Long-term care facilities (LTC) are not exempt from the rising costs of providing quality health care. Profound changes in the health care market place are forcing LTC facilities to carefully scrutinize the services they deliver. High costs and frequently inadequate reimbursements are forcing nursing homes to evaluate their delivery systems to use valuable resources in the most efficient and effective manner. To financially survive, LTC administrators and clinical professionals need to be aware of areas where potential exists to reduce costs without compromising quality care.
Numerous studies have been conducted in regard to the high costs associated with the treatment of pressure ulcers. The decubitus ulcer, or bedsore, is a clinical term often used interchangeably to describe an area of impaired skin integrity in which cellular necrosis has occurred (Kelley & Mobily, 1991). It develops when soft tissue is pressed between a bony prominence and a firm surface with the pressure causing capillaries to collapse. This, in turn, interrupts the tissue's supply of oxygen and nutrients (Colburn, 1990). If the capillaries remain closed, the surrounding tissue dies, sometimes within 2 hours (Colburn, 1990).
Pressure ulcers affect more than a million people each year. In terms of cost estimates for their treatment, the financial burden to LTC facilities is known to be quite significant. Estimates of total institutional costs for treatment of a person with pressure ulcers in a LTC facility ranged from $4,255 to $23,301 (Frantz, 1989) and between $14,000 and $25,000 (Valdez, 1990). Anticipating which residents are at risk for developing this problem and focusing staff attention on prevention, early detection, and intervention rather than costly and painful treatment, could result in cost savings and more effective use of scarce resources.
Literature and risk assessment protocols or tools regarding the prevention and treatment of pressure ulcers have saturated the market. In the midst of all the information and research, however, the effectiveness of using a risk assessment tool such as the Braden Scale (BS) in conjunction with the federally mandated Minimum Data Set (MDS) seems uncertain. The MDS is a component of the federally mandated Resident Assessment Instrument (RAI) (Morris, Murphy, & Nonemaker, 1995), and includes recognition of the presence or absence of pressure ulcers and factors that identify residents to be at risk for developing pressure ulcers. Because of the time element involved in completing the MDS assessment, it becomes important to know whether an additional assessment tool designed specifically to identify people at risk for developing pressure ulcers will help decrease their costly incidence. The combination of shorter lengths of stay in acute care facilities, increasing rates of acute care ^hospitalizations, the shift in location of death from the hospital to the nursing home, and higher numbers of required nursing procedures confirm that more seriously ill patients are receiving care in nursing home facilities (Langer, Drinka, & Voeks, 1991). As the acuity level increases, factors that contribute to pressure ulcer development will also potentially increase. Other outside forces such as reimbursement, staff shortages, expense of therapeutic beds, and aging of the patient population are driving the need to assess residents more precisely and in a timely manner.
Change in the care standards for the prevention and treatment of pressure ulcers has added a new legal dimension to LTC. Long-term care providers can legally be held liable if a resident in a nursing home develops pressure ulcers and is judged as receiving substandard care associated with pressure ulcers and probable neglectful attention (Kohnle, 1996). As a result, the LTC industry is being forced by a combination of governmental and legal pressures to monitor quite carefully all residents at risk for developing pressure ulcers.
Some providers have reacted to these growing regulatory requirements for more patient care assessment by creating their own specialized tools or using some already in place, such as the BS and the MDS. This combination practice is not a federal requirement and may not be desirable because of unnecessary use of human resources. Applying multiple assessment tools to help discover the same problem is not economically sensible (Morris, et al., 1995).
PRESSURE ULCER PREVALENCE AND GERIATRIC NURSING ISSUES
Nursing homes respond to the needs of elderly individuals and others requiring prolongea institutionalization for functional limitations exacerbated by chronic conditions by providing appropriate services according to patient needs. Expenses for essential care can vary significantly, as does payer reimbursement. Failure to identify residents at risk for developing pressure ulcers could result in even greater expenses for treatment protocols as well as increased demand on professional resources.
