Nursing Facilities, Staffing, Residents and Facility Deficiencies, 2009 Through 2016

Appendix: Technical Notes

Data Sources

OSCAR/CASPER is an on-line data system from the Centers for Medicare and Medicaid Services (CMS). The OSCAR data for this report are for 2009-2011, and these data were converted to CASPER data in 2012-2016 for all facilities. The OSCAR/CASPER systems include data for all certified nursing facilities in the U.S.  The data are collected in separate sets of files: (1) provider information, staffing data and health information on residents; and the (2) survey deficiencies. To create this report, the OSCAR/CASPER data from the annual surveys were combined with data from complaint surveys.

All nursing facilities federally certified for Medicare (skilled nursing care) and Medicaid (nursing facilities) and surveyed during the calendar year were included (about 15,400 to 15,600 facilities) in these data.  Intermediate care facilities for people with intellectual and developmental disabilities (ICF-I/DD) were excluded because they use different federal certification forms and have different standards of care.  Facilities located in the U.S. territories and Puerto Rico were excluded.

OSCAR/CASPER is a set of administrative databases that allow users to add, change, and delete data almost continually. These databases store no more than four standard surveys per provider. The system automatically deletes older surveys as new ones are entered. For these reasons, analysis of the same data elements may yield slightly different results depending upon the date the data are retrieved. Although facilities are surveyed every 9-15 months, this report used data during the calendar year only from both annual surveys and complaint surveys.1 Because data are overwritten in the database, OSCAR/CASPER data were retrieved twice a year and then merged to create the most complete file for each facility for each calendar year.  This report uses deficiency data that were obtained from the annual surveys and all complaint surveys. This approach gives a complete picture of all deficiencies throughout the year rather than at the time of the annual survey.  This report does not include the life safety code violations.  Therefore, this report varies from the CMS Nursing Home Compendium, which only uses data from the annual surveys.2

Outcomes Included

This report presents calendar year data on nursing facilities, staffing, resident characteristics, and surveyor reports of quality deficiencies by state.

  • Information on facility characteristics includes type of certification, bed size, occupancy, ownership, hospital-based and chain affiliations and other facility characteristics.
  • Resident characteristics are as reported by the nursing facilities. These include limitations in activities of daily living (ADLs), restraints, incontinence, psychological problems, and other special care needs of residents.
  • Nurse staffing (RNs, LPN/LVNs, and NAs) hours per resident day are presented for nursing facilities. These data are reported by facilities for a two-week period prior to when the state survey was conducted.  These data currently are the only major source of information for all facilities on staffing levels.
  • Finally, data are presented on facility deficiencies based on state surveyor evaluations of the process and outcomes of care in the facilities. Deficiencies are reported in 8 categories established by CMS for the Medicare Nursing Home Compare 5-Star website Deficiency data: (1) quality of care; (2) mistreatment; (3) resident assessment; (4) resident rights; (5) environment; (6) nutrition; (7) pharmacy; and (8) administration. The information compiled shows differences in the frequency of the deficiencies by type and category.3

Background on the Survey System and Data Collection

Every facility must have an initial survey to verify compliance with all federal regulatory requirements in order to be certified. Certified facilities are resurveyed no less often than every 15 months.  Follow-up surveys may be conducted to ensure that facilities correct identified deficiencies. In addition, surveys are required when there are substantial changes in a facility’s organization and management. Finally, surveys may be conducted to investigate complaints about violations and poor care.

Nursing facility data are collected in two different ways. First, the facility characteristics, resident characteristics, and staffing levels are completed on standardized forms by individual nursing homes at the beginning of each survey and are certified by the facility as being accurate. State staff enters the data into a computerized OSCAR/CASPER system data.

Second, state surveyors make decisions regarding whether the facility has met or not met each standard after the facility survey has been completed. If a facility is judged to not meet a standard, the facility is given a deficiency indicating that the regulation was not met.  Surveyors are also required to determine the scope and severity of each deficiency, and these are recorded as part of the process.

