This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation

Determinants and Effects of Nurse Staffing Intensity and Skill Mix in Residential Care/Assisted Living Settings

Correspondence: Address correspondence to Sally C. Stearns, PhD, Department of Health Policy and Administration, University of North Carolina at Chapel Hill, CB #7411, Chapel Hill, NC 27599-7411. E-mail: sally_stearns{at}unc.edu

	Abstract

TOP Abstract Methods Results Discussion References

 
Purpose: Residential care/assisted living facilities have become an alternative to nursing homes for many individuals, yet little information exists about staffing in these settings and the effect of staffing. This study analyzed the intensity and skill mix of nursing staff using data from a four-state study, and their relationship to outcomes. Design and Methods: We obtained longitudinal data for 1,894 residents of 170 residential care/assisted living facilities participating in the Collaborative Studies of Long-Term Care. Descriptive statistics assessed the levels of direct care staff (registered nurse, licensed practical nurse, personal care aide). Regression analyses evaluated the relationship between two staffing measures (intensity measured as care hours per resident and skill mix measured as the percentage of total care hours by licensed nurses), facility characteristics, and four health outcomes (mortality, nursing home transfer, hospitalization, and incident morbidity).  Results: Care hours per resident decreased with facility size (economies of scale) only for very small facilities and increased with dementia prevalence (case-mix effect). Licensed staff accounted for a greater proportion of total hours in nonprofit settings. Health outcomes did not vary by total care hours per resident, but hospitalization rates were significantly lower in facilities with higher proportions of skilled staff hours; this effect was stronger as dementia case mix increased. Implications: Current staffing levels for the outcomes analyzed meet most residents' needs. Reduced hospitalization in relation to greater use of licensed staff suggests that increased use of these workers might result in reductions in acute care expenditures.

Key Words: Resident outcomes • Licensed staff • Care hours

Nurse and personal care aide (PCA) staffing levels are indicators of quality in nursing homes, where minimum staffing ratios are required for participation in the Medicare or Medicaid programs (Harrington et al., 2000; Institute of Medicine, 1996). Over the past decade, states have paid increasing attention to mandating ratios for RC/AL as well (Mollica & Johnson-Lamarche, 2005), yet little evidence exists about facility and resident factors associated with staffing levels in RC/AL settings (Park, Zimmerman, Sloane, Gruber-Baldini, & Eckert, 2006) or about the effects of staffing on resident outcomes. To help address these issues, this article provides new evidence in three ways. First, it provides a descriptive analysis of staffing intensity and skill mix across a large and diverse sample of facilities. Second, it describes the results of a facility-level analysis of the relationship between facility characteristics and direct-care staffing intensity and skill mix, with a focus on economies of scale and the cost implications of varying staffing levels. Third, it uses quarterly observations over 1 year of resident health outcomes to examine whether staff intensity and skill mix are associated with mortality, nursing home transfer, morbidity, or hospitalization. We conducted the analyses using facility- and resident-level data from the Collaborative Studies of Long-Term Care (CS-LTC), a four-state study of RC/AL facilities.

Before presenting the study methods and results, we first turn attention to three areas of literature relevant to this study: the case mix and functional abilities of RC/AL residents, prior studies of staffing intensity and quality in nursing homes, and existing studies of staffing and health outcomes in RC/AL.

Studies of Case Mix Within RC/AL Facilities
Prior studies have shown that residents in RC/AL are, on average, less physically dependent and cognitively impaired than persons in nursing homes (Hawes, Rose, & Phillips, 1999; Manard & Cameron, 1997; Zimmerman et al., 2003). These differences are in part attributable to the fact that some states or facilities require that RC/AL residents have stable health conditions, not need 24-hr nursing care, or not be actively dying. In addition to having differences in functional impairment and case mix complexity, RC/AL residents tend to be slightly younger on average than nursing home residents. These factors result in a modestly healthier population in RC/AL facilities than in nursing homes (Zimmerman et al., 2003).