Pressure ulcers have been a problem since Hippocrates practiced ancient medicine. They have been found on Egyptian mummies and persist among bedridden patients despite remarkable technological and scientific advances (Colburn, 1990). Aggressive intervention is needed in the early stages of a pressure ulcer condition to reverse severe tissue damage. Otherwise, it will evolve into a damaged skin wound that can be very costly and difficult to heal. Among patients in skilled-care facilities and nursing home facilities, the prevalence of pressure ulcers was found to be 2.4% to 23% (Smith, 1994) (nursing home facilities may receive additional certification to provide skilled services payable by Medicare A). For those who do develop pressure ulcers after admission to these institutions, most occur within the first 2 weeks, frequently because the patient is still recovering from an acute bout of illness (Smith, 1995). Therefore, it is imperative for patients at risk to be detected early at the time of admission and that preventive interventions be initiated as quickly as possible.
The cost of pressure ulcer repair and healing depends on the extent of the condition. A chronic, full thickness wound can take as long as 6 weeks to 6 months to respond fully, or may not heal at all (Van Rijswijk & Cuzzell, 1991). Geriatric caregivers may end up absorbing some, all, or none of these costs, depending on the variance in reimbursement entitlements.
Not all residents of LTC facilities are at risk for developing pressure ulcers, and numerous studies have identified several factors that tend to place residents at greater risk. Because pressure ulcer development depends on the length of time surface pressure is applied, immobility is a major risk factor. Malnutrition has been linked to the development of pressure ulcers. Lower dietary protein intake and inability to feed oneself have been found to be reliable predictors of pressure ulcer development (Kelly & Mobily, 1991; Kresevic & Naylor, 1995; Smith, 1995). Urinary and fecal incontinence also have been considered factors contributing to prediction of pressure ulcers. The independent contribution of incontinence in predicting pressure ulcers, however, has not been shown consistently (Bergstrom & Braden, 1992; Berlowitz & Wilking, 1989; Brandeis, Ooi, Hossain, Morris, & Lipsitz, 1994). Urinary incontinence was not found to be a significant risk factor in these studies except as a component of skin moisture in the BS. Fecal incontinence was significant in one of three studies performed (Brandeis, et al., 1994), but not independently observed in others (Bergstrom & Braden, 1992; Berlowitz & Wilking, 1989). Older individuals who experience a decrease in their level of consciousness are at risk for pressure ulcers because of an altered ability to recognize and respond to the pain or a decreased ability to respond to the environment. Early identification of residents with risk factors linked to the development of pressure ulcers has the potential to decrease serious consequences and the time and cost to treat them.
MINIMUM DATA SET RESIDENT ASSESSMENT PROTOCOLS INSTRUMENTS
To improve the quality of care provided, the Nursing Home Reform Act of the Omnibus Budget Reconciliation Act of 1987 (OBRA '87) mandated the most comprehensive legislative requirements ever to affect LTC facilities for their 1.5 million residents (Marek et al., 1996). The intent of OBRA '87 was to improve nursing home quality by establishing a single set of certification conditions for all nursing homes. The RAI consists of three parts:
* The MDS.
* The Resident Assessment Protocols (RAPs).
* The Utilization Guidelines.
The MDS consists of federally required core elements and common definitions for resident assessment. It incorporates measures of physical health, functional status, psychosocial well being, dietary status, comprehension, vision, hearing, communication skills, activity preferences, potential for self-improvement, and indicators of quality of life.
COMPONENTS OF RISK FACTORS
The MDS contains triggers that cue users to areas requiring more indepth assessment. The triggers identify a single or a combination of MDS item responses specific to a resident that alert the assessor to the resident's possible problems, needs, or strengths (Morris, et al., 1995). Essentially, the triggers draw attention to conditions necessary for the interdisciplinary team members to consider in planning care decisions. When the resident's status on a particular MDS item(s) matches one of the triggers for a RAPs, the specific RAP is triggered and a review is required using the RAP Guidelines.
