Journal of Gerontological Nursing

Innovations in Long-Term Care 

Prevalence Rate of Pressure Ulcers in California Nursing Homes: Using the OSCAR Database to Develop a Risk-Adjustment Model

Mary Ellen Dellefield, PhD, RN

Abstract

This database contains approximateiy 700 variables that can be used to examine issues related to the quality of nursing home care.

Abstract

This database contains approximateiy 700 variables that can be used to examine issues related to the quality of nursing home care.

Nursing homes participating in the Medicare or Medicaid programs are required to achieve and maintain compliance with numerous regulatory standards. Compliance is determined during surveys conducted at each facility approximately every 12 months on average and no less than once every 15 months.

Surveyors from state agencies under contract with the Centers for Medicare and Medicaid (CMS) determine compliance. Substandard compliance results in citation of a deficiency. Deficiencies are characterized according to their level of scope and severity. Scope reflects the pervasiveness of the deficiency in the facility, and severity reflects the impact of the deficiency on the safety of the resident (State Operations Manual, 1999).

Data related to regulatory compliance are maintained and retrieved through a computerized national database called the Online Survey, Certification, and Reporting (OSCAR) system. The OSCAR database contains approximately 700 variables including facility, resident, staffing characteristics, and survey deficiencies with the accompanying scope and severity rating.

Both facility staff members and surveyors collect the data contained in the OSCAR database. At the beginning of a nursing home survey, staff members collect and record facility, resident, and staffing data on CMS Forms 671 and 672. Examples of facility variables are:

* Facility name and address.

* Type of certification (i.e., Medicare, Medicaid, dual program participation).

* Bed size.

* Occupancy.

* Ownership type (i.e., government, profit, not-for-profit).

* Affiliation with a chain or hospital.

* Presence of specialized units (AIDS, head trauma, Alzheimer's disease).

Resident characteristics include the number of residents with different levels of:

* Bowel/bladder function.

* Mental status.

* Mobility.

* Skin integrity.

* Special health care treatments such as chemotherapy or dialysis.

* Medication use.

The total number of residents having these characteristics within each facility is reported. Facility staffing levels for the 2-week period prior to the survey are recorded for nursing, physician, rehabilitation, and other clinical and administrative staff. Instructions for completing the forms and definitions of data items are provided by CMS and are available to staff members.

Surveyors record the deficiency information. Citation of a deficiency is based on the surveyor's assessment of the facility's compliance with regulations. Surveyors use extensive protocols established by CMS to determine whether the facility has met the regulatory standards. Each state office enters the deficiency data into the OSCAR database. The national data are compiled, edited, and maintained by CMS.

As with all administrative databases, the OSCAR database has some inherent limitations (Institute of Medicine, 2001). During the survey, the surveyors do not audit the accuracy of the clinical and staffing data that are collected. Staffing and resident data are aggregated to the facility and do not include information on individual nursing home residents or staff members. Clinical prevalence rates are calculated using aggregated resident data.

These rates reflect one point in time, making it difficult to determine whether the clinical characteristics were present on admission or were acquired by residents during their stay in the facility. The prevalence rates calculated from OSCAR data are not risk adjusted. Without risk adjustment, one cannot determine whether prevalence rates are related to differences in the acuity or case-mix of the residents or to other organizational characteristics of facilities (Berlowitz et al., 1996). Risk adjustment is a process of statistically controlling for the heterogeneity of clinical characteristics of nursing home residents in each facility, making it possible to make meaningful quality comparisons of facilities with one another.

The validity and reliability of deficiency data is problematic (Abt, 1996; U.S. General Accounting Office, 2002a). Individual state and regional variations in surveyors' determination of inadequate regulatory compliance have been found (Dellefield, 1999; Harrington, Zimmerman, Karon, Kob'mson & Beutel, 2000; Mukamel, 1997). CMS is making efforts to decrease the variation, to the extent that it is not reflective of actual quality differences within faculties. Increasingly refined survey protocols and greater surveyor training are being instituted to address this issue.

