encompass two additional health care settings: nursing homes and home health agencies.

Data Collection

Data are collected by the national network of 53 QIOs responsible for each U.S. State, territory, and the District of Columbia according to data collection tools developed by QIOs or related organizations. The systematic random sample size varies among years and measures. For example, in the 2002 data, all measures have up to 600 records per State; in the 2000-2001 data, acute myocardial infarction and pneumonia measures have up to 750 records per State, and the heart failure measures have up to 800 records per State.

Primary Content

Current national priorities include four health care settings: nursing homes, home health agencies, hospitals, and physician offices. QIO measures included in the NHQR are hospital settings, including those related to acute myocardial infarction, heart failure, and pneumonia.

Population Targeted

Medicare beneficiaries as required for the relevant measure.

Demographic Data

Age, gender, race.

Contact Information

Agency home page: http://www.cms.hhs.gov.

Data system home page: http://www.cms.hhs.gov/QualityImprovementOrgs/.

References

Go to the QIO Statement of Work at http://www.cms.hhs.gov/QualityImprovementOrgs/04_9thsow.asp#TopOfPage.

Go to the CMS and QIOs' public Web site at http://www.medqic.org/ and the quality measures detailed therein.

Sponsor

U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services (CMS).

Mode of Administration

Completed by facility personnel, with attestation of accuracy required.

Primary Content

The MDS is the core source of assessment information for the Resident Assessment Instrument (RAI). The MDS is a standardized primary screening and assessment tool of health status; it measures physical, medical, psychological, and social functioning of nursing home residents. The general categories of data and health status items in the MDS include demographics and patient history, cognitive function, communication/hearing, vision, mood/behavior patterns, psychosocial well-being, physical functioning, continence, disease diagnoses, health conditions, medications, nutritional and dental status, skin condition, activity patterns, special treatments and procedures, and discharge potential.

Chronic care refers to those types of patients who enter a nursing facility typically because they are no longer able to care for themselves at home. These patients (or residents) tend to remain in the nursing facility anywhere from several months to several years. The chronic care quality measures were calculated on any residents with a full or quarterly MDS assessment in the target quarter.

Post-acute care (PAC) refers to those types of patients who are admitted to a facility and typically stay less than 30 days. They are also referred to as “short-stay residents.” These admissions typically follow an acute care hospitalization and involve high-intensity rehabilitation or clinically complex care. The PAC measures were calculated on any patients with a 14-day prospective payment system (PPS) MDS assessment in the last 6 months.

Demographic Data

Age, gender, age, marital status, race or ethnicity, current payment sources, health status.

Population Targeted

All residents in Medicare or Medicaid certified nursing and long-term care facilities.

Years Collected

Nursing homes have been collecting the MDS since 1990; since June 1998, States have transmitted MDS to the CMS central repository.

Data Collection Schedule

Administered on admission, quarterly, annually, whenever the resident experiences a significant change in status, and whenever the facility identifies a significant error in a prior assessment. Also, residents receiving Medicare skilled nursing facility PPS payments require more frequent assessments (5, 14, 30, 60, and 90 days).

Facilities are required to electronically transmit MDS data to the States. The State agencies have the overall responsibility for collecting MDS data in accordance with CMS specifications. The State is also responsible for preparing MDS data for retrieval by a national repository established by CMS.

Geographic Estimates

National, State, facility.

Notes

The nursing home measures were changed in January 2004.

The data were downloaded from http://www.medicare.gov/download/downloaddb.asp on April 26, 2007, and reflect care for the period 10/1/06 to 12/30/06 for chronic care and 7/1/06 to 12/30/06 for short-stay measures.

State averages are calculated as a simple average (i.e., unweighted by number of residents) of all the facilities located in the State. Reported rates may be affected by the varying number of facilities for each measure in a State due to minimum denominator sizes (see the Table footnote). In addition, reported rates may not reflect the true rate of incidence within a State because the simple averages can only be interpreted as an average of facility rates.

Risk Adjustment for Measures

Three methods are used for risk adjusting the Nursing Home Care measures that use the MDS. One method is to exclude residents where the outcomes are not under facility control (e.g., the resident has a condition, such as a pressure ulcer, upon admission) or where the outcome may be unavoidable (e.g., the resident has end-stage disease or is comatose). Another method is adjustment based on resident-level covariates that have been found to increase risk for an outcome using logistic regression models. The third method is adjustment using logistic regression models based on a Facility Admission Profile (FAP), or the proportion of residents who enter the facility over the past year already with that outcome.

Contact Information

Agency home page: http://www.cms.hhs.gov.

Oversight home page: http://www.cms.hhs.gov/NursingHomeQualityInits/.

Data system home page: http://www.cms.hhs.gov/NursingHomeQualityInits/.

References

National Nursing Home Quality Measures, user's manual. V1.2 (consolidation of original user's manual and technical user's manual). Rockville, MD: CMS; November, 2004. Available at: http://www.cms.hhs.gov/NursingHomeQualityInits/Downloads/NHQIQMUsersManual.pdf. (Plugin Software Help)