The prevalence of chronic disease, functional impairment, and dementia in RC/AL is, however, substantial (Morgan et al., 2001), and some studies have questioned whether differences in case complexity are an artifact of less complete identification of RC/AL resident impairment relative to that of nursing home residents (Kane, Kane, Illston, Nyman, & Finch, 1991). Hawes and colleagues (1999) found cognitive and physical impairments present in residents in a majority of the facilities. Zimmerman and colleagues (2003) also found notable cognitive impairment; for example, 23% to 43% of residents had moderate or severe dementia, depending on the type of facility in which they resided. They also found variation in resident functional ability across facilities; for example, impairment in bathing (52%–63%), personal hygiene (23%–45%), and dressing (23%–43%) were common, whereas fewer residents had problems with eating, bed mobility, and locomotion, and 33% to 44% of residents had no ADL dependencies (Zimmerman et al., 2005). Common medical conditions reported by residents included arthritis, rheumatism, and related conditions (45%–52%), high blood pressure (42%–50%), and eye disease (e.g., glaucoma, cataracts, and macular degeneration; 35%–46%). Although most evidence indicates that residents in RC/AL are less impaired than residents in nursing homes, the aging in place of the current RC/AL population will lead to greater average levels of impairment over time.

Studies of Staffing in Nursing Homes
Although RC/AL living facilities differ in important ways from nursing homes, an understanding of staffing intensity and skill mix in nursing homes may provide insights into possible relationships in RC/AL facilities. Researchers consider nursing home staffing intensity and skill mix to be associated with facility characteristics (such as size, ownership, and reimbursement policy) and resident characteristics (such as demographics, payment source, and case mix) (Institute of Medicine, 1996). Consequently, both the Medicare program and state Medicaid programs typically provide some regulatory minimum staffing requirements (Harrington et al., 2000). Not surprisingly, higher staffing levels in terms of registered nurse (RN) staff and total time spent in resident care have been positively associated with a variety of process and outcome indicators of quality of care (such as physical restraint and antipsychotic drug use) as well as health outcomes (such as pressure sore or urinary tract infection rates). For example, Weech-Maldonado, Meret-Hanke, Neff, and Mor (2004) used data from 1,287 nursing homes in five states to show that having a higher RN skill mix was associated with better outcomes (in terms of pressure ulcers and cognitive functioning) and better processes of delivering care (such as lower use of restraints). In another study, highly trained staff and more direct care time was associated with improved resident assessment, monitoring, and care planning; efficient nurse–physician communication; and multidisciplinary team discussions associated with high quality of care (Schmidt & Svarstad, 2002). Svarstad and colleagues also found that use of antipsychotic drugs was negatively associated with the nurse-to-resident staffing ratio while controlling for case mix and treatment culture (Svarstad & Mount, 2001; Svarstad, Mount, & Bigelow, 2001).

The type or mix of nursing staff, however, may be more important than the total number of nursing staff hours per resident in affecting outcomes or processes of care (Institute of Medicine, 1996). For example, facilities with lower levels of restraint use employed more full-time equivalent RNs but fewer full-time equivalent nurse aides and licensed practical nurses (LPNs) per resident (Castle & Fogel, 1998; Kolanowski, Hurwitz, Taylor, Evans, & Strumpf, 1994; Sullivan-Marx, Strumpf, Evans, Baumgarten, & Maislin, 1999). Cohen and Spector (1996) found that a higher RN intensity was associated with a lower mortality rate, and a higher intensity of LPN staffing was significantly associated with improved resident functional outcomes as measured by ADLs. However, having more nurse aides had no impact on resident outcomes. In another study, Zimmerman Gruber-Baldini, Hebel, Sloane, and Magaziner (2002) found that high rates of incident infection were associated with high LPN staffing and low nurse aide staffing. Carter and Porell (2003) found that residents of facilities with nursing personnel expenses more heavily allocated to LPNs (as opposed to RNs) were at greater risk of hospitalization than otherwise similar residents of other nursing facilities. The optimal level or skill mix of staff for a given level of resident acuity, however, remains unclear.