Despite these limitations, OSCAR data are widely used by consumers and researchers. OSCAR data are publicly reported as part of an effort to use market forces to improve quality in nursing homes (U.S. General Accounting Office, 2002 b); these data are used in the "Nursing Home Compare" site found on the CMS Website (www.medicare.gov/Nhcompare/H ome.asp). Because of the variety of variables available in OSCAR, it has been used extensively by health services researchers in examining issues related to the quality of nursing home care (Harrington et al., 2000; Harrington, Woolhandler, Mullan, Carrillo, & Himmelstein, 2001; Hughes, Lapane, & Mor, 2000).

The purpose of this article is to demonstrate the importance of OSCAR data to gerontological nurses and its usefulness to nurse researchers and clinical managers by describing how OSCAR data were used to develop a risk-adjustment model of the prevalence rate of pressure ulcers. Model results are compared to protocols and a scale used for pressure ulcer risk assessment in nursing homes. Recommendations for further research using the OSCAR database and clinical management implications of the study findings are offered.

LITERATURE REVIEW

Pressure ulcers are skin lesions caused by unrelieved pressure that damages underlying tissue. The degree of tissue damage is categorized into four levels of increasing severity, or Stages I through IV. The definitions of each stage are based on recommendations published by the Agency for Health Care Policy and Research (AHCPR), now the Agency for Healthcare Research and Quality (AHRQ), in Clinical Practice Guideline Number 3. Pressure Ulcers in Adults: Prediction and Prevention (U.S. Department of Health and Human Services, 1992).

Table

TABLE 1RISK FACTORS ASSOCIATED WITN PRESSURE ULCER PREVALENCE RATES IN NURSING HOMES

TABLE 1

RISK FACTORS ASSOCIATED WITN PRESSURE ULCER PREVALENCE RATES IN NURSING HOMES

Pressure ulcer prevention and treatment is a specific regulatory standard of quality of care assessed by nursing home surveyors. Pressure ulcers are perceived by consumers and nursing home advocates as largely preventable, demonstrated by the increased number of medical malpractice cases involving pressure ulcers litigated in recent years (Bennett, O'Sullivan, DeVito, & Remsburg, 2000).

Clinicians understand pressure ulcers are largely preventable and are cosdy to treat without evidence-based protocols. Pressure ulcers are associated with increased mortality rates and suffering, and may be a quality marker of the intra-organizational effectiveness of the nursing home (Berlowitz, Bezerra, Brandeis, Kader, & Anderson, 2000). Effective pressure ulcer prevention and treatment requires coordinated interdisciplinary practice, effective management of clinical information, nursing clinical acumen, and role clarity among team members (Levine & Totolos, 1994).

The clinical risk factors associated with pressure ulcer prevalence rates in nursing homes may be grouped into two broad categories. One includes risk factors related to the amount and duration of pressure such as dependence in activities of daily living, limited mobility and transferring, Parkinson's disease, and paraplegia. The second category relates to the general susceptibility of the skin and includes risk factors such as having a history of healed pressure ulcers, the diagnoses of dementia and diabetes, urinary and fecal incontinence, the use of a catheter, and being recently hospitalized, underweight, African American, male, and older (Spector, 1994; Spector & Fortinsky, 1998; Spector, Kapp, Tucker, & Sternberg, 1988). Table 1 summarizes risk factors associated with pressure ulcer prevalence.

METHODS

A portion of the 1996 OSCAR reports for 883 for-profit and not-for-profit California nursing homes certified for participation in Medicare and Medicaid was used in this study. Specialty facilities including government-owned facilities, longterm care psychiatric hospitals, hospitals serving the developmentally disabled, intermediate care facilities, inpatient hospices, and pediatric subacute facilities were excluded.

The original sample included 1,404 facility reports. These reports were examined for missing values. Two hundred thirty-eight facility reports were removed because data were not recorded for resident and staffing characteristics. Another 108 facilities reports were excluded after logic checks were performed.

Additional tests were applied to the sample because staffing data were used for two other aspects of this study. Frequencies and distributions of the full-time, part-time, and contract nursing staff hours and each category of nursing staff (registered nurse, LVN, and certified nursing assistant) were examined. Facility reports with values of staffing variables in the lower and upper 2% of each category were deleted because their values appeared erroneously extreme, resulting in 158 facility reports being removed.