There are 18 problem-oriented RAPs, each of which includes MDSbased trigger conditions that signal the need for additional assessment and review. The MDS and RAPs provide the health care worker with information to assist in the care planning process. The MDS ensures that staff have access to specific assessment data and the RAPs provide criteria that trigger review of possible problem conditions to ensure that staff identify problems in a consistent and systematic manner. The following items on the MDS will trigger the potential for risk of developing a pressure ulcer: bed mobility, bedfast, bowel incontinence, peripheral vascular disease, pressure ulcer, previous pressure ulcer, impaired tactical sense, and trunk restraint. These risk factors, along with BS risk factors, are illustrated in Table 1.
The MDS and RAPs provide a substantial amount of patient care information. The average time to complete the MDS in one study was 1 hour, 20 minutes (Morris, et al., 1995). Completion of the initial assessment is required within 14 days of admission. Annual reassessments must be completed within 12 months of the most recent full assessment. A short-form quarterly assessment must be completed no less frequendy than once every 3 months and a long form must be completed within 14 days following a significant change in resident status.
RISK ASSESSMENT TOOLS
The initial step in any prevention strategy is the identification of those patients at risk. Assessment tools have gained increasing popularity as a means to implement efficient risk identification. Assessment tools should address an important problem; identify when preventive measures might be most useful; and be simple, convenient, reliable, and cost effective (Goodridge, 1993).
A risk assessment tool commonly used in LTC facilities is the BS. This tool quantifies a number of risk factors using a rating scale with summative scores that serve as the basis for risk prediction. The BS addresses sensory perception, moisture, activity, mobility, nutrition, and friction or shear. The rating scale of 1 (least favorable) to 4 (most favorable) is used for each subscale and these combine for a maximum possible score of 23. A low score indicates high risk for developing pressure ulcers and a high score indicates low risk. The BS scores are continuous - those of 16 or 17 indicate mild risk, while 12 or lower indicate high risk (Bergstrom & Braden, 1992).
The BS requires approximately 10 minutes to administer initially. Reassessments with the BS require approximately 1 minute to perform when conducted by a health care worker who delivers nursing care on a regular basis to the same resident (Burd, Langemo, Olson, Hanson, Hunter, & Sauvage, 1992).
For a risk assessment tool to be effective and efficient, it must provide information that the federally mandated MDS cannot. Accuracy in identifying residents at risk for pressure ulcer development would allow staff to focus on this high-risk group in regard to implementing preventive interventions. As a result, both time and dollars would be saved.
The study population included the MDS/RAPs and the BS risk assessment tool records of all residents living in eight randomly selected Medicare/Medicaid-certified LTC facilities in the northeast section of Iowa. Other than being Medicare/Medicaid certified, the only other criterion for facility selection was current or previous use of a risk assessment tool to identify residents at risk for developing pressure ulcers for at least the past year (from beginning of study).
A two-item questionnaire was sent to each of the 91 LTC facilities located in northeast Iowa asking whether the facility used a pressure ulcer risk assessment tool; and if so, which one. Random selection of facilities for participation in this study was then determined by assigning each facility a number obtained from a Table of Random Numbers (Babbie, 1992). Starting from the top of the order, facilities were sent a letter requesting data collection at that specific facility. In the event of refusal from the facility for data collection, the next one on the list was contacted. One facility contacted refused to participate in this study.
Data from the MDS/RAPs record and the risk assessment tool record were obtained from the resident's permanent chart. Data from the RAP Summary sheet indicated whether the person was identified to be at risk for developing pressure ulcers. A score of 16 or lower on the BS indicated the person to be at risk for developing pressure ulcers. The results of both represented a binomial variable of either yes or no. If a review of either record indicated that the presence or absence of these two data elements were neither documented nor clearly stated, that case was excluded from the analysis of data.
A two-sample t test was conducted to compare the difference in the two mean values calculated from two different sources of data for a given value. Results were considered significant 2Xp < .05. Each observation was sampled independently and each observation came from a population with normal distribution. Validity was established by the ability of the MDS and the BS to predict the incidence of pressure ulcers.
BRADEN SCALE SCORES OF 16 OR BELOW
BASED ON MDS TRIGGERS
Data were also collected on the risk factors- that were triggered or identified for those residents who did develop pressure ulcers. These data were listed in rank order and either supported or disputed previous research in this area, and also demonstrated similar and dissimilar characteristics between the two assessment tools.