Studies of Staffing in RC/AL Settings
Regulatory requirements for staffing in RC/AL vary widely from state to state. As of the year 2000 (the year data collection ended for this study), 15 states had developed minimum staffing requirements for RC/AL, and another 15 required the presence or on-call availability of a direct care staff member 24 hr a day. The minimum staffing requirements are separate for waking hours and sleeping hours in most states. For example, Alabama requires 1 staff person per 6 residents at all times, whereas South Carolina requires 1 staff person per 10 residents during waking hours and 1 staff member per 44 residents during sleeping hours. Some states, such as Pennsylvania and South Dakota, require a minimum number of care hours per resident day rather than the number of staff (Hodlewsky, 2001). Staff training or licensure may also be specified.

Researchers have reported little about the relationship between staffing levels and care outcomes in RC/AL settings, however, and the extent to which the relationships observed in nursing homes also hold for RC/AL is not known. Zimmerman and colleagues (2005) used data from the CS-LTC to show that simply having an RN/LPN was associated with a significantly higher likelihood of nursing home placement, and that the amount of RN staff time per resident was very slightly protective of hospitalization. They found no effects of the presence of a licensed nurse or of aide staff hours per resident. Those analyses did not consider, however, total direct staff care hours per resident or the potential implications of staff mix (measured as the proportion of resident care hours provided by licensed staff) for resident health outcomes. Aside from the study by Zimmerman and colleagues (2005), the existing literature provides little information on factors associated with staffing levels and subsequent resident outcomes.

The current study extends the literature by using additional data from the CS-LTC to take a closer look at staffing in RC/AL settings and to explore the relationships between staffing intensity, skill mix, facility characteristics (especially size and ownership, in consideration of the costs of increasing staffing levels or skill mix), and resident outcomes. Findings have implications for practice and policy related to care provision in RC/AL.

	Methods

TOP Abstract Methods Results Discussion References

 
Data Sources
These analyses used data collected by the CS-LTC for RC/AL residents and facilities in four states: Florida, Maryland, New Jersey, and North Carolina. The CS-LTC defined RC/AL as facilities or discrete portions of facilities that are licensed by the state at a non-nursing-home level of care and that provide room, board, 24-hr oversight, and assistance with ADLs. A multilevel sampling frame was used to select facilities and residents for participation (Figure 1). Within each state, a region of counties was identified that represented the state in terms of rural/urban mix and eight demographic and health-related variables (e.g., proportion of elderly population, proportion of health care providers). The sampling frame consisted of all licensed RC/AL facilities within each region, stratified into three types: (a) facilities with fewer than 16 beds, (b) facilities with 16 or more beds of a "new-model" type (i.e., built after January 1, 1987, and having at least one of the following four components: at least two different monthly private pay rates, 20% or more of residents requiring assistance with transfer, 25% or more of residents who are incontinent daily, and either an RN or an LPN on duty at all times), and (c) "traditional" facilities with 16 or more beds not meeting new-model criteria. Approximately 59% of facilities contacted by the study agreed to participate. Nonparticipating facilities had more owners working more hours in the facility, had more rate levels, and housed a slightly less impaired resident population. Participating and nonparticipating facilities were similar with respect to proprietary status; affiliation with other long-term-care facilities; facility age, size, and occupancy rate; and resident characteristics (age, ethnicity, and race).

