Journal of Gerontological Nursing

Nurse Staffing and QUALITY OF CARE IN NURSING FACILITIES

Jean Johnson-Pawlson, PHD, NP-C; Donna Lind Infeld, PHD

Abstract

ABSTRACT

A study of 198 nursing facilities in Maryland tested the hypotheses that: 1) the presence of more RNs improves the quality of nursing care; and 2) increased numbers of all types of nursing staff improve the qualify of nursing care, based on a multidimensional measure of quality of nursing care. Findings indicate that the ratio of RNs to residents is directly related to a measure of resident rights deficiencies. In addition, the ratio of total nursing staff to residents is directly related to a lower overall deficiency index and a higher quality of care score.

Abstract

ABSTRACT

A study of 198 nursing facilities in Maryland tested the hypotheses that: 1) the presence of more RNs improves the quality of nursing care; and 2) increased numbers of all types of nursing staff improve the qualify of nursing care, based on a multidimensional measure of quality of nursing care. Findings indicate that the ratio of RNs to residents is directly related to a measure of resident rights deficiencies. In addition, the ratio of total nursing staff to residents is directly related to a lower overall deficiency index and a higher quality of care score.

Quality of nursing (care and levels of nurse staffing in nursing facilities are critical public policy concerns. At the request of Congress, the Health Care Financing Administration (HCFA) is currently developing a report on the status of nurse staffing in nursing homes. Further, legislation, passed in 1993, directs the Secretary of the Department of Health and Human Services to conduct a study of the relationship of nurse staffing patterns in hospitals and nursing homes to quality of patient care. The study described here is an early effort to examine these concerns.

The issues of nurse staffing and quality of nursing care are inextricably interrelated. How can quality nursing care be provided if there are not enough nurses to deliver the care? There are currently two conflicting positions regarding nurse staffing levels. Professional nursing organizations and consumer advocacy groups contend mat more RNs and more total nursing staff than currently required by states are needed to provide adequate care. Nursing home residents are sicker and more debilitated than in the past (Morrisey, Sloan, Valvona, 1988; Shaughnessy & Kramer, 1990). In addition, end-of-life care is increasingly provided in nursing facilities and often requires more sophisticated and timeintensive nursing care (Sager, Easterling, Kindig, & Anderson, 1989). These changes put new and complex demands on nursing staff.

Previous research has shown that increased numbers of RNs are related to better resident outcomes including fewer pressure sores, less functional decline, and a higher rate of community discharge, as well as fewer survey deficiencies for facilities (Cherry, 1991; Mezey & Lynaugh, 1989; Munroe, 1990; Small & Walsh, 1988). Total nurse staffing has also been related to functional improvement (Specter & Takada, 1991).

The other perspective on nurse staffing is that taken by state and federal payers concerned about nursing home costs. Medicaid agencies are facing growing demands and limited resources. There is a strong incentive to limit nursing home payments by establishing maximum reimbursement levels through the use of per them rates or caps on nursing cost centers. These incentives provide an impetus for facilities to staff with the least expensive mixture of nursing stati. Evidence supports that there is a substitution of lower paid workers in response to relative wages. That is, since RNs and LPNs cost more, facilities tend to hire more nursing aides instead (McCaffree, Baker, & Perrin, 1979; Zinn, 1990). This staffing pattern could affect the quality of care provided.

RESEARCH MODEL

This study explores two questions related to current policies governing nurse staffing in nursing facilities. The first question centers on whether facilities that staff at higher levels of registered nurses (RNs) provide higher quality nursing care than facilities that staff with fewer RNs. The second question is whether facilities that staff at levels above the minimum total nursing staff required by state regulations provide a higher quality of nursing care than facilities that staff at or near the minimum. The relationships hypothesized are: 1) more RNs lead to higher quality of care, and 2) more total nursing staff leads to higher quality of care. Facility characteristics (case mix, ownership, and payer mix) are included as control variables (Figure).

Independent Variables

The independent variables examined are nurse staffing measured as RN and total nursing staff full-time equivalent (FTE) positions per resident. This staff to resident ratio was based on the facility report to the Maryland state survey agency for nurse staffing time recorded during the last 14 days of the most recent pay period immediately prior to or during the annual standard survey. Full-time equivalent RN staff includes all RNs in the facility except the director of nurses. The database does not include descriptions of RN positions. Nursing staff FTEs include all RNs in the facility except the director of nurses. Types of positions nurses are in are not included in the database.