Upon completion of data cleaning, 900 facility reports remained, allowing sufficient power (at least .96) to detect small effects of the predictors used in the multiple regression analysis. The analysis was examined for outlier observations of the prevalence rate of pressure ulcers, defined as a facility having a studentized residual greater than the absolute value of 2.5 and a Cook's distance greater than 4/N (Schlotzhauer & Littell, 1987). Seventeen facilities were removed, with the final sample consisting of 883 facilities.

Table

TABLE 2DEFINITIONS AND F-TACS FOR DEPENDENT AND INDEPENDENT VARIABLES

TABLE 2

DEFINITIONS AND F-TACS FOR DEPENDENT AND INDEPENDENT VARIABLES

Study Variables

Each component of the federal requirements for nursing homes, whether a definition of a term or a statement of a standard, is identified by a numeric code called the F-tag. The risk factors included in the model were operationally defined by the instructions given by CMS for Form 672. The risk factors included facility residents having:

* Pressure ulcers on admission.

* Difficulty feeding themselves.

* Difficulty with mobility and transfers.

* Incontinence of bladder or bowel.

* No rehabilitation potential.

* Dementia.

* Physical restraints.

All of these were among the risk factors identified in the literature and were variables included in the portion of the database that was used in the study. The F-tag labels and numbers are listed in Table 2. Because data on resident age, ethnicity, gender, and the diagnosis of diabetes are not components of the OSCAR database, these variables could not be included in the model.

Analysis

Multiple regression was used in the analysis of the risk-adjustment model. The full model included the variables listed in Table 2. The goals of the risk-adjustment analysis were to explain the variation in the prevalence rate of pressure ulcers attributable to clinical risk factors and to include only statistically significant variables in the model.

RESULTS

Facility Characteristics and Prevalence Rates of Clinical Characteristics

The study sample consisted of 652 (74%) for-profit and 231 (26%) notfor-profit nursing homes. Eightynine percent were freestanding nursing homes, and 58% of the facilities were located in urban counties.

All of California's 58 counties were represented, although one third of the facilities were located in Los Angeles County. Several rural counties had only one facility.

Facility size ranged from 1 1 to 269 beds. Forty-nine percent of the facilities had 60 to 120 beds. The two largest federal certification categories of nursing facilities included 451 (51%) nursing facilities for Medicaid, with a distinct part for skilled nursing facility care, and 323 (37%) facilities dually certified for Medicaid and Medicare (Table 3). The prevalence rates of clinical characteristics summarized in Table 4 are similar to rates observed in California facilities in 1996 by Harrington et al. (2001), who have analyzed OSCAR data extensively.

Risk-Adjustment Model

Table 5 shows the results of the regression analysis of the final riskadjustment model. Forty-four percent of the variation in pressure ulcer prevalence rates in facilities was explained by the model (R2 = 0.44). The proportion of residents having a pressure ulcer on admission was the most important risk factor among all others included in the model. The prevalence rates of incontinence and dementia were removed from the final model because of their statistical insignificance.

Removed Observations

Because of concerns regarding validity, the dependent variables from facility reports removed as outliers (17) and for logic errors (266), totaling 283 facility reports, were compared to those of the study sample. Wilcoxon's Rank Sum test was used because of the non-normal distribution of the variables in the removed group of observations. The test was used to determine whether the means of study variables of the two groups of sample observations and removed observations were significantly different at the 5% significance level.

Several of the means of the study variables in the removed observation group were significantly different in the two groups, including the variables of dementia, incontinence of bladder or bowel, no rehabilitation potential, and physical restraint use. However, the means of the prevalence rate of pressure ulcers, pressure ulcers on admission, difficulty feeding, and difficulty with mobility and transfers were not significantly different for both groups. These findings lend some support to the generalizability of the study findings, despite the number of removed observations.

DISCUSSION

This study illustrates how the OSCAR database was used to develop a risk-adjustment model for the prevalence rate of pressure ulcers in 883 California nursing homes. The risk-adjustment model explained some of the variation in the prevalence rate of pressure ulcers.

Clinical characteristics that were significantly associated with an increased pressure ulcer prevalence rate included the proportion of residents in each facility who had pressure ulcers on admission, had difficulty eating independently, had difficulty walking and transferring independently, and were restrained. Statistically significant risk factors in the model were similar to several risk factors included in other instruments used for risk assessment of pressure ulcers in nursing homes (Table 6).