View larger version (12K): [in this window] [in a new window]   Figure 1. Sampling overview

Measures
At baseline, the facility administrator provided data on facility characteristics and staffing, specifically staff hours defined as time spent by facility or contract staff (RNs, LPNs, or PCAs) in resident care tasks (nursing or personal care tasks done), as distinct from time spent doing other tasks (e.g., administration, planned activities, informal activities, social work, meal planning/preparation, housekeeping, maintenance, transportation). Data on time spent in specific nursing or personal care tasks (e.g., medicine administration, catheter care, bathing, dressing) were not collected due to concerns about burdens of such detailed collection. If an RN did a task that was typically considered to be a personal care task, such as bathing, then we counted the time spent on that task as personal care task time, not as RN time. Resident-level baseline and follow-up information (at 3, 6, 9, and 12 months post-baseline) was obtained by interview with the direct care staff who knew the resident the best. Baseline data included resident age; gender; race; marital status; education; tenure in facility; dementia diagnosis; comorbid conditions; and established measures of ADLs (Morris, Fries, & Morris, 1999), cognition (Hartmaier, Sloane, Guess, & Koch, 1994), and depressive symptoms (Alexopolous, Abrams, Young, & Shamoian, 1988). Outcomes obtained at telephone follow-up included mortality, nursing home transfer, hospitalization (excluding psychiatric), and new or worsening morbidity (fracture, skin ulcer, paralysis of arm/leg, bleeding from stomach/bowel, diabetes, stroke, congestive heart failure, heart condition, nonfracture accident/injury, infection, and temporary nursing home placement). Further details about the CS-LTC sampling, data collection procedures, and participation rates appear elsewhere (Zimmerman, Sloane, & Eckert, 2001).

Of the 193 participating facilities, 23 did not contribute data to these analyses for the following reasons: 10 administrators did not provide staffing information, 6 were missing critical facility information, and 7 did not report any paid staff in any category. The final sample for this analysis included 170 facilities (88%): 94 facilities with fewer than 16 beds, 37 traditional facilities, and 39 new-model facilities. Data were available for 1,894 residents at these facilities, representing 91.1% of the total residents in the study.

Methods for Analysis of Staffing Intensity and Skill Mix
Analysis of staffing sought to provide a descriptive assessment of (a) overall staff availability as measured by staff hours per day, (b) staffing intensity as measured by care hours per resident per week, and (c) skill mix as measured by the percentage of total care hours provided by licensed staff (RNs or LPNs). We created a staffing hierarchy for considering these measures, defined by the highest level of trained staff: RN, LPN, or other direct resident care staff (PCA or certified nursing aide).

We conducted regression analyses to determine the relationship between staffing intensity and staff skill mix and facility characteristics including state, type, size, age, ownership, affiliation with a hospital or chain of facilities, and resident case mix (percentage of residents reported as incontinent, chairbound, or with dementia). These case-mix measures were estimates of the percentages of residents with the different conditions. We measured facility size using splines to examine different effects of very small facilities (3–7 residents), small facilities (7–15 residents), and facilities with 16 or more residents, as preliminary analyses indicated that economies of scale were reached once the number of residents exceeded 15.

Methods for Analysis of Resident Outcomes in Relation to Staffing Intensity and Skill Mix
We performed longitudinal analyses using resident-level data to assess the association between (a) staffing intensity as measured by direct care hours or skill mix and (b) the incidence of mortality, nursing home transfer, hospitalization, and incident morbidity. We used resident-level data to control for within-facility variation of resident case mix. All analytic models accounted for facility-level clustering (i.e., lack of statistical independence between residents). We viewed mortality and nursing home transfer as endpoint events and thus assessed them using Cox proportional hazards methods (Cox & Oakes, 1984). We assessed hospitalization and morbidity, measured on a quarterly basis, using repeated measures analysis. We employed generalized estimating equations to fit a Poisson regression model for count data of number of events (Liang & Zeger, 1986). We estimated relative risks associated with mortality, nursing home transfer, hospitalization, and incident morbidity using facility staffing intensity as the exposure variable and resident-level covariates to adjust for case-mix differences among facilities. Analyses controlled for clustering by facility and exposure time, as well as resident age, gender, race, marital status, education, tenure in facility, functional status as reflected by the Minimum Data Set ADL measure (Morris et al., 1999), cognition as measured by dementia diagnosis and the MDS Cognition Scale (Hartmaier et al., 1994), depression as measured by the Cornell Scale for Depression in Dementia (Alexopolous et al., 1988), and number of morbidities. A relative risk greater than 1 indicates that the rate in the exposed group (i.e., the per-unit increase in the referent group) is greater than that in the unexposed group; values less than 1 indicate that the rate is lower in the exposed group than in the referent.