The survey evaluations are based upon data from a combination of sources including, but not limited to, the assessment of a selected sample of individual residents; interviews with a sample of residents, family members and staff; a review of the resident records and facility documents; and other data.  After these judgments are made, state surveyors record and enter the data for each item into OSCAR/CASPER.  Thus, determinations of deficiencies are made by state surveyors independent of the facility, with standard forms, sampling and survey procedures to ensure accuracy.  Team members and state supervisors subsequently review state surveyor deficiencies.  Facilities have the option to challenge and appeal deficiency decisions through administrative review procedures.

Because of these checks in the system, the likelihood of false positive deficiencies is low, and errors tend to be in under-reporting of failures to meet standards.4,5,6,7,8,9,10,11,12,13  Thus, a note of caution is needed that under-reporting of deficiencies is more likely to be a problem than over-reporting.

CMS Procedures and State Survey Variation

The Centers for Medicare & Medicaid Services (CMS) uses “front-end” edit screens to ensure the accuracy of the OSCAR/CASPER data. State staff enter the data for each survey item into the OSCAR/CASPER data set within 45 days of each survey, leading to time delays in obtaining the data.

One concern about the OSCAR/CASPER data has been with survey reliability (both across and within states) in judging the quality of facilities.14 These issues have been addressed over time by CMS with its survey process.  First, the procedures require some accuracy checks by surveyors.15 CMS provides federal training for new state surveyors. Recently, CMS has been testing ways to improve the efficiency and effectiveness of the process using a new Quality Indicator Survey (QIS) in comparison to the traditional procedures that are used.16 In addition, CMS uses federal survey teams from the regional offices to conduct periodic oversight surveys of state agencies.  States that fall below the concurrence standards established by CMS are critiqued and monitored by CMS.

Even though CMS has made efforts to standardize the reporting of deficiencies by state survey staff, regional variations in issuing deficiencies continue and enforcement procedures have been inconsistent and ineffective.17,18,19,20,21,22,23,24,25,26,27,28,29  Problems have also been found with the complaint investigation process, 30,31,32,33  and quality of care problems continue to exist in nursing facilities.34,35 In spite of the improvements that have been made, there is a need to further improve both the state survey process and the enforcement system. Staff turnover and recruitment problems and fiscal problems at the state agency level may hamper survey and enforcement efforts.36,37,38

Data Cleaning and Duplicate Records

In preparing the data for this report, discussions were held with CMS officials as necessary to discuss data acquisition, formatting, and cleaning issues.  Frequency distributions were developed for all the indicators on the data set and a series of cleaning activities undertaken.

The first step in cleaning was to eliminate any duplicate provider records.  Duplicate records were considered generally to be the result of changes in certification for Medicare and Medicaid (Title VIII/XIX) facilities or Medicare only (Title XVIII) facilities.  Since 1990, because of the OBRA 1987 legislation, Medicaid only (Title XIX) SNF and ICF facilities are certified together as nursing facilities (NFs).  With this change, some problems with duplicate SNF and ICF reporting were essentially eliminated.

To correct the duplicate problems, we identified all those facilities showing identical values for the following areas: state, city, facility name, and facility address.  Where duplicate records were identified, a decision-making procedure was invoked as follows:

(1)  The most recent record within a calendar year was selected over earlier records; and

(2)  A record reporting a category of either: (01) skilled nursing facility (SNF) — Medicare participation; (02) nursing facility (NF) — Medicaid participation; or (03) SNF/ICF — Medicare/Medicaid participation was selected over facilities reporting as (10) hospital based.

For those remaining facilities with the same name and/or address, data on telephone numbers and survey data were then examined.  Where there appeared to be two facilities at one site with different data, neither facility was eliminated.  This overall process resulted in dropping about 100 facilities annually prior to 2012. Fewer duplicates were found among the CASPER records each year since 2011.