A human subjects review committee approval was not applicable because this research involved data collection without identifying specific residents. Written consent from the administrator of each facility where data were collected was obtained prior to the review.
Of the 91 surveys sent out, 81 were returned. Two facilities were excluded (letters were returned to sender for unknown reasons) from the original 91, which reduced the number used for statistical purposes to 89, resulting in a return rate of 91%.
Of the 81 surveys returned, 88.9% used a risk assessment tool (n = 72): 77.8% used the BS (n = 63); 6.2% used a Briggs form (Briggs Corporation, Des Moines, IA) (n = 5); and 4.9% used a facility specific tool (n = 4). Nine facilities (11.1%) responded that they did not use a pressure ulcer risk assessment tool. The results of this data collection approach are presented in Figure 1.
The BS was used by 87.5% of the facilities, the Briggs form was used by 6.9% of the facilities, and a facility specific tool was used by 5.6% of the facilities. These data are expressed in Figure 2.
The eight facilities in northeast Iowa where data were collected ranged in number of licensed beds from 66 to 110. The number of resident charts reviewed totaled 561. Of the 561 charts reviewed, 6 were eliminated because of incomplete or insufficient data relative to the BS or the MDS, resulting in complete data collection from a total of 555 resident charts. Stage I-IV pressure ulcers were considered for this study with staging consistent with the recommendations of the National Pressure Ulcer Advisory Panel (Walker, 1996). Pressure ulcers not staged because of eschar were also included in the study.
SUMMARY OF MPS DATA
SUMMARY OF BRADEN SCALE DATA
A two-sample t test with 95% confidence was conducted to compare the difference in the two mean values of the MDS and the BS (i = 3.77; df = 97; ? < .003) and (t = 3.77; df = 97; ? < 001). Each observation was sampled independently and each observation came from a population with a normal distribution. This statistical application demonstrated a significant difference between the predictive results of the MDS and the BS.
Chi-square application indicated both the MDS and the BS were significantly correct in identifying risk factors triggering risk for pressure ulcers: (?2 = 43.68; df=i;p< .0001) and (?2 = 52.47; df=\;p< .0001).
Sensitivity and specificity resulted in the following: MDS sensitivity 20%, specificity 98%; and BS sensitivity 27%, specificity 95%. (Sensitivity refers to the percent of residents who developed pressure ulcers from those predicted to do so, and specificity refers to the percent of those who did not develop pressure ulcers and were predicted not to be at risk.)
BRADEN SCALE TRIGGER PERSPECTIVE
Braden scales of 16 or less (mild risk) totaled 172, or 31%, of all charts reviewed. On the MDS, risk for pressure ulcers was triggered on 166 charts of the 172 with a BS score of 16 or less (96%). The number in this grouping with a BS score of 16 or less that developed or had a history of a pressure ulcer within the past 90 days totaled 46, or 27% of BS scores of 16 or lower. The MDS triggered 45 in this grouping, or 98% of BS scores of 16 or lower that indicated either development of a pressure ulcer or a history of one within the past 90 days. The BS did not identify 383 residents to be at mild risk for pressure ulcers. However, 20 pressure ulcers did develop, or there was a history of such within the past 90 days. In comparison, the MDS did not identify 21 residents. Table 2 illustrates these data.
Further data breakdown on the BS resulting in a score of 12 or lower revealed 43 to be at high risk for developing pressure ulcers or 8% of the 555 charts where data were collected. All 43 also triggered risk for pressure ulcers on the MDS. Of the 43 with a BS score of 12 or lower, 1 1 either developed a pressure ulcer, or had a history of such within the past 90 days (26%). All of these were also triggered on the MDS.
MDS TRIGGER PERSPECTIVE
In contrast to the 1 72 BS scores of 16 or lower, which represented mild risk for pressure ulcers, the MDS identified 311 residents to be at risk for pressure ulcers. This represents 56% of all MDSs reviewed (n = 555) compared to 31% of BS scores of 16 or lower (n = 555). Of the 311 predicted to be at risk for pressure ulcers through the MDS, 166 were identified to be at risk with a BS score of 16 or lower. Out of the 311 MDS triggers that cued residents at risk for pressure ulcers, 62 developed them or there was a pressure ulcer history within the past 90 days, representing 20% of this group. Four additional pressure ulcers developed that the MDS did not predict. The BS score of 16 or lower was unable to predict 3 of these pressure ulcers. However, it did predict 1 that the MDS failed to predict. The results of these data are presented in Table 3.
Of the 62 residents who actually developed a pressure ulcer, or had a history of pressure ulcers within the past 90 days, 45 were identified to be at risk for ulcer development. They had a BS score of 16 or lower, or 73% of what the MDS predicted. Of the 62 pressure ulcers that developed after being predicted by the MDS triggers, 17 were not identified by a BS score of 16 or lower (27% comparatively). This shows a four-fold (425%) inability of BS to predict risk and subsequent development of pressure ulcers.
Overall, the MDS triggers predicted 62 of the total 66 pressure ulcers, or history of such within the past 90 days (94%). The BS score of 16 or lower identified 46 to be at risk for pressure ulcers (70%). This count of 46 includes the one correct prediction of the BS that the MDS did not predict.
The MDS triggered risk for pressure ulcers in a total of 311; 249 of these did not develop pressure ulcers (80%) and 62 (20%) did. The MDS did not predict a pressure ulcer risk factor in 244 records (44%). Four of these resulted in pressure ulcer development that the MDS failed to predict (2%) and 240 (98%) did not develop pressure ulcers. These data indicate major statistical differences among the observations summarized in Table 4 (?2 = 43.68; df=l; p <. 0001).
In contrast, a BS score of 16 or lower identified 172 residents to be at risk for pressure ulcers, with 46 (27%) developing them and 126 (73%) non-occurring. Of the 383 not identified to be at risk, 20 developed pressure ulcers (5%) and 363 (95%) did not. These data indicate high statistical differences among the observations as summarized in Table 5 (?2=52.47; df= 1;p< .0001).
Even though the MDS and the BS were both correct in identifying residents at risk for developing pressure ulcers, this investigation found a significant difference between the predictive value of the MDS and the BS as demonstrated by the two-sample t test statistical application. The BS ranked slightly higher than the MDS in regard to sensitivity, while the MDS ranked slightly higher than the BS in regard to specificity.
A BS score of 16 or lower identified 172 residents of 555 to be at mild risk for developing pressure ulcers (31%), and the MDS triggered risk for 311 of the 555 (56%). The BS correctly predicted 46 ulcers of the 172 identified to be at pressure ulcer risk. However, the BS failed to identify 20 resident to be at risk who did develop pressure ulcers, or had a medical history of them within the past 90 days (5%). The difference in the predictive ability of the MDS and the BS is that the MDS tends to overpredict residents at risk for pressure ulcer development by a ratio of nearly 2 to 1. Either a "yes" or "no" information response triggers risk factors on the MDS. The BS factors in severity by assigning a choice of 4 numbers that must accurately describe the resident's condition.
The MDS predicted the occurrence of pressure ulcers more accurately than the BS by a ratio of 1.3 to 1. But again, the MDS triggered risk for pressure ulcers in 56% of all residents (predicting 94% of all pressure ulcers), while the BS identified 31% of all residents to be at risk (predicting 70% of all pressure ulcers). These data would suggest that the greater the number of residents triggered or identified to be at risk for pressure ulcer development, the greater the chances of this group containing residents who are likely to develop pressure ulcers.
The BS based on a score of 16 or lower failed to identify 20 residents at risk who actually did develop pressure ulcers, while the MDS failed to identify 4 such cases. The MDS triggered risk for pressure ulcers in nearly the same number of residents as did the BS scores of 16 or lower (166 and 172, respectively). There was only a difference of 1 between the MDS and BS for residents who actually developed pressure ulcers or had such histories in the past 90 days along with a difference of one case that was not triggered at risk.