	Results

TOP Abstract Methods Results Discussion References

 
Descriptive Assessment of Staffing Intensity and Skill Mix
Table 1 shows the distribution of the facility sample by the nursing staff member with the most training and by facility type. In all, 77 facilities (45%) in the sample had an RN available for resident care, whereas 70 facilities (41%) had only PCAs providing resident care. The majority (70%) of the small facilities had only PCAs providing care, whereas roughly three quarters of traditional and new-model facilities had RNs.

View larger version (15K): [in this window] [in a new window]   Figure 2. Task hours per resident per week by facility and staff type for facilities with registered nurses (n = 77). RN = registered nurse; LPN = licensed practical nurse

View larger version (12K): [in this window] [in a new window]   Figure 3. Skill mix: Licensed (RN or LPN) care hours as a percentage of total care hours per resident by facility type. RN = registered nurse; LPN = licensed practical nurse

	Discussion

TOP Abstract Methods Results Discussion References

 
This study found that the mean number of direct care hours per resident per week ranged from 0.87 to 1.80, with the higher figure pertaining to facilities that did not have licensed staff. Although these differences may be due to the fact that facilities without RNs or LPNs are almost all smaller facilities with fewer than 16 beds, the patterns may also be due to differences in responses by administrators if direct care in these smaller settings is more broadly defined as opposed to facilities where RN practices and norms may shape the culture and perceptions as to what constitutes direct care. By comparison, data from the Online Survey, Certification and Reporting System have shown approximately 3 nurse/nurse aide staff hours per patient day in that nursing home setting (although this estimate includes administration time and other tasks in addition to direct patient care time). The modest level of direct care hours is consistent with the high level of functional ability of many RC/AL residents and the less intensive staff structure of RC/AL facilities when compared to nursing homes. Indeed, analyses of the association between direct care hours per resident and four health outcomes showed no evidence of benefits from more hours of care per resident, suggesting that given the generally high functional level and limited medical needs of many RC/AL residents, the vast majority of facilities may be sufficiently staffed to cover residents' needs. Of course, these medically related outcomes might be less sensitive to the effects of staffing hours than other outcomes, such as depression or quality of life. Furthermore, the acuity level of RC/AL facilities has been increasing, and RC/AL settings strive to provide supportive care and to maintain or improve functioning of residents. Also, many residences encourage aging in place rather than transfer to nursing homes as functional needs increase. Thus, the extent to which more care hours may be needed as acuity increases is an important issue for RC/AL residents who seek quality living situations at affordable cost, as well as for state policy makers involved in facility regulation, payment, and access issues. Another area of future work could consider additional aspects of staffing, such as use of contract rather than facility staff.

With respect to skill mix, this study found that the LPN time in facilities without RNs was very close to the combined time for RNs and LPNs in facilities with RNs (roughly 0.22 hr per resident per week, as shown in Table 2), leading one to hypothesize that RNs and LPNs may be very close substitutes in RC/AL settings. The analyses with respect to skill mix produced one important statistically significant finding. Having a greater proportion of direct care hours provided by licensed (RN or LPN) staff was highly protective against hospitalization on average over all facilities, although the effect varied with facility case mix as measured by percentage of residents with dementia. No protective effect of skill mix was estimated for facilities with a very low percentage of residents with dementia, but the effect was extremely strong in facilities with a very high percentage of residents with dementia. The overall results are consistent with studies from nursing homes that have demonstrated reduced hospitalizations in relation to greater skill mix (Carter & Porell, 2003), and the results held even when analyzing only facilities with some licensed staff (RNs or LPNs). The explanation for this finding is not clear from our analyses, and, given the complex interactions between skill level, case mix, and care practices, we can only speculate on the underlying issues. The findings for skill mix suggest that greater levels of supervision or involvement in resident care by more highly trained (licensed) staff may result in timely identification of medical problems (thereby enabling prevention of acute care admissions) and in greater ability to administer treatments (e.g., parenteral antibiotics), to monitor acutely ill patients, or perhaps to provide appropriate terminal care. This interpretation has implications for staffing decisions and possibly for the training and role of PCAs. However, further research is required to understand the exact mechanism by which reduced hospitalizations occur among RC/AL residents in facilities with greater skill mix. Furthermore, the fact that nonprofit facilities had greater levels of skill mix, controlling for other factors including the percentage of residents with dementia, may mean that nonprofit RC/AL facilities may be more likely to have lower rates of subsequent hospitalizations than for-profit facilities; this suggestion is consistent with findings in nursing homes (Zimmerman et al., 2002).