Table

FIGUREStudy Variables

FIGURE

Study Variables

Control Variables

The variables that were controlled in this study include case mix, facility ownership, and payer mix. A facility with a high case mix will likely staff at higher rates to meet greater resident needs. In addition, recent findings suggest that surveyors may cite more deficiencies in facilities with higher case mix (Abt Associates, 1993).

Ownership has also been associated with different staffing levels. Non-profit facilities staff at higher levels of RNs than proprietary facilities (Kanda & Mezey, 1991; National Center for Health Statistics, & Strahar, 1988).

Payer mix is controlled in order to account for potential differences in facility quality of care that could be attributed to the effect of Medicaid. Munroe (1990) found that a higher proportion of Medicaid reimbursed days was directly related to poorer quality of care.

Data from the Health Care Financing Administration (HCFA) Online Survey Certification and Reporting database were used to calculate a facility case-mix index. A classification system (Burke, 1991) based on measures of activities of daily living deficits and need for special treatment was used to calculate a case-mix index for each facility.

Scores for ADLs, namely toileting, transfer, and eating, were calculated based on five levels from: (1) independent to (5) total dependence. Mobility was scored on a three-level scale of (1) ambulatory, (2) chair bound, and (3) bed bound. The total index score for each resident could range from 4 to 18. Bathing was not included in this index because regulatory requirements in many states require full assistance for bathing regardless of a resident's capability and therefore distort the usefulness of this specific ADL. A facility ADL index was computed by multiplying the limitation level of each ADL deficit by the percentage of residents needing each level of care for that ADL. Table 1 provides an example calculation of the numeric facility score for the transfer ADL. A total ADL index for a facility was computed by summing the ADL scores for toileting, transferring, eating, and mobility. The percentage of residents receiving special treatments, including parenteral feeding, suctioning, tracheostomy care, and ventilator care was added to the facility ADL score. For instance, if 15% of residents needed parenteral feeding, 5% needed suctioning, 3% needed tracheostomy care, and no one was on a ventilator, the total for all of the special procedures would be .23. This number was added to the facility ADL index to construct a facility case mix index.

Table

TABLE lCalculation of ADÌ Score for Transfer

TABLE l

Calculation of ADÌ Score for Transfer

Activities of daily living measures have been the foundation of all classification systems proposed and account for a significant amount of variance in nursing staff requirements (Fries & Cooney, 1985; Weissert & Musliner, 1992). Not included in the facility case mix index however is a measure of psychobehavioral problems. The extent that psychobehavioral variables explain variance in resource need is not clear and continues to be explored. There was no attempt to include these variables since information was not accessible from the database used for this study.

The second control variable, ownership, is categorized as: 1) nonprofit; 2) proprietary (one or two facilities); 3) proprietary-multifacility (single ownership of three or more facilities); and 4) public-nonfederal.

Finally, to control payer mix, the effect of Medicaid reimbursement was measured using the percent of residents covered by Medicaid, which captures the financial impact of Medicaid on the facility.

Dependent Variable

The multidimensional nature of the concept of quality of care leads to problems of definition and measurement. To address these difficulties, quality criteria were developed based on multiattribute utility (MAU) theory (Farquhar, 1977; Fishburn, 1978). MAU is used to build an operational, quantitative measure of a subjective, complex, multidimensional concept by capturing the judgments, values, and preferences of experts (Zeleny, 1982).

For this study, measures of nursing care quality were constructed with the input of nursing home ombudsmen and records of facility survey deficiencies. Ombudsmen input was used rather than expert RN input based on the results of a comparison of responses of 15 nurse experts and 15 ombudsmen (Johnson-Pawlson, 1993). Each nurse and ombudsman was asked to rate on a scale of 0 to 4, each item for two separate dimensions: 1) appropriateness of each item as an indicator of quality of care; and 2) the extent to which each item reflects nursing care. The responses were similar with the exception that ombudsmen rated resident assessment and resident rights higher as indicators of quality of nursing care than did the expert RNs. The ombudsmen responses were used for this study since the ombudsmen responses, while very similar to nurses, provided a broader measure of quality. Based on the ombudsmen ratings, 78 items from the 1991 Long Term Care Survey were identified as specifically relating to nursing care.

The items on the questionnaire were used to determine a comprehensive measure of nursing care as well as measures of specific dimensions of nursing care included in the HCFA Long Term Care Survey. These dimensions include resident rights, resident behavior, quality of life, resident assessment, and quality of care.

Using only Long Term Care Survey items identified by the ombudsmen as appropriate, an overall measure of nursing care quality was calculated for each facility by multiplying each item receiving a deficiency on the last survey by the average score for the strength of the relationship to nursing. For instance, if a facility received a deficiency for the item in the survey related to the prevention of pressure sores and the average of the ombudsmen rating for how strongly this item related to nursing was 4.0, 1 (representing the deficiency) was multiplied by 4. All of the 78 items that received a deficiency were similarly scored and the total was added. The total score is referred to as the overall deficiency index. For instance, similar calculations were done for each HCFA dimension, thus producing five nursing deficiency indices.

While the authors recognize that there is controversy over the accuracy of survey data, limiting the analysis to a single state eliminates the cross state variation. In addition, there is evidence that the quality of a facility identified as high or low quality using a combined measure of pressure sores per bed bound residents, restraint use, and medication error rate is related to numbers of deficiencies cited (Johnson-Pawlson & Cowles, 1995).

Data Sources

This study is based on cross-sectional data from the last standard survey conducted between October 1, 1991, and September 30, 1992, for all Medicare /Medicaid certified facilities in Maryland. Data on deficiencies, staffing, case mix, ownership, and payer mix were obtained from the Online Survey Certification and Reporting (OSCAR) database developed and maintained by HCFA's Division of Evaluation.

Findings: Maryland Nursing Facilities

A total of 217 certified Maryland facilities were identified in the OSCAR data set. Ten facilities with fewer than 10 residents and nine with significant data errors were eliminated. The remaining 198 facilities include a total of 25,918 beds and 23,343 residents. The mean size of a facility is 130 beds with a standard deviation of 74.54 and a range of 22 to 550 beds. The average number of residents is 118.

Nursing Staff Characteristics - A total of over 12,000 nursing staff are employed in Maryland facilities with an average of 7.04 FTE RNs, 10.86 FTE LPNs; and 46.70 FTE nursing assistants (NAs) per facility. Thus 64 nursing staff care for 118 residents in a typical facility. The mean RN, LPN, and NA to resident ratios of .063, .096, and .394, respectively, result in an overall staffing of slightly over one-half of a nursing staff member (.553) per resident (Table 2).

Deficiency Description - Data summarizing the deficiency rates for each of the HCFA dimensions are shown in Table 3. It should be noted that some deficiency dimensions include more items and therefore are likely to have higher scores. The average number of overall deficiencies cited in Maryland facilities for the items related to nursing care is 2.32 (sd=3.24), with a range of zero to 14.

Ownership and Payer Mix - Most nursing faculties in Maryland are proprietary and are not part of multifacility corporations (53.00%). Next are nonprofit facilities (27.80%), proprietary multi-organization facilities (16.20%), and government facilities (3.00%). This distribution of ownership is consistent with the ownership pattern in the rest of the nation (Table 4).

Payer mix was analyzed both for Medicaid and Medicare reimbursement. The data indicate that slightly fewer (65.10%) of Maryland nursing facility residents receive some level of Medicaid benefits compared to the national average (67.20%). Medicare also supports fewer (4.30%) Maryland residents compared to the national average (4.56%) (Table 4).

Resident Characteristics - Resident characteristics that served as the basis for calculating the facility case mix index are presented in Tables 5 and 6. Table 5 depicts a very disabled population with most residents disabled in toileting and transferring. However, the mean case-mix index for residents in Maryland (10.44) is slightly lower than that of the nation (10.66). In sum, Maryland facilities and residents are basically like those across the nation although slightly less likely to receive Medicaid or Medicare and slightly less disabled.

Data in Table 6 indicate that relatively few residents receive special care treatments. Injections comprise the greatest number of special care treatments with a facility mean of 11.30 residents (9.48% of residents in a facUity). This is followed by tube feedings, respiratory care, ostomy care, and suctioning, all of which are provided to fewer than 5% of nursing home residents. Tracheostomy and IV care are even more rare, affecting fewer than 1% of residents.

Table

TABLE 2Nursing Stoff in Maryland Facilities

TABLE 2

Nursing Stoff in Maryland Facilities

Table

TABLE 3Deficiencies per Facility In Maryland

TABLE 3

Deficiencies per Facility In Maryland

Table

TABLE 4Facility Ownership and Public Payer Mix

TABLE 4

Facility Ownership and Public Payer Mix

Relationship of Nursing Staff to Deficiency Indices

Results of multiple regression analysis of the relationship between the ratio of RNs to residents and overall deficiency index, controlling for case mix, ownership, and payer mix, is not statistically significant (?=-0.64, ?=2?2). However, the ratio of total nursing staff to residents and the overall deficiency index, with the control variables, is statistically significant (f=-1.86, p=.032). As the total staffing ratio increased, the deficiency index decreased. While staffing with more RNs does not appear to improve nursing quality, more nursing staff of all types does.

Analysis of the relationship between the HCFA dimensions of nursing care and the RN to resident ratio, with control variables, is presented in Table 7. The relationship is significant for the resident rights category (f=1.71, p=.044) but is not in the direction hypothesized. As the ratio of RNs to residents increases, the resident rights deficiency index increases. There is also a statistically significant relationship between the total nursing staff ratio and the quality of care index (r=-1.92, p=.028) which suggests that the relationship identified with the overall deficiency index may be focused on the quality of care dimension.

Relationship of Control Variables to Deficiency Indices

The model developed to explore the relationship between total nursing staff and RN staff to quality of care controls for facility case mix, ownership, and payer mix. As shown in Table 8, there was no statistically significant relationship between any of the control variables and the overall deficiency index (Table 8).

The lack of statistically significant relationships between case mix and payer mix with any of the deficiency indices is important. Even though the literature suggests that these variables may influence quality of care, neither provided a unique contribution to explaining the difference in the overall deficiency index or the sub-indices. However, there was a significant relationship between nonprofit facilities and the quality of life deficiency index. Nonprofit facilities had significantly fewer quality of life and resident rights deficiencies than facilities with other types of ownership (Table 8).

DISCUSSION

RNs and Quality of Nursing Care

The findings of this study do not support the hypothesis that a higher RN to resident ratio is related to better care. The number of RNs in nursing facilities is extremely small. On average, there are only 20 minutes of RN time per resident per day. This time is allocated to numerous responsibilities including scheduling staff, attending meetings, supervising other staff, and conferring with family members, physicians and other health professionals, as well as providing direct resident care. This limited time allocated to multiple responsibilities could result in RNs not having an impact on quality of nursing care.

Another aspect of limited RN staff is the narrow range of RN staff levels found in Maryland homes. The result is that the RN staff level is a truncated variable which distorts possible real effects.

Other variables that affect RN performance, including skill levels, turnover, and experience could affect RN impact on quality. In addition, facility-related characteristics which were not controlled, such as differences in management philosophy, availability of resources to nursing staff, or participation of family members or volunteers, could account for lack of an RN effect.

Table

TABLE 5Residents per Facility Needing Assistance with ADLs

TABLE 5

Residents per Facility Needing Assistance with ADLs

Table

TABLE 6Residents Receiving Special Care

TABLE 6

Residents Receiving Special Care

Policy implications for enhancing the role of RNs in nursing facilities depend on the reason for the lack of relationship. If RNs really have no impact on quality of care, then requiring one or more additional RNs will make no difference. However, this is counter-intuitive. To improve care significantly, substantially greater numbers of RNs may be needed, greater numbers than were present in the Maryland sample or in most nursing facilities in the U.S. today.

Table

TABLE 7Relationship Between Nurse Staffing and HCFA Deficiency Indices

TABLE 7

Relationship Between Nurse Staffing and HCFA Deficiency Indices

Table

TABLE 8Relationship Between Control Variables and HCFA

TABLE 8

Relationship Between Control Variables and HCFA

A surprising finding was that as the number of RNs increased, so did the number of resident rights deficiencies. This may be more than a spurious finding. One possible explanation is that most RNs work in isolation to some extent from other professional staff. A consequence may be that RNs may feel they must have the answers and demonstrate their knowledge and skills at the expense of resident rights. However, this finding is particularly significant to RNs and nursing organizations since resident rights are recognized as an important aspect of nursing home care. RNs are the key staff members who must recognize and facilitate resident rights if residents are to have the level of autonomy that is consistent with even minimally acceptable quality of care. If RNs, in fact, tend to infringe on residents' rights, changes in training or in practice would clearly be indicated.

Total Nursing Staff and Quality of Nursing Care

It would seem obvious that greater numbers of nursing staff will produce a higher quality of care up to some undefined point of inefficiency. However, this relationship has not been supported until recently. Results of this study indicate that the overall deficiency index as well as the quality of care deficiency index are significantly related to the total nursing staff ratio. These findings are consistent with those reported by Spector and Takada (1991) which demonstrated that higher staff levels were related to functional improvement.

States have established staffing levels which are supposed to allow for provision of care to meet federal requirements. Therefore, facilities that staff at the minimum should receive no more deficiencies than facilities that staff at higher levels. This study shows, however, that facilities staffing at or close to a minimum level of required nursing staff are more likely to provide a poorer quality of care based on deficiency citations than those that staff at higher levels. This finding calls into question the current staffing level that is the basis of Medicaid and Medicare reimbursement. This finding suggests the need for higher staffing levels, which has huge financial implications for Medicare, Medicaid, as well as private payers of nursing facility services. Staffing is at the crux of the tradeoff between cost and quality.

Control Variables and Quality of Care

Ownership of the facility was significantly related to both the quality of life and residents' rights measures. These findings are at least partially consistent with the relationships discussed by Scanlon (1980) and Palmer (1985) who hypothesized that non-profit facilities want to maximize size and quality, subject to no loss constraints. In other words, nonprofit facilities may provide better care because they are not as driven by the profit motive as for-profit facilities. Other possible explanations for these relationships is that nonprofit facilities value quality of life and resident rights more highly or that they use their fundraising capacity to supplement programs which enhance both domains.

LIMITATIONS OF THE STUDY AND IMPLICATIONS FOR FUTURE RESEARCH

One limitation of this study is the use of survey deficiencies as a measure of quality of care. Concern has been expressed about the reliability and validity of state survey data. However, a documentation review of survey or deficiency reports indicated that of 21 sample states Maryland had the second lowest rate of insupportable deficiencies suggesting some evidence that use of deficiency citations in Maryland is a valid measure of quality (JohnsonPawlson, 1992).

The inverse relationship between the total nursing staff ratio and the overall deficiency index adds some credibility to survey findings since one would expect this relationship to exist. However, the validity of the individual indices reflecting different dimensions of care has not been validated.

An additional concern relates to the impact of historic events on internal validity. The results of this study are based on new regulations for nursing facilities implemented on October 1, 1990. This study should be repeated to determine the reliability of the findings, particularly to determine if the relationships change as both facilities and survey agencies gain experience with the new regulations.

Limiting this study to a single state has the advantage of eliminating variance caused by survey procedures. However, it also limits study generalizability. This study should be replicated in other states to explore the generalizability of the findings.

Finally, examination of the relationship of RNs to quality of care needs to be extended to include additional variables such as staff turnover, education, experience, and management style of RNs. Further, the RN effect wül need to be evaluated within a framework that allows for significant differences in RN to resident ratios. The marginal benefit of an RN compared with either an LPN or NA should also be examined.

NURSING IMPUCATIONS

The most significant implication of this study for nursing resides in the areas of public policy activities and research. Nursing as a profession needs to examine its commitment to providing safe, high quality nursing care to nursing home residents. It is a sad commentary that there may not be a sufficient number of RNs or adequately prepared RNs in any facility to make a significant difference in the care of residents. While there is a role for nursing assistants and LPNs, having a requirement of one RN for 8 hours a day, 7 days a week is simply not an adequate requirement. Nursing needs to continue to evaluate the relationships between RN staff and quality and press for requirements that truly provide adequate care.

Nursing also has a responsibility to review the educational programs preparing all levels of RNs for a role in a nursing facility. Currently, while many programs have gerontologie content incorporated, few have advanced clinical courses offered in nursing facilities. Roles in nursing facilities require extensive clinical knowledge as well as management and leadership skills.

This study also indicates that the total number of staff, including LPNs and NAs makes a difference in quality of care, yet, even those facilities that provided a lower quality of care had higher than the minimum staffing levels required by law. In general nursing needs to make a concerted effort to establish guidelines for staffing patterns at all levels that will yield acceptable care.

There also needs to be a continued effort to establish better measures of nursing quality on which to base judgments about the effect of staffing patterns. Survey data, while available and which have some credibility, are at best gross measures of how well nursing care is provided. Continued work to define quality of care in nursing facilities and measurement methods is needed.

REFERENCES

  • Abt Associates, Inc. (1993). Briefing points on preliminary evaluation requests. HCFA Leadership Conference.
  • Burke, R. (1991). Multistate nursing home case-mix and quality demonstration. Unpublished manuscript.
  • Cherry, R. (1991). Agents of nursing home quality of care: Ombudsmen and staff ratios revisited. The Gerontobgist, 31(3), 302-308.
  • Farquhar, RH. (1977). A survey of multiattribute utility theory and applications. In M.K. Starr & M. Zeleny (Eds.), Studies in the management sciences: vol. 6. Multiple criteria decision making (pp. 59-90). Amsterdam: North Holland Publishing Co.
  • Fishburn, PC. (1978). A survey of multiattribute/multicriterion evaluation meories. In S. Zionts (Ed.), Lecture notes in economics and mathematical systems: Vol. 155. Multiple criteria problem solving (pp. 181-224). New York, NY: Springer-Verlag.
  • Fries, B.E. & Cooney, LM. (1985). Resource utilization groups: A patient classification system for long term care. Medical Care, 23, 110122.
  • Johnson-Pawlson, J. (1992). Surveyor performance study. Washington, DC: American Health Care Association.
  • Johnson-Pawlson, J. (1993). The relationship between nursing staff variables and quality of care in nursing facilities. Ann Arbor, MI: University Microfilms Dissertation Services.
  • Johnson-Pawlson, J. & Cowles, CM. (1995). The context of nursing care in nursing homes. Manuscript submitted for publication.
  • Kanda, K., & Mezey, M. (1991). Registered nurse staffing in Pennsylvania nursing home: Comparison before and after implementation of Medicare's prospective payment system. The Gerontologist, 32(3), 318-324.
  • McCaffree, K., Baker, J. & Perrin E. (1979). Long-term care, case mix, employee time and costs (Contract No. 230-76-0285). Washington, DC: Department of Health, Education and Welfare.
  • Mezey, M.D., & Lynaugh J.E. (1989). The teaching nursing home program. Nursing Clinics of North America, 24(3), 769-780.
  • Munroe, DJ. (1990). The influence of registered nurse staffing on the quality of nursing home care. Research in Nursing and Health, 13, 263-270.
  • Morrisey, M.A., Sloan, RA., & Valvona, J. (1988). Medicare prospective payment and posthospital transfers to subacute care. Medical Care, 26, 685-698.
  • National Center for Health Statistics, & Strahar, G. (1988). Characteristics of registered nurses in nursing homes: Preliminary data from the 1985 National Nursing Home Survey. Advance Data from Vital and Health Stetistics, No. 152, (DHHS Pub. No. [PHS] 87-1250). Hyattsville, MD: Public Health Service.
  • Palmer, HC. (1985). Studies of nursing home costs. In RC. Vogel & HC. Palmer (Eds.), Long term care perspectives from research and demonstrations (pp. 665-722). Rockville, MD: Aspen.
  • Sager, M.A., Easterling, D. V, Kindig, D.A., & Anderson, O. W. (1989). Changes in the location of death after passage of Medicare's prospective payment system. The New England Journal of Medicine, 320(7), 433-439.
  • Scanlon, W. (1980). A theory of the nursing home market. Inquiry, 17(1), 25-41.
  • Shaughnessy, P., & Kramer, A. (1990). The increased needs of patients in nursing homes and patients receiving home health care. The New England Journal of Medicine, 322(1), 21-27.
  • Small, N., & Walsh, M. (Eds.). (1988). Teaching nursing homes, The nursing perspective. Owings Mill, MD: National Health Publishers.
  • Spector, W.D., & Takada, H (1991). Characteristics of nursing homes that affect resident outcomes. Journal of Aging and Health, 3(4), 427-454.
  • Weissert, W. & Musliner, M. (1992). Access quality and cost consequences of case-mix adjusted reimbursement for nursing homes: A critical review of the evidence. Washington, D.C.: Public Policy Institute, & American Association of Retired People.
  • Zeleny, M. (1982). Multiple criteria decision making. New York, NY: McGraw-Hill.
  • Zinn, J.S. (1990). The extent and the determinants of variations in nursing home staffing and clinical practice. Unpublished doctoral dissertation, University of Pennsylvania.

FIGURE

Study Variables

TABLE l

Calculation of ADÌ Score for Transfer

TABLE 2

Nursing Stoff in Maryland Facilities

TABLE 3

Deficiencies per Facility In Maryland

TABLE 4

Facility Ownership and Public Payer Mix

TABLE 5

Residents per Facility Needing Assistance with ADLs

TABLE 6

Residents Receiving Special Care

TABLE 7

Relationship Between Nurse Staffing and HCFA Deficiency Indices

TABLE 8

Relationship Between Control Variables and HCFA

10.3928/0098-9134-19960801-11

Sign up to receive

Journal E-contents