Outcome Measurement, Data Errors and Corrections

The data were examined for missing values and gross errors.  Means and standard deviations of the data were computed and examined.  Logic checks confirming the reasonableness of data were conducted.  Preliminary work identified some missing data and errors, primarily in reporting the beds, staffing and the number of residents.  Each problem area and the cleaning procedures are discussed below.

Total Number of Beds

During the preliminary work, the total number of beds in each state reported on OSCAR/CASPER was compared to the total licensed bed supply for each state from an independent survey of states.  The total number of beds reported by facilities was significantly higher than the total licensed beds in the U.S.  An analysis of this discrepancy found that some hospital-based facilities had reported the total number of acute care beds rather than the total number of skilled nursing beds for their facility.  To correct this error, the maximum number of beds for a hospital-based facility was set to equal the maximum number of certified skilled nursing beds in the facility.  This process made the total certified nursing facility beds more comparable to the total state licensed nursing facility beds in each state.

Total Number of Residents

Several problems concerning resident data were detected.  First, some facilities had missing data for their total number of residents.  These facilities were left in the data set, but where resident data were not available, these facilities were not included in the tables reported.

Second, some facilities reported extremely low numbers of residents.  In order to identify facilities with possible errors in reporting residents, occupancy rates were computed for all nursing facilities.  Free-standing (non-hospital based) facilities reporting 50 or fewer residents than total beds were considered to have erroneous data and were eliminated from the staffing and resident characteristics’ analysis.

Third, some facilities reported more residents than beds, suggesting more than 100 percent occupancy.  Hospital-based facilities may have had approval for swing beds, which would allow the hospital to use an acute care bed for a nursing facility resident.  Therefore, hospital-based facilities with more residents than beds were left in the data set, but the maximum occupancy rate for such facilities was reported to be 100 percent. Finally, facilities with numbers of residents reported at greater than 100 percent of total beds, which were not hospital-based, were dropped from the analysis.

Resident Characteristics

Minimum Data Set (MDS) assessments are required for all residents in nursing facilities on a periodic basis and are used to describe resident characteristics. The first MDS assessments were developed and implemented and sent to CMS electronically.39  Over time the MDS was improved using the revised MDS 2.0.40  In 2011, CMS implemented an improved MDS 3.0 version.41,42 These data are used by nursing facilities for quality improvement and by CMS to create quality measures.

The MDS is also used by Medicare and many state Medicaid programs to take resident casemix into account in reimbursement rates.43  The Medicare developed resource utilization groups (RUGS) to measure case mix and to estimate the amount of staffing time needed to care for residents in each RUG category.44,45,46 The MDS and RUGs data are separate from the OSCAR/CASPER data and not shown in this report.

This OSCAR/CASPER report has summary data on residents at the facility level describing the residents’ need for assistance with activities of daily living (ADL).  Two types of summary data are presented. First, a simple summary of three major activities of daily living (ADLs) was compiled.   The facilities were asked to rate each resident’s ADLs on a scale of 1 to 3 from “needs little or no assistance” to “needs extensive assistance.”  The three ADL scores were for those residents who needed assistance in: (1) eating, (2) toileting, and (3) transferring.  A score of 1 was assigned to residents who were independent.  A score of 2 was assigned to those that needed some supervision.  A score of 3 was assigned to those who were dependent.

Each ADL score was multiplied by the number of residents in that category for each facility.  An average composite score was developed by adding each of the three scores together and dividing by the total number of residents in the facility to compute each facility’s index score.  Thus, a summary case mix score ranging from 3-9 was compiled for each facility based on resident ADL characteristics. Individual facility scores were then summarized for each state.

Finally, the report shows resident needs for more advanced care including the: percent of clients receiving special treatments (injections, ostomy care, IV feedings, tube feedings, or suctioning), and percentage with organic psychiatric or other psychiatric conditions.  Other characteristics included are the percentage who receive psychotropic drugs and who have pressure ulcers, contractures, incontinence, or catheters.