One interesting nonsignificant finding was that the proportion of care hours provided by an RN or LPN was not related to nursing home placement, though greater skill mix showed a trend toward being protective of placement. Hawes and colleagues (1999) found that residents in RC/AL facilities that did not have a full-time RN and did not offer nursing care with their own staff were twice as likely to enter a nursing home between baseline and follow-up. In contrast, prior analyses of the CS-LTC data found that having an RN or LPN was associated with a 50% greater risk of nursing home transfer for a resident, though the proportion of care hours provided by an RN or LPN was not related to nursing home placement (Zimmerman et al., 2005). Therefore, the current and prior findings using CS-LTC data are consistent with respect to the staffing intensity measure, and it may be that the discrepancy between the finding by Hawes and colleagues and Zimmerman and associates (2005) with respect to the presence of an RN is attributable to the importance of skill mix rather than simply the presence of an RN. The skill mix employed in these analyses was created using numbers of direct care hours and captures more variation in levels of care than a simple indicator.

It is also somewhat surprising that the hospitalization results are so strong given the lack of finding for morbidity. Morbidity was defined as onset of a number of conditions: fracture, skin ulcer, paralysis of arm/leg, bleeding from stomach/bowel, diabetes, stroke, congestive heart failure, heart condition, nonfracture accident/injury, and infection. Many of these conditions would, of course, require hospitalization, although these conditions can occur at some level without hospitalization. The difference may be one of data quality, in that the staff who provided the data may have been more certain that a hospitalization had occurred than that a morbid event had transpired. Alternatively, increased staffing may mainly improve a facility's ability to manage, rather than prevent, acute conditions.

The results of these analyses raise questions related to the costs of increasing skill mix, challenges in maintaining an optimal skill mix, and payment for such services. Facility-level analyses did not show evidence of economies of scale in skill mix, aside from the fact that smaller facilities (<16 beds) were less likely to have RNs or LPNs on staff. We did find evidence of economies of scale, however, with respect to staffing intensity (i.e., Figure 2 indicated economies of scale in use of licensed staff, with greater efficiencies of hiring more highly skilled staff in larger facilities). Therefore, the ability to offer some licensed staff involvement may be more feasible for facilities of at least a certain minimum size (e.g., facilities with at least seven or more residents). This finding is further documentation of the challenges of smaller RC/AL facilities, which have created concern in the field (Morgan, Eckert, Gruber-Baldini, & Zimmerman, 2004; Zimmerman et al., 2003). Yet the benefits from reduced hospitalization in relation to greater skill mix remained strong in facilities with RNs or RNs/LPNs rather than just facilities with licensed staff compared to facilities without licensed staff, and the evidence from these analyses may be of special interest to public payers for health care. Although Medicare does not cover RC/AL services, the Medicare program would benefit from reduced financial liability related to hospitalization. Thus, although residents, family, and RC/AL facilities all benefit from avoided hospitalizations, further determination of the specific contributions of greater skill mix may be useful for understanding the mechanisms and trade-offs (e.g., additional costs for nursing care within RC/AL in comparison to Medicare savings) for preventing hospitalization in RC/AL residents and concomitant implications for public spending.