A BS score of 12 or lower (indicating high risk) identified 43 cases to be at risk, with 1 1 pressure ulcers actually developing or being present in the past 90 days. All of these were also triggered at risk for pressure ulcers by the MDS. Even representing high risk for pressure ulcers on the BS, it was only able to predict 17% on a score of 12 or lower. These data suggest that the predictive value of pressure ulcer development is more accurate on the BS at 16 or lower (indicating mild risk) as opposed to 12 or lower (indicating high risk).
These data would imply that preventive measures incorporated by facilities using the MDS as the sole means of identifying residents at risk for pressure ulcer development would include a higher number than the BS. This would result in more cost to the facility for pressure ulcer prevention. Treatment of pressure ulcers was found to be 2.5 times more costly than prevention. Therefore, it still would have been less expensive to incorporate preventive measures for the 4 pressure ulcers that the MDS failed to predict compared to the 20 that the BS failed to predict.
Conversations with staff at all eight LTC facilities where data were collected indicated the BS was completed at the same time as the MDS. As noted earlier, the MDS must be completed within 14 days of admission. Taking this into consideration, the BS may be of more value when performed immediately upon admission. This would alert staff to the residents who are at greater pressure ulcer risk, thus allowing preventive measures to be introduced right away. A study based on this, and comparing facilities that use this strategy compared to facilities that do not use a risk assessment tool immediately upon admission, would provide additional information regarding the cost effectiveness of risk assessment tools in conjunction with the MDS. A number of factors were not included in this investigation that could also influence the results of future studies. These include staff ratios, acuity of care, emphasis of management on prevention, education and knowledge of staff regarding causes and prevention of pressure ulcers, and determining interrater reliability for both the MDS and BS. The best assessment tools are of no value unless the results are used to improve timely caregiving for residents at risk for pressure ulcers.
More conclusive data derived from this study support previous research regarding factors that contribute to pressure ulcers. The three primary factors that put residents at risk for pressure ulcers on the BS with scores of 16 and lower and 12 and lower were:
* Limitations in activity.
* Friction or shear.
Two of these key factors were similarly identified in a Kingston, Ontario study (McNaughton & Brazil, 1995). Compromised activity and mobility levels also support research conducted by others (Litchford, 1993; Smith, 1995). Although last in rank order, compromised nutrition places residents at risk for developing pressure ulcers, which, again, support several previous studies (Kresevic & Naylor, 1995; Smith, 1995).
The two most significant factors that triggered the MDS for pressure ulcer risk (in conjunction with both a BS of 16 or lower and 12 or lower) were bed mobility and bowel incontinence. Urinary incontinence did not trigger a resident to be at risk as did the BS. The third and fourth most frequent factors on the MDS that triggered risk for pressure ulcers in conjunction with a BS of 16 or lower were: pressure ulcer present or experienced within the past 90 days and skin desensitization to pain or pressure, respectively. These two factors were reversed in BS scores of 12 or lower. The last three MDS factors that triggered risk were peripheral vascular disease, daily trunk restraint, and bedfast. These factors represented only a slight variation between baseline BS scores of 16 or lower and 12 or lower. The significance of ranking these trigger factors was not only to make a comparison between the MDS and BS, but to also identify the dissimilar characteristics between the two assessment tools.
Data indicate the MDS was able to identify a greater number of pressure ulcers from those identified to be at risk as opposed to the BS. However, besides compromised mobility and activity, both tools measure different risk factors. The MDS fails to address friction and shear and urinary incontinence while the BS fails to address pressure ulcers (present or experienced in the past 90 days), skin desensitized to pain or pressure, daily trunk restraint, or peripheral vascular disease as risk factors for pressure ulcers.
Even though the statistical examination demonstrated a significant difference between the predictive value of the MDS and the BS, the results were the opposite of what the investigators expected. As a result, further research is warranted on the efficient and effective use of risk assessment tools in conjunction with the federally mandated MDS. More study is also needed to determine whether the 14day timeframe for completion of the MDS (and in this research, completion of the BS at the same time), represents a sufficiently critical interval where early identification of those residents at risk for pressure ulcer development would result in a lower incidence rate. A study such as this would help to determine whether costs associated with treatment could be further reduced by using pressure ulcer risk assessment tools more efficiently and effectively.
Research in the area of adding those risk factors found to be significant on the BS to the MDS (but currently excluded) could result in reducing or eliminating the number of additional, more costly, and timeconsuming risk assessment tools. But for this to occur, it would be necessary to determine whether the 14-day timeframe to complete the MDS represents a significant negative factor in regard to early pressure ulcer risk detection in certain residents.
The results of this investigation may spur further interest in similar studies that explore whether risk assessment tools (other than pressure ulcer tools) might result in duplication of mandated forms or assessments. Scarce geriatric care resources should always be managed as efficiently and effectively as possible.
- Babbie, E. (1992). The practice of social research (6th ed.). Belmont, CA: Wads worth.
- Bergstrom, N., & Braden, B. (1992). A prospective study of pressure sore risk among institutionalized elderly. Journal of the American Geriatrics Society, 40, 747758.
- Berlowitz, D., & Wilking, S. (1989). Risk factors for pressure sores. A comparison of cross-sectional and cohort-derived data. Journal of the American Geriatrics Society, 37, 1043-1050.
- Brandeis, G., Ooi, E., 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, 388-393.
- Burd, C, Langemo, D., Olson, B., Hanson, D., Hunter, S., & Sauvage, T. (1992). Skin problems: Epidemiology of pressure ulcers in a skilled care facility. Journal of Gerontological Nursing, 18(9), 30-37.
- Colburn, L. (1990). Preventing pressure. Nursing 90(December), 60-63.
- Frantz, R. (1989). Pressure ulcer costs in long term care. Decubitus, 2, 56.
- Goodridge, D. (1993). Pressure ulcer risk assessment tools: What's new for gerontological nurses. Journal of Gerontological Nursing, /9(1), 23-27.
- Kelley, L., &: Mobily, P. (1991). Iatrogenesis in the elderly: Impaired skin integrity. Journal of Gerontological Nursing, 17(9), 24-28.
- Kohnle, R. (1996). Pressure Sores, the problem. DermaSafe's study of pressure sores, October, 1-2.
- Kresevic, D., & Naylor, M. (1995). Preventing pressure ulcers through use of protocols in a mentored nursing model. Geriatric Nursing, 16(5), 225-229.
- Langer, E., Drinka, P., & Voeks, S. (1991). Readmissions and acuity in the nursing home: How will the nursing homes manage? Journal of Gerontological Nursing, 17(7), 15-19.
- Litchford, M. (1993). Respecting nutrition: Ulcer link crucial in treatment success. Provider, 63.
- Marek, K.D., Rantz, M.J., Fagin, CM., Krejci, J.W. (1996). OBRA *87: Has it resulted in better quality of care? Journal of Gerontological Nursing, 22(10), 28-36.
- McNaughton, V., & Brazil, K. (1995) Wound and skin team impact on pressure ulcer in chronic care, Journal of Gerontological Nursing, 21(2) 45-47.
- Morris, J., Murphy, K., & Nonemaker, S. (1995). Long term care resident assessment instrument user's manual (Version 2.0). Washington, DC: Health Care Finance Administration^- 19, 4-3, 301.
- Smith, D. (1995). Pressure ulcers in the nursing home. Annals of Internal Mediane, 123(6), 337-338.
- Smith, P. (1994). Infection control in long-term care facilities (2nd ed.). New York: Delmar.
- Valdez, L. (1990). A cost-effective alternative for relieving pressure. Ostomy/Wound Management, Sept/Oct, 30-34.
- Van Rijswijk, L., & Cuzzell, J. (1991). Managing full-thickness wounds. American Journal of Nursing, June, 18-22.
- Walker, D. (1996). Choosing the correct wound dressing. American Journal of Nursing, September 9, 38.
COMPONENTS OF RISK FACTORS
BRADEN SCALE SCORES OF 16 OR BELOW
BASED ON MDS TRIGGERS
SUMMARY OF MPS DATA
SUMMARY OF BRADEN SCALE DATA