Staffing Data

Nursing personnel in nursing facilities were of particular interest for this report.  Nursing personnel included: registered nurses (RNs); licensed practical/ vocational nurses (LPN/LVNs), and nursing aides/orderlies/ assistants (NAs). Staffing hours (including full-time, part-time, and contract staff) are reported by facilities as total hours worked in a fourteen-day period.  The staff time includes all administrative and direct care time. To compute the staffing ratios for this report, the total number of staffing payroll hours reported in a two-week period was divided by the total number of residents and by the 14 days in the reporting period.  For this report, the total hours of staffing per resident day were examined for all dually certified facilities (Title XVIII/XIX), for Medicare-only facilities (Title XVIII), and for Medicaid-only facilities (Title XIX).  It should be noted that the reported staffing ratios reflect reported hours per resident day and not the actual hours of care delivered directly to residents.  In the future, CMS plans to require payroll data for reporting staffing in nursing facilities, which should increase the accuracy for staffing data.47

There were a number of problems identified with the facility staffing data.  Some facilities reported extremely high staff hours per resident day while others reported no registered nurses or no nursing staff hours.  Where a facility reported nurse staffing hours per resident day that did not fall within a reasonable range, the data for that item were considered invalid.  The following uniform decision-making rules were created for eliminating facility staffing data which clearly appeared to be too high or too low:

First, facilities with average nursing hours per resident day that were greater than 24 hours of nursing per resident were considered erroneous and eliminated from the analysis. Distributions of the nursing hours per resident were then examined.

To correct further for staffing levels that were unreasonably high, facilities reporting staffing hours per resident day in the upper 2 percent by type of facility (separately for Medicaid only and for Medicare only/dually certified) and by type of staff (RN, LVN/LPN, and nursing assistants) were eliminated from the staffing analysis.

Facilities reporting extremely low staffing hours per resident day were identified.  Since some Title XIX facilities and Title XVIII/XIX facilities were given federal waivers from the staffing requirements, these facilities may have few or no RN staff.

Since all facilities are required to have some licensed nurses, nursing facilities with no licensed staff (RNs and/or LVN/LPNs) and/or no nursing staff were eliminated from these analyses.  In addition, facilities with computed staffing levels lower than 1 percent for licensed or combined nursing personnel for each type of facility (separately for Medicaid only and for Medicare only and/or dually certified) were removed from the staffing report because some of these may have been erroneous.

Other reporting errors in staffing data may occur.  For example, facility errors in the reporting of time periods may have occurred or rounding errors may have occurred.  These types of errors cannot be detected in the data set.  Thus, because further accuracy checks could not be conducted, only the high and low outlier facilities were removed from the tables on staff.

RNs, LPN/LVNs, and Aides are presented separately.  Total licensed nurses are also presented; these include RNs and LPN/LVNs added together.  Total combined nursing personnel are included as combined RNs, LPN/LVNs, and nursing aide hours.

Deficiency Data

State surveyors assess both the process and the outcomes of nursing home care in 8 categories: administration, environment, mistreatment, nutrition, pharmacy, quality of care, resident assessment, and resident rights. Each of these categories has specific regulations that state surveyors review to determine whether or not facilities have met the standards. Each of the specific requirements that goes into a deficiency area has a measurement and an identifying number (F-tag). In July 1995, the federal government consolidated the 325 measures of quality to about 185 measures, and additional standards have been consolidated over time, so this report includes data for about 175 F-tags. Some of the definitions of requirements have changed over time. A detailed list of all the F-tags and longer descriptions are shown in the Supplemental Tables. Detailed definitions of deficiencies are given in the CMS State Operations Manual.48

Where a facility fails to meet a requirement, a deficiency or citation is given to the facility for that individual requirement. The deficiencies are given for problems that can result in a negative impact on the health and safety of residents. Since 1995, surveyors have rated each deficiency based on scope and severity for purposes of enforcement.  The deficiencies rated as causing actual harm or immediate jeopardy are the most serious (rated at a G level or higher).49

The tables in this report and in the Supplemental Tables include information for the following deficiencies:

  • Accident Environment (F323): Facilities must ensure that the environment is as free of accident hazards as possible. This is designed to prevent unexpected and unintended injury.
  • Activities of Daily Living Services (F312): Residents who are unable to carry out activities of daily living (ADL) should be given the necessary services to maintain nutrition, grooming, and personal and oral hygiene.
  • Activities Program (F248): Facilities must provide residents with ongoing activities that meet the interests and the physical, mental, and psychosocial well-being needs of each resident.
  • Bladder Incontinence Care (F315): Residents who have bladder incontinence should receive appropriate treatment and services to prevent incontinence and to restore as much bladder functioning as possible.
  • Clinical Records (F514): The facility must maintain clinical records on each resident in accordance with accepted professional standards and practices that are: (i) complete; (ii) accurately documented; (iii) readily accessible; and (iv) systematically organized.
  • Comprehensive Care Plan (F279): The facility must develop a comprehensive care plan for each resident that includes measurable objectives and timetables to meet a resident’s medical, nursing, and mental and psychosocial needs that are identified in the comprehensive assessment.
  • Dignity (F241): Facilities must promote care for residents in a manner and in an environment that maintains or enhances dignity and respect in full recognition of each resident’s individuality. This involves assisting residents to be well groomed and dressed appropriately; promoting independence in dining; allowing private space and property; speaking and listening respectfully; and facilitating resident communications.
  • Food Sanitation (F371): Sanitary conditions must be ensured in storing, preparing, distributing, and serving food to prevent food borne illness.
  • Housekeeping (F253): Housekeeping and maintenance services must be provided to maintain a sanitary, orderly, and comfortable environment.
  • Infection Control (F441): The facility must establish and maintain an Infection Control Program designed to provide a safe, sanitary and comfortable environment and to help prevent the development and transmission of disease and infection.
  • Limited Range of Motion Services (F318): Residents with limited range of motion must receive appropriate treatment and services to increase and/or to prevent declines in range of motion.
  • Nutrition (F325): Facilities must ensure that residents receive acceptable nutrition to maintain body weight unless a resident’s condition makes this impossible.
  • Pharmacy Consultation (F431): The facility must employ or obtain the services of a licensed pharmacist who establishes a system of records of receipt and disposition of all controlled drugs in sufficient detail to enable an accurate reconciliation; and determines that drug records are in order and that an account of all controlled drugs is maintained and periodically reconciled.
  • Physical Restraints (F221): Residents have the right to be free of physical restraints imposed for purposes of discipline or convenience and not required to treat the resident’s medical symptoms. Restraints are defined as mechanical devices, materials, or equipment that restricts freedom of movement or normal access to one’s body.
  • Pressure Sores (F314): Facilities must ensure that residents without pressure sores do not develop them if this is avoidable.
  • Qualified Personnel (F282): Care must be provided by qualified persons in accordance with each resident’s written plan of care.
  • Quality of Care (F309): Each resident must receive and the facility must provide the necessary care and services to attain or maintain the highest practicable physical, mental, and psychosocial well-being, in accordance with the comprehensive assessment and plan of care.
  • Sufficient Nursing Staff (F353): Facilities must have sufficient nursing staff to provide nursing and related services to attain or maintain the highest practicable physical, mental, and psychosocial well-being of residents.
  • Unnecessary Drugs (329): Each resident’s drug regimen must be free from unnecessary drugs including (i) in excessive dose (including duplicate therapy); or (ii) for excessive duration; or (iii) without adequate monitoring; or (iv) without adequate indications for its use; or (v) in the presence of adverse consequences which indicate the dose should be reduced or discontinued; or (vi) any combinations of these reasons.
Tables Supplemental Tables

KFF Headquarters: 185 Berry St., Suite 2000, San Francisco, CA 94107 | Phone 650-854-9400
Washington Offices and Barbara Jordan Conference Center: 1330 G Street, NW, Washington, DC 20005 | Phone 202-347-5270 | Email Alerts: | |

The independent source for health policy research, polling, and news, KFF is a nonprofit organization based in San Francisco, California.