	Footnotes

 
This research was supported by Grants R01 AG13871, R01 AG13863, and K02 AG00970 from the National Institute on Aging. We acknowledge and note appreciation for the cooperation of the facilities, residents, and families participating in the Collaborative Studies of Long-Term Care (CS-LTC). We also extend gratitude to Dr. Verita Custis Buie, who was part of the original CS-LTC study team and who was integrally involved in data collection for this cohort.

¹ Department of Health Policy and Administration, University of North Carolina at Chapel Hill.

² Division of General Internal Medicine, University of Pennsylvania, Philadelphia.

³ Cecil G. Sheps Center for Health Services Research, University of North Carolina at Chapel Hill.

⁴ School of Social Work, University of North Carolina at Chapel Hill.

⁵ Division of Gerontology, Department of Epidemiology and Preventive Medicine, University of Maryland School of Medicine, Baltimore.

Decision Editor: Linda S. Noelker, PhD

Received for publication November 8, 2006. Accepted for publication April 16, 2007.

	References

TOP Abstract Methods Results Discussion References

 

• Alexopolous, G. S., Abrams, R. C., Young, R. C., & Shamoian, C. A., (1988). Cornell Scale for Depression in Dementia. Biological Psychiatry, 23, 271-284.[Medline]
• Carter, M. W., & Porell, F. W., (2003). Variations in hospitalization rates among nursing home residents: The role of facility and market attributes. The Gerontologist, 43, 178-191.
• Castle, N. G., & Fogel, B., (1998). Characteristics of nursing homes that are restraint free. The Gerontologist, 38, 181-188.[Abstract]
• Cohen, J. W., & Spector, W. D., (1996). The effect of Medicaid reimbursement on quality of care in nursing homes. Journal of Health Economics, 15, 23-48.[Medline]
• Cox, D. R., & Oakes, D., (1984). Analysis of survival data. London: Chapman & Hall.
• Harrington, C., Kovner, C., Mezey, M., Kayser-Jones, J., Burger, S., & Mohler, M., et al (2000). Experts recommend minimum nurse staffing standards for nursing facilities in the United States. The Gerontologist, 40, 5-16.[Abstract]
• Hartmaier, S. L., Sloane, P. D., Guess, H. A., & Koch, G. G., (1994). The MDS Cognition Scale: A valid instrument for identifying and staging nursing home residents with dementia using the Minimum Data Set. Journal of the American Geriatrics Society, 42, 1173-1179.[Medline]
• Hawes, C., Rose, M., & Phillips, C., (1999). A national study of assisted living for the frail elderly: Results of a national survey of facilities. Washington, DC: U.S. Department of Health and Human Services; Office of Office of the Assistant Secretary for Planning and Evaluation; Office of Disability, Aging, and Long-Term Care Policy.
• Institute of Medicine. (1996). Nursing staff in hospitals and nursing homes: Is it adequate?. Washington, DC: National Academy Press.
• Kane, R. A., Kane, R. L., Illston, L. H., Nyman, J. A., & Finch, M. D., (1991). Adult foster care for the elderly in Oregon: A mainstream alternative to nursing homes? American Journal of Public Health, 81, 1113-1120.[Abstract/Free Full Text]
• Kolanowski, A., Hurwitz, S., Taylor, L. A., Evans, L., & Strumpf, N., (1994). Contextual factors associated with disturbing behaviours in institutionalized elders. Nursing Research, 43, (2), 73-79.[Medline]
• Hodlewsky, R. T., (2001). Staffing problems and strategies in assisted living. In S. Zimmerman, P. D. Sloane, & J. K. Eckert (Eds.), Assisted living: Needs, practices, and polities in residential care for the elderly (pp. 78–91). Baltimore: Johns Hopkins University Press.