Table

TABLE 3FACILITY CHARACTERISTICS (JV = 883)

TABLE 3

FACILITY CHARACTERISTICS (JV = 883)

Nursing home residents who did not receive rehabilitative services provided by, or under the direction of, a rehabilitation professional (i.e., physical therapist, occupational therapist, speech-language pathologist) were defined as having no rehabilitation potential in this study. These residents were associated with a reduced prevalence rate of pressure ulcers.

This finding is better understood by recalling that some residents who no longer receive rehabilitative services from professional service providers may be participating in restorative programs in the facility. Such nurse-run programs focus on grooming, eating, and mobility skill maintenance. Also, patients receiving rehabilitative services may represent a more acutely ill and less medically stable segment of each nursing home's resident census. This finding suggests the proportion of residents receiving rehabilitative services in a facility may reflect the proportion of short-stay to long-stay residents in a facility at the time the survey was conducted. The rehabilitative services rate may function as a proxy for case-mix.

The prevalence rate of residents having no rehabilitation potential, as defined in this study, was correlated with the prevalence rate of residents having dementia (Pearson correlation coefficient =. 37, p < .001). This may have resulted in the variable of "residents having no rehabilitation potential" overlapping with any unique contribution that the diagnosis of dementia would have contributed to explaining the variation in the prevalence rate of pressure ulcers. Perhaps the functional limitations that may accompany the diagnosis of dementia, rather than the diagnosis of dementia itself, better characterize the way in which dementia functions as a risk factor for the pressure ulcer prevalence rate in a nursing home.

Table

TABLE 4PREVALENCE RATES OF CLINICAL CHARACTERISTICS (/V= 883)

TABLE 4

PREVALENCE RATES OF CLINICAL CHARACTERISTICS (/V= 883)

Table

TABLE 5ORDINARY LEAST SQUARES REGRESSION MODEL PREDICTORS OF PREVALENCE RATE OF PRESSURE ULCERS (N = 883)

TABLE 5

ORDINARY LEAST SQUARES REGRESSION MODEL PREDICTORS OF PREVALENCE RATE OF PRESSURE ULCERS (N = 883)

Incontinence was not a significant predictor in this study, and its role as a risk factor in the prevalence rate of pressure ulcers is unclear (Berlowitz & Wilking, 1993). In this study, the distribution of incontinence was bimodal, with most facilities having residents with a higher prevalence of incontinence and a small grouping of faculties having residents with a lower prevalence rate of incontinence. The F-tag label for incontinence did not discriminate die levels of incontinence experienced by residents. Those who were occasionally or frequendy incontinent were categorized together. Consequently, most residents were characterized as being incontinent

LIMITATIONS

As with all administrative databases, this database presented both advantages and disadvantages to the researcher. The advantage of using the OSCAR database was daat it provided access to numerous non-specialized profit and not-for-profit California nursing homes surveyed during 19%. The disadvantages were the cross-sectional nature of the data, the restriction of working with variables diat were pre-determined, data inaccuracies, and missing entries.

The portion of the database made available for this study did not contain all of the approximately 700 variables within OSCAR. Consequendy, risk factors that, on hindsight might have been included in the risk-adjustment model could not be examined. This included the F-tags of the number of residents receiving hospice care benefit (F-1 19), radiation therapy (F-120), chemotherapy (F-121), dialysis (F-122), and tube feedings (F-129).

Although 238 (42%) of all observations removed from the study were removed because of missing data, it was unlikely that missing data reflected information not provided by nursing home staff or not collected by surveyors. The missing data were likely due to a problem with data entry (Mr. C. Hermann, personal communication, March 23, 1999).

IMPLICATIONS FOR RESEARCH AND CLINICAL MANAGEMENT

Research Implications

To study the prevalence rate of pressure ulcers in California nursing homes or other states using OSCAR data, access to the complete OSCAR database is recommended. Since this study was conducted, several vendors have begun to sell OSCAR data on individual facilities and states for single or multiple years. Vendors may be located on the Internet or in advertisements found in nursing home industry journals.

It is noteworthy that much of the variation in the prevalence rate of pressure ulcers in dus study was not explained by clinical characteristics of the nursing home residents. This may be attributable, in part, to random variation and limitations in the type of variables that were included in the model, for reasons already described.

The unexplained variation may be attributable to non-resident characteristics nSat warrant further study. Such characteristics include structural and process attributes of the nursing home. For example, structural attributes include nursing staffing levels, interdisciplinary staffing levels, the level of skin care education for nursing home staff, and the perceptions of the nursing staff regarding the work environment. Process attributes include the use of wound/skin teams, prevention and treatment protocols, specific types of beds, equipment, and skin care products.

Table

TABLE 6COMPARISON OF SIGNIFICANT RISK FACTORS IN FINAL RISKADJUSTMENT MODEL, OTHER PROTOCOLS, AND THE BRADEN SCALE

TABLE 6

COMPARISON OF SIGNIFICANT RISK FACTORS IN FINAL RISKADJUSTMENT MODEL, OTHER PROTOCOLS, AND THE BRADEN SCALE

Recommended study of these organizational attributes is consistent with other researchers' descriptions of a "facility effect" related to pressure ulcer incidence, prevention, and treatment (Brandeis, Ooi, Hossain, Morris, & Lipsitz, 1994; Ooi, Morris, Brandeis, Hossain, & Lipsitz, 1999). Better understanding of the relationships between facilities with low risk-adjusted prevalence rates and their organizational attributes is essential to the development of interventions that promote best practices (Rubenstein, Mittman, Yano, & Mulrow, 2000). Some of these organizational attributes are included in the OSCAR database.

However, OSCAR data need to be augmented by the use of facility-level questionnaires, medical record reviews, or observational data-collecting techniques to acquire data on clinical interventions and processes. The OSCAR database is useful as a data platform on which regional or state data of clinical interventions and processes could be included in studies that use the nursing facility as the unit of analysis.

Clinical /Management Implications

Risk assessment and an understanding of the concept of risk adjustment for prevalence rates of pressure ulcers are important components of a clinical management program for pressure ulcer prevention and treatment in a nursing facility. The risk factors that were significant predictors of the prevalence rate of pressure ulcers in this study are similar to those identified in risk-assessment instruments used by surveyors and clinicians. These include the risk factors identified in the Resident Assessment mstt^ment/Minimum Data Set (RAI/MDS) Resident Assessment Protocol (RAP) for pressure ulcers, the investigative protocol for pressure ulcers/ulcers used by surveyors to identify whether a pressure ulcer is avoidable or unavoidable, and the Braden Scale.

In the RAP for pressure ulcers, risk factors are called triggers and are summarized for clinicians. The intent is to prompt clinicians to perform a more focused clinical examination of the resident who has the triggers (Morris, Murphy, & Nonemaker, 1995). The Braden Scale is a widely used valid and reliable risk-assessment tool (Table 6). Whether the RAI/MDS is used alone, as Vap and Dunaye (2000) recommended, or in conjunction with the Braden Scale, risk assessment of pressure ulcers in nursing home residents continues to be an important component of an effective pressure ulcer prevention and treatment program.

Given that much of the variation in the prevalence rate of pressure ulcers was not explained by resident clinical characteristics, clinical managers may benefit from critically reviewing their pressure ulcer programs to identify aspects that may be modified to better promote best practices. Although the essential structural and process attributes of these programs are not fully understood, there is literature to assist clinical managers in critically examining the components of a model program.

For example, Levine and Totolos (1994) provided an excellent summary of a pressure ulcer management program that was successfully implemented in a large nursing facility. The program had educational, documentation, data flow, and administrative components for each member of the nursing staff and specific interdisciplinary team members. The AHCPR guidelines for pressure ulcer prevention and treatment provide an important tool for clinical managers, akhough much remains to be learned on how facilities successfully integrate die guidelines into dieir operations on a sustained basis (Berlowitz et ah, 2001; Xakellis, Frantz, Lewis, & Harvey, 2001).

An examination of the resources devoted to pressure ulcer prevention and treatment is recommended. Existing resources may be enhanced. For example, evaluation of the competence of nursing staff in clinical skills related to pressure ulcer prevention and treatment can be conducted, and deficiencies in skill performance may be enhanced through in-service education. Routine or periodic use of a skin care consultant and an evaluation of the efficacy of skin care products used may be considered.

Although study findings of the relationship between staffing levels and the prevalence rate of pressure ulcers are inconsistent (Dellefield, 2000), staffing threshold levels below which the quality of pressure ulcer prevention is reduced have been identified (Kramer & Fish, 2001). A clinical manager might observe staffing levels that correspond with recorded increases in the prevalence rate of pressure ulcers over time.

These are rudimentary examples of how clinical managers might find creative ways to better use the OSCAR data resources, applying them to performance assessment and improvement activities. Traditionally, nursing homes have had limited resources for data analysis focused on quality improvement. However, this situation is changing as clinical managers move toward using data-driven performance assessment and improvement programs that are an integral part of the facility's operations (Karanick, 2003). While data from the RAI/MDS are most commonly associated with these efforts, the OSCAR database is available as another data resource.

Finally, nurses in facility leadership positions such as nursing directors, staff development nurses, and corporate level nurses involved in administration and quality improvement activities need to create opportunities for enhancing nursing staff familiarity with the instructions and definitions of CMS Forms 671 and 672. Gerontological nurses need to understand that die OSCAR data collected during the annual survey are important. Gerontological nurses need to be knowledgeable about OSCAR, promoting accuracy and competence among nursing staff who contribute to it, and exploring ways to use the OSCAR database for performance improvement activities and research. In doing so, nurses help advance the science of nursing through the collection and use of accurate data to promote quality in nursing homes.

REFERENCES

  • Abt Associates Inc. (1996, December). Evaluation of the long term care survey process: Final report. Bethesda, MD: Author.
  • Bennett, R., O'Sullivan, J., DeVito, E., & Remsburg, R. (2000). The increasing medical malpractice risk related to pressure ulcers in the United States. Journal of the American Geriatrics Society, 48, 73-81.
  • Berlowitz, D., Ash, A., Brandéis, G., Brand, H., Halpern, J., & Moskowitz, M. (1996). Rating long-term care facilities on pressure ulcer development: Importance of case-mix adjustment. Annals of Internal Medicine, 124, 557-563.
  • Berlowitz, D., Bezerra, H., Brandeis, G., Kader, B., & Anderson, J. (2000). Are we improving the quality of nursing home care: The case of pressure ulcers. Journal of the American Geriatrics Society, 48(1), 59-62.
  • Berlowitz, D., & Wilking, S. (1993). Pressure ulcers in nursing homes. In L. Rubenstein & D. "Wieland (Eds.), Improving care in the nursing borne: Comprehensive reviews of clinical research. Newbury Park, CA: Sage.
  • Berlowitz, D., Young, G., Hickey, E., Anderson, J., Ash, A., Sc Moskowitz, M. (2001). Clinical practice guidelines in the nursing home. American Journal of Medical Quality, /6(6), 189-195.
  • Brandeis, G., Ooi, W., Hossain, M., Morris, J., & Lipsitz, L. (1994). A longitudinal study of risk factors associated with the formation of pressure ulcers in nursing homes. Journal of the American Geriatrics Society, 42, 388393.
  • Dellefield, M. (1999). Predictors of quality of care in California nursing homes. Dissertation Abstracts International (UMI No. 9943785), 60-08B, 3848.
  • Dellefield, M. (2000). A review of the literature: The relationship between nurse staffing in nursing homes and quality indicators. Journal of Gerontological Nursing, 26(6), 15-28.
  • Harrington, C, Woolhandler, S., Mullan, J., Carrillo, Hn & Himmelstein, D. (2001). Does investor ownership of nursing homes compromise the quality of care? American Journal of Public Health, 9/(9), 1452-1455.
  • Harrington, C-, Zimmerman, D., Karon, Sn Robinson, Jn & Beutel, P. (2000). Nursing home staffing and its relationship to deficiencies. Journal of Gerontology: Social Sciences, 555(5), S278-S287.
  • Hughes, C., Lapane, K., & Mor, V. (2000). Influence of facility characteristics on use of antipsychotic medications in nursing homes. Medical Care, 38(12), 1 164-1 173.
  • Karanick, J. (2003). Performance improvement: The new quality paradigm. Nursing Homes: Long Term Care Management, 52(1), 54-58.
  • Kramer, A., & Fish, R. (2001). The relationship between nurse staffing levels and the quality of nursing home care (pp. 1-26). In Appropriateness of minimum nurse staffing ratios in nursing homes phase II: Final report. Washington, DC: Health Care Financing Administration.
  • Levine, J., & Totolos, E. (1994). A quality-oriented approach to pressure ulcer management in a nursing facility. The Gerontologist, 34(3), 413-417.
  • Morris, J., Murphy, K., & Nonemaker, S. (1995). Long term care facility resident assessment instrument (RAI) user's manual version 2.0. Washington, DC: Health Standards and Quality Bureau, Health Care Financing Administration.
  • Mukamel, D. (1997). Risk-adjusted outcome measures and quality of care in nursing homes. Medical Care, 35(4), 367-385.
  • Ooi, W, Morris, J., Brandeis, G., Hossain, M., & Lipsitz, L. (1999). Nursing home characteristics and the development of pressure ulcers and disruptive behaviour. Age and Ageing, 28( 1), 45-52.
  • Rubenstein, L., Mittman, B., Yano, E., & Mulrow, C. (2000). From understanding health care provider behavior to improving healtJi care: The QUERI framework for quality improvement. Medical Care, 38(b), I129-I141.
  • Schlotzhauer, S., & Littell, R. (1987). SAS system for elementary statistical analysis. Cary, NC: SAS Institute Inc.
  • Spector, W. (1994). Correlates of pressure ulcers in nursing homes: Evidence from the National Medical Expenditure Survey. Journal of Investigative Dermatology, 102(6), 42S-45S.
  • Spector, W, & Fortinsky, R (1998). Pressure ulcer prevalence in Ohio nursing homes: Clinical and facility correlates. Journal of Aging and Health, /0(1), 62-80.
  • Spector, W, Kapp, M., Tucker, R., & Sternberg, J. (1988). Factors associated with the presence of decubitus ulcers at admission to nursing homes. The Gerontologist, 28, 830834.
  • State operations manual provider certification (Transmittal No. 10). (1999). Washington, DC: Health Care Financing Administration.
  • U.S. Department of Health and Human Services, Public Health Services, Agency for Health Care Policy and Research. (1992). Clinical practice guideline number 3. Pressure uUers in adults: Prediction and prevention. Rockville, MD: Agency for Health Care Policy and Research.
  • U.S. General Accounting Office. (2002a). Nursing homes: Quality of care more rented to staffing than spending (U.S. GAO-02431R). Washington, DC: Author.
  • U.S. General Accounting Office. (2002b). Nursing homes: Public reporting of quality indicators has merit, but national implementation is premature (U.S. GAO-03-187). Washington, DC: Author.
  • Vap, P., & Dunaye, T. (2000). Pressure ulcer risk assessment in long-term care nursing. Journal of Gerontological Nursing, 26(6), 37-45.
  • Wunderlich, G.S., & Kohler, P.O. (Eds.). (2001). Improving the quality of long-term care. Report by the Committee on Improving the Quality in Long-Term Care, Division of Health Care Services, Institute of Medicine. Washington, DC: National Academy of Sciences, National Academy Press.
  • Xakellis, G., Frantz, R., Lewis, A., & Harvey, P. (2001). Translating pressure ulcer guidelines into practice: It's harder than it sounds. Advances in Skin and Wound Care, 14(5), 249-256.

TABLE 1

RISK FACTORS ASSOCIATED WITN PRESSURE ULCER PREVALENCE RATES IN NURSING HOMES

TABLE 2

DEFINITIONS AND F-TACS FOR DEPENDENT AND INDEPENDENT VARIABLES

TABLE 3

FACILITY CHARACTERISTICS (JV = 883)

TABLE 4

PREVALENCE RATES OF CLINICAL CHARACTERISTICS (/V= 883)

TABLE 5

ORDINARY LEAST SQUARES REGRESSION MODEL PREDICTORS OF PREVALENCE RATE OF PRESSURE ULCERS (N = 883)

TABLE 6

COMPARISON OF SIGNIFICANT RISK FACTORS IN FINAL RISKADJUSTMENT MODEL, OTHER PROTOCOLS, AND THE BRADEN SCALE

10.3928/0098-9134-20041101-08

Sign up to receive

Journal E-contents