• Liang, K. Y., & Zeger, S. L., (1986). Longitudinal data analysis using generalized linear models. Biometrica, 73, 13-22.
• Manard, B. B., & Cameron, R., (1997). A national study of assisted living for the frail elderly: Report on in-depth interviews with developers. Washington, DC: Office of the Assistant Secretary for Planning and Evaluation; Administration on Aging and the National Institute on Aging; U.S. Department of Health and Human Services.
• Mollica, R., & Johnson-Lamarche, H., (2005). State residential care and assisted living policy: 2004. Retrieved March 31, 2005, from http://aspe.hhs.gov/daltcp/reports/04alcom.htm.
• Morgan, L. A., Eckert, J. K., Gruber-Baldini, A. L., & Zimmerman, S., (2004). Policy and research issues for small assisted living facilities. Journal of Aging & Social Policy, 16, (4), 1-16.
• Morgan, L. A., Gruber-Baldini, A. L., & Magaziner, J., (2001). Resident characteristics. In S. Zimmerman, P. D. Sloane, & J. K. Eckert (Ed.), Assisted living: Needs, practices, and policies in residential care for the elderly (pp. 144–172). Baltimore: Johns Hopkins University Press.
• Morris, J. N., Fries, B. E., & Morris, S. A., (1999). Scaling ADLs within the MDS. Journal of Gerontology: Medical Sciences, 54A, M546-M553.[Abstract]
• Park, N. S., Zimmerman, S., Sloane, P. D., Gruber-Baldini, A. L., & Eckert, J. K., (2006). An empirical typology of residential care/assisted living based on a four state study. The Gerontologist, 46, 238-248.[Abstract/Free Full Text]
• Schmidt, I. K., & Svarstad, B. L., (2002). Nurse–physician communication and quality of drug use in Swedish nursing homes. Social Science & Medicine, 54, 1767-1777.
• Sullivan-Marx, E., Strumpf, N. E., Evans, L. K., Baumgarten, M., & Maislin, G., (1999). Predictors of continued physical restraint use in nursing home residents following restraint reduction efforts. Journal of the American Geriatrics Society, 47, 342-348.[Medline]
• Svarstad, B. L., & Mount, J. K., (2001). Chronic benzodiazepine use in nursing homes: Effects of federal guidelines, restraint mix, and nurse staffing. Journal of the American Geriatrics Society, 49, 1673-1678.[Medline]
• Svarstad, B. L., Mount, J. K., & Bigelow, W., (2001). Variations in the treatment culture of nursing homes and responses to regulations to reduce drug use. Psychiatric Services, 52, 666-672.[Abstract/Free Full Text]
• Weech-Maldonado, R., Meret-Hanke, L., Neff, M. C., & Mor, V., (2004). Nurse staffing patterns and quality of care in nursing homes. Health Care Management Review, 29, (2), 107-116.[Medline]
• Zimmerman, S., Gruber-Baldini, A. L., Hebel, J. R., Sloane, P. D., & Magaziner, J., (2002). Nursing home facility risk factors for infection and hospitalization: Importance of RN turnover, administration and social factors. Journal of the American Geriatrics Society, 50, 1987-1995.[Medline]
• Zimmerman, S., Gruber-Baldini, A. L., Sloane, P. D., Eckert, J. K., Hebel, J. R., & Morgan, L. A., (2003). Assisted living and nursing homes: Apples and oranges? The Gerontologist, 43, (Special Issue II), 107-117.[Abstract/Free Full Text]
• Zimmerman, S., Sloane, P. D., & Eckert, J. K., (Eds.) (2001). Assisted living: Needs, practices, and polities in residential care for the elderly. Baltimore: Johns Hopkins University Press.
• Zimmerman, S., Sloane, P. D., Eckert, J. K., Gruber-Baldini, A. L., Morgan, L. A., & Hebel, J. R., et al (2005). How good is assisted living? Findings and implications from an outcomes study. Journal of Gerontology: Social Sciences, 60B, S195-S204.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation