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

Does Cognitive Status Moderate the Health Effects of Single-Person Room Transfers on Nursing Home Residents?

^a Department of Health and Nutrition Sciences, Brooklyn College, NY

Correspondence: Jerrold Mirotznik, PhD,MPH, Department of Health and Nutrition Sciences, Brooklyn College, 2900 Bedford Avenue, Brooklyn, NY 11210. E-mail: Jerrym{at}brooklyn.cuny.edu.

Decision Editor: Laurence G. Branch, PhD

	Abstract

TOP Abstract Theoretical Models Methods Results Discussion References

 
Purpose: This study investigated whether cognitively impaired nursing home residents are particularly vulnerable to harmful effects during single-person, intrabuilding room transfers. Design and Methods: A variation of a pretest–posttest experimental-comparison group design was used. Data on cognitive status, mortality, and seven morbidity outcome measures were abstracted from the Minimum Data Set Plus and were analyzed by means of event history analyses, controlling for covariates as well as baseline status of outcome variables. Results: None of the Relocation x Cognitive Status interaction effects were significant at the Bonferroni corrected p value. Implications:These findings suggest that cognitive status may not moderate the health effects of single-person room transfers.

Key Words: Cognitive impairment • Morbidity • Mortality • Intrainstitutional relocation • Room transfers

Over the past 40 years, a growing body of research has addressed the question of whether relocation of nursing home residents causes deleterious health effects. Although several studies have documented an increase in death posttransfer (Bourestom and Pastalan 1975; Robertson, Warrington, and Eagles 1993), others have not (Conroy and Adler 1998; McAuslane and Sperlinger 1994; Mirotznik and Los Kamp 2000). Studies focusing on morbidity and psychosocial outcomes have yielded equally conflicting results (Friedman et al. 1995; Holzapfel, Schoch, Dodman, and Grant 1992; Lander, Brazill, and Ladrigan 1997; Mallick and Whipple 2000; Mirotznik and Los Kamp 2000; Rogers, Stuart, Sheffield, Swee, and Formica 1990; Thorson and Davis 2000). By way of explaining these contradictory findings, it has been suggested that the impact of relocation depends on the conditions of the relocation as well as on the characteristics of the residents being relocated (Bourestom and Pastalan 1981).

Probably no individual characteristic is more frequently cited as moderating relocation stress than cognitive status (Kowalski 1981). Indeed, it has become somewhat of a truism that cognitively impaired residents are at particular risk for deleterious relocation effects. Yet, in a recent review of 17 studies investigating the cognitive status vulnerability hypothesis, Mirotznik and Los Kamp 2000 found 12 studies lacked a comparison group of nonmovers (Aldrich 1964; Friedman et al. 1995; Guttman and Herbert 1976; Kral, Grad, and Berenson 1968; Lander et al. 1997; Lieberman and Tobin 1983, pp. 151–152; Markus, Blenkner, Bloom, and Downs 1972; Marlowe 1974; Miller and Lieberman 1965; Nirenberg 1983; Ogren and Linn 1971). Without a comparison group, one cannot rule out the possibility that the greater mortality and/or morbidity among relocated residents who are cognitively impaired may merely reflect the fact that such residents deteriorate at a higher rate in general, regardless of the occurrence of relocation (Coffman 1983; van Dijk, Dippel, and Habbema 1991).

The remaining five studies included a comparison group but yielded contradictory findings. One found moderately cognitively impaired residents (Pruchno and Resch 1988) and another severely impaired residents (Csank and Zweig 1980) to be at greatest risk. A third study found that moderately impaired and unimpaired residents as a group experienced more effects, including a positive prerelocation and negative postrelocation response, than residents who were severely impaired (Mirotznik 1995). And two studies found no evidence of a moderating effect of cognitive status at all (Goldfarb, Shahinian, and Burr 1972; Lieberman and Tobin 1983, p. 149). Reporting on their own investigation of a mass intrainstitutional, interbuilding transfer, Mirotznik and Los Kamp 2000 also found no evidence that cognitive status influenced relocation outcome.

The question of whether cognitive status moderates relocation outcome is currently of relevance for at least two reasons. First, it has recently been estimated that 48% of nursing home residents have a diagnosed dementia (Leon and Moyer 1999), suggesting that large numbers of residents may be vulnerable to untoward outcomes if relocated. And second, there has been an enormous proliferation of dedicated special care units (SCUs) for nursing home residents with dementia. Leon 1994 and Leon, Cheng, and Alvarez 1997 reported that in their 1991 national survey of nursing homes about 10% of 15,553 facilities were found to have SCUs for an estimated 47,857 residents, and in their follow-up 1995 survey 19.5% of 16,827 facilities offered SCUs for 113,505 residents. Undoubtedly, the greater availability of SCU beds has led to a concomitant increase in intrainstitutional transfers of residents into and out of SCUs (Kovach 1998).

In particular, two kinds of intrainstitutional relocations have been noted to occur as a result of the establishment of a SCU in a facility: intrabuilding transfers (i.e., moving residents from one room to another within the same building) and interbuilding transfers (i.e., moving residents out of one building into another building within the same nursing home complex; Reingold and Werner 1994). Intrainstitutional relocations may further differ in terms of the number of residents being moved. The establishment of a new SCU at a facility or the modernization of an old SCU, for instance, might involve the transfer of large numbers of residents. Transfers between a nonspecialized unit and an already existing SCU, on the other hand, may involve moving only one person at a time.

Interbuilding versus intrabuilding transfers and mass versus single-person transfers may comprise different stressors, involve different supports, allow for different coping strategies, and consequently have different potential for harmful outcomes. Yet, not all types of intrainstitutional relocations have been equally investigated. Few studies have focused on single-person, within-building room changes. Indeed, of the six studies mentioned earlier that used a comparison group to test the cognitive status vulnerability hypothesis, only one investigated this type of move (Pruchno and Resch 1988). Given the likelihood that most transfers into SCUs are of this sort, there is a need for further research on the question of whether cognitive status moderates outcome during single-person, intrabuilding relocation.

	Theoretical Models

TOP Abstract Theoretical Models Methods Results Discussion References

 
Three broad types of theoretical models have been suggested to explain how cognitive status influences relocation outcome: stress-reaction models, coping models, and stressor-coping models (Mirotznik and Los Kamp 2000). Stress-reaction models assume that relocation involves a differential degree of stress for cognitively impaired than for cognitively unimpaired individuals. Lawton and Simon 1968 environmental docility hypothesis, for example, suggests that because cognitively impaired individuals are more dependent on their external environment, relocation involves more of a disruption for them (Pruchno and Resch 1988).

Coping models emphasize the limited abilities and resources of impaired individuals for dealing with relocation. Schulz and Brenner 1977 hypothesized that perceived control and predictability during relocation influence outcome. This model suggests that cognitively impaired persons may not be able to buffer themselves from the stress of relocation by developing a sense of personal control or by rendering the change more predictable by availing themselves of preparation programs (Pruchno and Resch 1988).

Stressor-coping models focus on the specific aspects of relocation that induce stress as well as incorporate the notion of limited coping resources. Csank and Zweig 1980 suggested that cognitively impaired residents are not capable of experiencing anticipatory relocation stress. Rather, for them stress results from the actual relocation itself, specifically, from loss of a familiar environment, which triggers an anxiety reaction. It is this anxiety reaction coupled with a lack of coping resources that results in harmful relocation effects.

Two stressor-coping models offer different predictions for those who are moderately impaired than for those who are severely impaired. Lieberman and Tobin 1983 pathognomonic sign model suggests that severely impaired residents, given their limited coping capacity, will experience deleterious effects regardless of the degree of stress involved in a particular relocation. The effect of relocation on those with adequate coping resources (i.e., those who are moderately impaired or unimpaired) would largely depend on the exact nature of the relocation stressor. In contrast, Pruchno and Resch 1988 modified environmental docility hypothesis suggests that highly impaired individuals will not be affected by relocation because they lack the threshold capacity needed for dependency on environmental cues. Moderately impaired individuals, in contrast, are very dependent on such cues and, as such, experience relocation as highly stressful. Unimpaired residents are also unlikely to be affected because they have the coping reserves for adjusting to change.

Hypotheses
Relocation can be conceptualized as a process consisting of three stages: (a) a decision and preparation stage before relocation, also known as an anticipatory stage; (b) an impact or short-term adjustment stage within which the actual physical transfer occurs; and (c) a settling-in or long-term adjustment stage (Borup 1981; Tobin and Lieberman 1976). Each stage may have its own potential for stress (Tobin and Lieberman 1976). During Stage 1, stress may result from anticipation of the relocation. During Stage 2, stress may result from the physical move itself and the actual loss of a familiar environment, and during Stage 3, it may result from adjusting to a new environment.

The theoretical models described earlier to explain the greater vulnerability of those who are cognitively impaired focus on events that occur during Stages 2 and 3 of the relocation process. They suggest that cognitively impaired residents are likely to exhibit negative effects as a result of an actual change in their environment. Indeed, one of the models explicitly argues against the possibility of cognitively impaired persons experiencing stress in anticipation of relocation (Csank and Zweig 1980). On the basis of these models, therefore, one would predict that the cognitively impaired would be more likely than the cognitively unimpaired to exhibit deleterious effects postrelocation.

Furthermore, on the basis of the various models, competing predictions can be derived about the pattern of these postrelocation effects. Lawton and Simon 1968 environmental docility hypothesis and Schulz and Brenner 1977 each suggested a positive linear relationship between cognitive impairment and negative effects; the more impaired individuals are, the more dependent they are on their environment, the more limited their coping capacities are, the greater their deleterious effects. Pruchno and Resch 1988 modified docility hypothesis suggested a curvilinear pattern, with those highest and lowest on impairment exhibiting the fewest effects and those with moderate impairment exhibiting the greatest effects. Last, the pathognomonic sign model by Lieberman and Tobin 1983 suggested a nonlinear pattern, whereby individuals with high impairment would necessarily experience harmful effects, and those below that threshold level (i.e., the moderately impaired and the unimpaired), would not. Hence, I addressed the following specific questions in this study:

1. Are cognitively impaired residents more likely than cognitively unimpaired residents to exhibit negative relocation effects during the impact and/or settling-in stages of single-person, intrabuilding transfer?
   
2. Do severely or moderately cognitively impaired residents exhibit more harmful postrelocation effects? Put otherwise, do posttransfer effects manifest a linear, curvilinear, or a nonlinear pattern with regard to cognitive status?
   

	Methods

TOP Abstract Theoretical Models Methods Results Discussion References

 
Setting
The study was conducted at two nursing homes that were participating agencies of Metropolitan Jewish Health System (MJHS): Shorefront Jewish Geriatric Center, a 360-bed facility, and MJG Nursing Home Co., Inc., a 529-bed facility. Both nursing homes are located in urban residential neighborhoods of Brooklyn, New York. The nursing staff at these facilities were similar and included a culturally and ethnically diverse group of individuals. Although nursing (registered nurse, licensed practical nurse, and nursing assistant) staffing ratios differed between them, the same nursing staffing methodology was used in both facilities. Floors with residents with more acute conditions were allocated a larger number of staff. At both facilities, similar criteria were used for hiring, training, and managing staff. Both institutions provided full-time onsite medical staff as well as the full range of ancillary staff. Specialty consultants most often held privileges at both facilities. Policies and procedures for the two facilities were similar if not identical and reflected a shared administrative perspective.

At both Shorefront and MJG, the residents were mainly female (68% and 72%, respectively), White (94% and 98%, respectively), widowed (63% and 56%, respectively), spoke English as a primary language (83% and 74%, respectively), and had a mean age of 82.5 and a mean length of stay of 1.5 and 2.2 years, respectively. In all analyses, nursing home residence was controlled for to ensure that any differences between Shorefront and MJG did not confound the study's results.

Study Design
The study used a variation of the pretest–posttest experimental-comparison group design. Although intrabuilding transfers routinely occurred at Shorefront and MJG, at any one point in time few residents experienced such a transfer. To accumulate a large enough number of participants who experienced an intrabuilding transfer, Campbell and Hudson 1985 technique for investigating infrequent events was used. This technique, the creation of synthetic cohorts, yields a multiwave data set centered on the event of interest with pre- and postevent data. Thus, the intrabuilding moves investigated in this study did not occur at the same time but over a period of time. Specifically, all residents in the two facilities who experienced an intrabuilding transfer at any point during the 4-year period from July 1, 1993, through June 30, 1997, were identified.

Two types of moves were distinguished: those involving transfers from one room to another within the same floor and those involving transfers between floors. As both Shorefront and MJG had heterogeneous staffing patterns, with floors with sicker residents having a larger concentration of personnel, moves between floors typically occurred in response to a change in a resident's health status. If a resident's health declined, he or she was moved to a floor with a heavier nursing pattern to accommodate the more intensive needs of sicker residents. If a resident's health status improved, he or she was moved to a floor with a lighter nursing pattern. Moves between rooms on the same floor typically occurred for nonhealth reasons (i.e., roommates were not getting along and either they or their family requested the transfer).

Residents who moved between floors were excluded from the study for two reasons. First, to determine if intrabuilding moves affected health status, it was necessary to remove residents for whom a change in health was known to have caused a change in rooms. Second, as mentioned, residents whose health status worsened were moved to floors that provided more intensive care. This intensive care could confound any effects that might result from being transferred. Thus, for a clearer assessment of transfer effects themselves, it was preferable to study residents transferred between rooms on the same floor. The experimental group was, therefore, limited to only residents who moved between rooms within a floor. To ensure that worsening health did not precipitate within-floor room changes, the experimental group was further reduced by excluding any resident who had a Minimum Data Set clinical assessment (discussed below) triggered by a significant decline in health within a 1-month period before room transfer. The comparison group consisted of residents who had not undergone either a within- or between-floor transfer during the 4-year study period. Given my focus on long-term nursing home residents, I also excluded all subacute residents as well as hospice and maximum care unit residents. The experimental group so constituted consisted of 150 residents, and the comparison group consisted of 626 residents, for a total sample size of 776.

Data
The Minimum Data Set Plus (MDS+) served as the main source of data. The MDS+ is a federally mandated, standardized clinical assessment form for monitoring the functional and medical status of nursing home residents (Morris et al. 1990). Research on the psychometric properties of the instrument has indicated that the MDS+ measures certain dimensions better than others. The MDS items for cognition, activities of daily living (ADLs), disease diagnoses, administered medication, and psychosocial well-being have been found to exhibit good to excellent interrater reliability (Hawes et al. 1995). A recent study also documented temporal stability for ADLs over a 7-day period (Graney and Engle 2000). The MDS items for mood have been found to exhibit relatively poor reliability (Hawes et al. 1995). The MDS measures for ADLs, cognitive status, and communication correlate highly with comparable rating scales (Frederiksen, Tariot, and De Jonghe 1996; Hartmaier et al. 1995; Morris et al. 1994; Phillips, Chu, Morris, and Hawes 1993), although the measures for problem behavior and mood do not (Frederiksen et al. 1996; Schnelle, Wood, Schnelle, and Simmons 2001). Also, some evidence attests to the factorial validity of the MDS items for ADLs and cognitive status (Casten, Lawton, Parmelee, and Kleban 1998). Inconsistent results have been uncovered for the factorial validity of mood, behavior problems, and social engagement (Casten et al. 1998; Mor et al. 1995). International research on the MDS has also found good validity for ADLs and cognitive function, but not for mood and behavior (Achterberg, Pot, van Campen, and Ribbe 1999). MDS diagnostic and drug data have been found to exhibit good predictive positive values when assessed, respectively, in terms of diagnoses on hospital discharge claims and the probability of drug use on the basis of age, gender, and labeled indications (Gambassi et al. 1998). Although exhibiting excellent interrater reliability (Hawes et al. 1995; Resnick, Brandeis, Baumann, and Morris 1996), the MDS continence items were found to have questionable validity (Crooks, Schnelle, Ouslander and McNees 1995).

Evidence has suggested that MDS items may be less reliable for those who are cognitively impaired (Casten et al. 1998), particularly those items based on staff observation and assessment (Phillips et al. 1993). MDS items based on medical records exhibit the highest interrater reliabilities that in practical terms differ little between the cognitively impaired and the intact (Phillips et al. 1993). Evidence has also indicated that the internal consistency reliability of the MDS items representing ADLs does not vary across levels of cognitive impairment (Phillips and Morris 1997).

Historically, questions have been raised about the use in research of data collected for clinical/administrative purposes as opposed to that collected by a specially trained research staff for the purpose of investigating a particular hypothesis. Teresi and Holmes 1992, for instance, have questioned the research quality and, therefore, the appropriateness for scientific investigation of routinely collected MDS data. To address this issue, Phillips and Morris 1997 compared the reliability and predictive validity of MDS measures of cognition and ADLs for four databases, one collected as part of a formal research study and the other three collected as part of the routine clinical/ administrative procedures of three states. The Cronbach's alphas for these two measures varied little between the research and clinical/administrative databases, as did the correlations between the cognition and ADLs measures, indicating the appropriateness of using MDS clinical/administrative databases for research purposes.

MDS+ assessments at Shorefront and MJG were completed by nurses and other staff who were specially trained for that purpose. Further, at each facility an MDS coordinator reviewed all MDS+ assessments to ensure completeness and accuracy.

Measures
A comprehensive MDS+ assessment was completed at admission, annually, and in response to a significant change in a resident's status. In addition, quarterly reviews were completed every 3 months on a large subset of the MDS+ items. To have the requisite data to measure change in health status over time, only those items included in the quarterly assessment were used.

MDS+ items recorded quarterly were further screened in terms of their reported psychometric properties. Items suggested by the literature to have questionable reliability and/or validity, such as those pertaining to mood, problem behavior, social engagement, and continence, were excluded.

To operationalize cognitive status, the MDS Cognitive Performance Scale (CPS; Morris et al. 1994) was used. The CPS, based on five MDS items (comatose status, ability to make decisions, short-term memory, making self understood, and self-performance in eating), classifies residents into seven categories: intact (0), borderline intact (1), mild impairment (2), moderate impairment (3), moderately severe impairment (4), severe impairment (5), and very severe impairment (6). The CPS has been found to explain 74% and 75% of the variance, respectively, in Mini-Mental Status Exam (MMSE) scores and scores derived from a combination of the MMSE and the Test for Severe Impairment (Morris et al. 1994). When assessed against the MMSE as a criterion variable, the CPS showed a sensitivity of 94%, a specificity of 94%, and a diagnostic accuracy of 96% (Hartmaier et al. 1995). More recently, Gruber-Baldini, Zimmerman, Mortimore, and Magaziner 2000 investigated the construct validity of the CPS when completed by regular nursing home personnel, as opposed to members of a research-trained staff. Accordingly, they found the CPS had a .65 correlation with the MMSE and a .63 correlation with the Psychogeriatric Dependency Rating Scale. Mirotznik and Los Kamp 2000, using a sample of Shorefront and MJG residents, assessed the association of the CPS measure with diagnosed dementia, ADLs, and mortality. After collapsing the seven CPS categories into three broader groupings, intact (0–1), moderate impairment (2–3), and severe impairment (4–6), they observed that level of cognitive impairment was, as expected, linearly associated with each criterion variable. Specifically, although 15.9% of intact residents had a diagnosis of dementia, the respective figures for moderately and severely cognitively impaired residents were 53.4% and 85.9%. Similarly, as cognitive impairment increased, mean ADL dependence scores increased from 1.6 to 2.3 to 3.4, as did mortality rates, 9.8% to 14.0% to 23.0%. Following Mirotznik and Los Kamp 2000, the current study also collapsed the seven CPS levels into three broader categories.

ADLs consisted of five items concerning bed mobility, transfer ability, locomotion, dressing, and toilet use, each of which was scored from 0, indicating independence, to 4, indicating total dependence. Responses to the items were summed and divided by the number of answered items. The Cronbach's alpha for the ADLs measure was .93. Mirotznik and Los Kamp 2000 reported an alpha for ADLs of .94.

Psychological well-being, derived from the "Sense of Initiative/Involvement" section of the MDS+, consisted of seven items (e.g., "at ease interacting with others," "at ease doing planned or structured activities"), each coded in the current study so that 0 = yes and 1 = no. All seven items were summed for a composite score that ranged from 0 to 7. The Cronbach's alpha for this scale was .68. Mirotznik and Los Kamp 2000 reported an alpha for psychological well-being of .73.

Physical health status was operationalized on the basis of several MDS+ items abstracted from patients' medical records. Specifically, measures were constructed of the number of (a) diagnosed diseases, (b) conditions/signs/symptoms, (c) administered medications, (d) emergency room transfers, and (e) hospital admissions. Data were also available on whether residents expired during the study period.

Data were abstracted on the following covariates as well: age, gender, length of stay, Medicare or Medicaid payment source for nursing home stay, primary language (English or other), marital status (ever or never married), and race (White or non-White).

Analytic Approach
I used event history analysis for all eight outcome variables. This approach made use of all of the assessment data for each individual, and it allowed the modeling of the effects of relocation status, cognitive status, and the interaction of Relocation Status x Cognitive Status while controlling for baseline values of outcome variables as well as of extraneous covariates. Specifically, for mortality I conducted a continuous survival analysis using Cox regression. As death was recorded as occurring on a specific day, a relatively exact measurement, it was reasonable to model time to death as a continuous hazard function. In contrast, the timing of changes in ADLs, psychological well-being, and the five physical health status measures was not precisely known. Rather, these changes were essentially recorded as occurring during quarter- yearly periods. If an event either occurs or is measured at discrete time intervals, it is not appropriate to use continuous survival analysis. Doing so leads to errors in the estimate of the hazard ratio, referred to as time-aggregation bias (Petersen 1991). As such, for the morbidity outcome measures I conducted discrete-time survival analysis using logistic regression (Allison 1982; Singer and Willett 1991). Such an approach correctly modeled time to decline as occurring in quarter-yearly intervals.

As implied above, the time-to-event for mortality was defined as number of days from the date of the baseline assessment. For the experimental group, the MDS+ immediately preceding their intrabuilding move constituted the baseline assessment. For the comparison group, the first MDS+ completed during the study period constituted the baseline assessment. For mortality, failure was defined as the occurrence of death.

For the seven morbidity outcome measures, time to event was defined as the number of 3-month intervals from the baseline MDS+ assessment, given that MDS+ assessments were typically conducted quarterly and hence morbidity events were recorded quarterly. Because the study extended over 4 years, there were 16 such time intervals. For the morbidity outcome measures, failure was defined as a decline in a resident's health status from baseline assessment. Decline was operationalized as a one-unit increase in a scale or count score (Phillips, Sloane, et al. 1997). So a resident whose ADLs score or number of administered medications had changed since baseline from 3 to 4 exhibited such a decline. Residents whose baseline ADLs and psychological well-being scores indicated they could not decline further were excluded from the analysis of these two outcome variables.

Residents who did not die or decline from baseline or who exited from the two facilities were considered censored at the time of their last assessment. The 1-year exit rate was found to be associated with relocation status (i.e., nonmovers were more likely to exit within 1 year from baseline than movers) but not with cognitive status or with the Relocation Status x Cognitive Status interactions. The exit rates for the remaining years of the study period were found to be unrelated to relocation status, cognitive status, or their interactions.

In Step 1 of the Cox regression for mortality, relocation (movers vs nonmovers), cognitive status (represented by two dummy variables for moderate and severe impairment), and the two Relocation Status x Cognitive Status interaction terms were entered. To address the issue of resident selection bias, in Step 2 the covariates that reflected baseline health status and initial differences between movers and nonmovers were entered. The proportional hazard assumption was evaluated by entering in Step 3 the interaction terms representing the product of each independent variable with survival time. Significant Covariate x Survival Time interactions were to be retained in the final version of the equation, thereby adjusting for violations of the proportional hazard assumption (Kleinbaum 1996; Singer and Willett 1991). Significant interactions between theoretically central independent variables (relocation status, cognitive status, and their interactions) and survival time were to be further explored by segmenting time into two periods, the year immediately following baseline (the short-term adjustment period) versus the next 3 years (the long-term adjustment period) and testing the relative risk of death within each period separately (Kleinbaum 1996). So the interaction terms representing the product of, say, relocation status with survival time tested whether relocation was associated with mortality similarly in the first year after baseline as it was in the next 3 years.

A separate logistic regression was run for each morbidity outcome measure. These regressions followed the same steps as the analysis for mortality, with one exception. Before entering the theoretically central independent variables, the outcome variable was regressed on 16 dummy variables, each signifying a specific 3-month interval of the study's 4-year follow-up period. In discrete-time survival analysis, the regression coefficients for these time-dummy variables, without any substantive predictors entered into the equation, represent the risk profile over time for the sample as a whole. Substantive predictors then entered are tested by determining if they effect a vertical shift in the overall profile (Singer and Willett 1991).

Measuring relocation effects in terms of eight distinct outcome variables inflated the study's investigationwise Type I error rate beyond the conventional .05 level, increasing the likelihood of yielding a significant result when in fact nothing but random variation was at work. To adjust for this, any effects that were significant at the .05 level were subjected to a Bonferroni correction and then assessed for significance at the (.05/8) = .006 level (Harris 2001).

	Results

TOP Abstract Theoretical Models Methods Results Discussion References

 
Of the total sample of 776 residents, 74% were female and 97% were White, with a mean age of 83 and a 2.3-year length of stay. Seventy-seven percent spoke English as their primary language. For most, the source of payment for their nursing home stay was a combination of Medicaid and self-pay/private insurance (72% and 69%, respectively). Only 8% had their stays initially covered by Medicare. Thirteen percent had never married, 22% were still married, 60% were widowed, and 5% were separated or divorced. Twenty-five percent were cognitively intact, 31% were moderately cognitively impaired, and 44% were severely impaired. About 49% needed extensive assistance or were totally dependent in terms of ADLs. The most frequent diagnoses were arteriosclerotic heart disease (42%), cataracts (34%), Alzheimer's disease (33%,) dementias other than Alzheimer's disease (30%), anemia (23%), arthritis (22%), cerebrovascular accident (18%), congestive heart failure (15%), and diabetes mellitus (15%).

As seen in Table 1 , residents who underwent a room change and those who did not were similar in terms of age, gender, race, marital status, cognitive status, length of stay, psychological well-being, number of conditions/signs/symptoms, number of administered medications, number of emergency room transfers, having Medicaid as a payment source for their nursing home stay, and speaking English as a primary language. Movers were more likely to have had a baseline MDS+ assessment in 1995, 1996, and 1997; nonmovers, in 1993 and 1994. Movers were somewhat less likely to have their stay initially paid by Medicare. They were more ADL independent, had a greater number of diagnosed diseases, and were more likely to have had a hospital admission. Last, movers were more likely to be residents at Shorefront Nursing Home and to have undergone the interbuilding move previously reported on by Mirotznik and Los Kamp 2000.

	Discussion

TOP Abstract Theoretical Models Methods Results Discussion References

 
A number of important questions remain unanswered in the relocation stress literature. This study addressed one such question, namely, does cognitive status moderate the health effects of relocation? Although several articles have been published on the cognitive status vulnerability hypothesis, those studies have either exhibited methodological limitations or yielded conflicting results. Further, few studies have investigated this hypothesis with regard to single-person, intrabuilding room transfers.

To test the vulnerability hypothesis, two interaction terms, Relocation Status x Moderate Cognitive Impairment and Relocation Status x Severe Cognitive Impairment, were constructed. Each term was tested against eight outcome variables. Thus, the vulnerability hypothesis was given 16 opportunities to emerge. Only one of those interaction effects reached the .05 significance level. The pattern of the effect, however, indicated that cognitively intact movers exhibited a reduced risk of mortality, and severely cognitively impaired movers exhibited no difference in their risk of dying. Further, this effect failed to remain significant after correcting for an inflated investigationwise Type I error rate. It would be worth testing in future studies if this particular interaction was in fact a chance occurrence. If the interaction replicates, that would suggest cognitive status does moderate relocation effect, but in a way quite different than expected. In either case, this study found no evidence, in spite of the number of interactions tested, to suggest that individuals with either severe or moderate cognitive impairment are at greater risk of harmful health effects during single-person, intrabuilding room transfers. Indeed, none of the remaining 15 interaction terms investigated reached significance even at the .05 level. As noted above, Mirotznik and Los Kamp 2000, in an earlier investigation at Shorefront and MJG, found that cognitive status did not moderate relocation outcome in a mass interbuilding transfer. Thus, the current study yielded results consistent with this earlier research.

It was mentioned above that MDS items, especially those based on subjective assessment, are less reliable for cognitively impaired residents. This might have attenuated the Relocation Status x Cognitive Status interaction effects, particularly for ADLs and psychological well-being. However, the outcome variables operationalized on more objective medical record data (i.e., number of administered medications, hospital admissions, etc.) also failed to exhibit significant interaction effects.

Given several theories that suggest that cognitively impaired individuals would experience deleterious effects postrelocation, the present study focused on the impact and long-term adjustment stages of the relocation process. Furthermore, documentation was not available as to exactly when each resident was notified about the upcoming transfer, so it was not possible to test for anticipatory effects. It would be interesting to investigate, though, if cognitive status moderates outcome in anticipation of room transfer. A number of studies have provided evidence to suggest that residents in general may experience stress in anticipation of relocation (Tobin and Lieberman 1976; Zweig and Csank 1976). Furthermore, some of the theoretical models discussed earlier would suggest that cognitively impaired residents may be less likely than those who are intact to experience such anticipatory stress. In general, residents at Shorefront and MJG were given short notice of room changes. An additional question worth investigating is whether the number of days of notice given for a future room transfer affects outcome (i.e., if longer or shorter pretransfer periods are more stressful)? In the current study, documentation regarding the specific reasons residents were transferred to a different room was also unavailable. Yet those reasons might theoretically influence outcome. As such, this issue also warrants attention. An additional limitation of this study involved the use of the MDS+. Outcome measures were constructed on the basis of those MDS+ items that previous research has found to be psychometrically sound. Consequently, this study did not investigate whether cognitive status moderated the effects of room changes on mood, problem behavior, social engagement, and continence. It would be worthwhile to investigate if cognitively impaired individuals are particularly susceptible to deleterious changes in these outcome measures during room transfers.

What implications do this study's results have for the transfer of cognitively impaired residents into SCUs? Before addressing this question, one qualification must be stated. It is possible that cognitively impaired residents selected into SCUs may differ from those who remain on nonspecialized units. Although the literature is inconsistent, studies have found differences in cognitive functioning, ADLs, problem behaviors, and so forth. (Holmes et al. 1990; Riter and Fries 1992). To the degree that cognitively impaired residents selected for SCUs constitute a unique subset, they may in fact respond to transfer in unique ways. With this caveat and the study's limitations in mind, the current findings would suggest that cognitively impaired residents may not be at unusual risk of experiencing harmful effects as a consequence of being transferred into SCUs. But clearly our knowledge would be greatly extended if future research would study more directly the impact of transferring cognitively impaired residents into SCUs.

	Acknowledgments

 
This study was supported by the Alzheimer's Association (Grant RG3-96-003). I thank the administration of Metropolitan Jewish Health System for its cooperation and Martin Piccochi for data management.

Received for publication July 19, 2001. Accepted for publication March 28, 2002.

	References

TOP Abstract Theoretical Models Methods Results Discussion References

 

• Achterberg W., Pot A. M., van Campen C., Ribbe M., 1999. Residential Assessment Instrument (RAI): A review of international research on the psychometric qualities and effects of implementation in nursing homes. Tijdschrift Voor Gerontologie En Geriatrie 30:264-70. [Medline]
• Aldrich C. K., 1964. Personality factors and mortality in the relocation of the aged. The Gerontologist 4:92-93.
• Allison P. D., 1982. Discrete-time methods for the analysis of event histories. Leinhardt S., , ed.Sociological methodology 61-98. Jossey-Bass, San Francisco.
• Borup J. H., 1981. Relocation: Attitudes, information, network and problems encountered. The Gerontologist 21:501-511.
• Bourestom, N., & Pastalan, L. (1975). Final report, forced relocation: Setting, staff and patient effects. Institute of Gerontology, University of Michigan, Ann Arbor.
• Bourestom, N., & Pastalan, L. (1981). The effects of relocation on the elderly: A reply to Borup, J. H., Gallego, D. T., & Heffernan, P. G. The Gerontologist, 21, 4–7.
• Campbell R. I., Hudson C. M., 1985. Synthetic cohorts from panel surveys: An approach to studying rare events. Research on Aging 7:81-93. [Abstract]
• Casten R., Lawton M. P., Parmelee P. A., Kleban M. H., 1998. Psychometric characteristics of the Minimum Data Set I: Confirmatory factor analysis. Journal of the American Geriatrics Society 46:726-735. [Medline]
• Coffman T. L., 1983. Toward an understanding of geriatric relocation. The Gerontologist 23:453-459.
• Conroy J. W., Adler M., 1998. Mortality among Penhurst class members, 1978 to 1989: A brief report. Mental Retardation 36:380-385. [Medline]
• Crooks V. C., Schnelle J. F., Ouslander J. P., McNees M. P., 1995. Use of the Minimum Data Set to rate incontinence severity. Journal of the American Geriatrics Society 43:1363-1369. [Medline]
• Csank J. Z., Zweig J. P., 1980. Relative mortality of chronically ill geriatric patients with organic brain damage before and after relocation. Journal of the American Geriatrics Society 28:76-83. [Medline]
• Frederiksen K., Tariot P., De Jonghe E., 1996. Minimum Data Set Plus (MDS+) scores compared with scores from five rating scales. Journal of the American Geriatrics Society 44:305-309. [Medline]
• Friedman S. M., Williamson J. D., Lee B. H., Ankrom M. A., Ryan S. D., Denman S. J., 1995. Increased fall rates in nursing home residents after relocation to a new facility. Journal of the American Geriatrics Society 43:1237-1242. [Medline]
• Gambassi G., Landi F., Peng L., Brostrup-Jensen C., Calore K., Hiris J., et al. 1998. Validity of diagnostic and drug data in standardized nursing home resident assessments: Potential for geriatric pharmaco-epidemiology. Medical Care 36:167-179. [Medline]
• Goldfarb A. I., Shahinian S. P., Burr H. T., 1972. Death rate of relocated nursing home residents. Kent D. P., Kastenbaum R., Sherwood S., , ed.Research planning and action for the elderly: The power and potential of social science 525-537. Behavioral Publications, New York.
• Graney J. G., Engle V. F., 2000. Stability of performance of activities of daily life using the MDS. The Gerontologist 40:582-586. [Abstract/Free Full Text]
• Gruber-Baldini A. L., Zimmerman S. I., Mortimore E., Magaziner J., 2000. The validity of the Minimum Data Set in measuring the cognitive impairment of persons admitted to nursing homes. Journal of the American Geriatrics Society 48:1734-1736. [Medline]
• Guttman G. H., Herbert C. P., 1976. Mortality rates among relocated extended-care patients. Journal of Gerontology 31:352-357.
• Harris R. J., 2001. A primer of multivariate statistics Erlbaum, Mahwah, NJ.
• Hartmaier S. L., Sloane P. D., Guess H. A., Koch G. G., Mitchell C. M., Phillips C. D., 1995. Validation of the Minimum Data Set Cognitive Performance Scale: Agreement with the Mini-Mental Status Examination. Journal of Gerontology: Medical Sciences 50A:M128-M133.
• Hawes C., Morris J. N., Phillips C. D., Mor V., Fries B. E., Nonemaker S., 1995. Reliability estimates for the Minimum Data Set for nursing home resident assessment and care screening (MDS). The Gerontologist 35:172-178. [Abstract]
• Holmes D., Teresi J., Weiner A., Monaco C., Ronch J., Vickers R., 1990. Impacts associated with special care units in long-term care facilities. The Gerontologist 30:178-183. [Abstract]
• Holzapfel S. K., Schoch C. P., Dodman J. B., Grant M. M., 1992. Responses of nursing home residents to intrainstitutional relocation. Geriatric Nursing July: (August) 192-195.
• Kleinbaum D. G., 1996. Survival analysis Springer, New York.
• Kovach C. R., 1998. Nursing home dementia care units: Providing a continuum of care rather than aging in place. Journal of Gerontological Nursing 24:30-36.
• Kowalski C. N., 1981. Institutional relocation: Current programs and applied approaches. The Gerontologist 21:512-519.
• Kral V. A., Grad B., Berenson J., 1968. Stress reactions resulting from relocation of an aged population. Canadian Psychiatric Association Journal 13:201-209. [Medline]
• Lander S. M., Brazill A. L., Ladrigan P. M., 1997. Intrainstitutional relocation. Effects on residents' behavior and psychosocial functioning. Journal of Gerontological Nursing 23:35-41. [Medline]
• Lawton M. P., Simon F., 1968. The ecology of social relationships in housing for the elderly. The Gerontologist 8:108-115. [Medline]
• Leon J., 1994. The 1990/1991 national survey of special care units in nursing homes. Alzheimer's Disease and Associated Disorders 8:S72-S86.
• Leon J., Cheng C. K., Alvarez R. J., 1997. Trends in special care: Changes in SCU from 1991 to 1995 ('95/96 TSC). Journal of Mental Health and Aging 3:149-168.
• Leon J., Moyer D., 1999. Potential cost savings in residential care for Alzheimer's disease patients. The Gerontologist 39:440-449. [Abstract]
• Lieberman M. A., Tobin S. S., 1983. The experience of old age Basic Books, New York.
• Mallick M. J., Whipple T. W., 2000. Validity of nursing diagnosis of relocation stress syndrome. Nursing Research 49:97-100. [Medline]
• Markus E., Blenkner M., Bloom M., Downs T., 1972. Some factors and their association with post-relocation mortality among institutionalized aged persons. Journal of Gerontology 27:376-382.
• Marlowe, R. A. (1974, February). When they closed the doors at Modesto. Paper presented at the National Institute on Mental Health Conference on the closure of state hospitals, Scottsdale, AZ.
• McAuslane L., Sperlinger D., 1994. The effects of relocation on elderly people with dementia and their nursing staff. International Journal of Geriatric Psychiatry 9:981-984.
• Miller D., Lieberman M. A., 1965. The relationship of affect state and adaptive capacity to reactions to stress. Journal of Gerontology 20:492-497. [Medline]
• Mirotznik J., 1995. Response of noninterviewable long-term care patients before and after interinstitutional relocation. Psychological Reports 76:1267-1280. [Medline]
• Mirotznik J., Los Kamp L., 2000. Cognitive status and relocation stress: A test of the vulnerability hypothesis. The Gerontologist 40:531-539. [Abstract/Free Full Text]
• Mor V., Branco K., Fleishman J., Hawes C., Phillips C., Morris J., et al. 1995. The structure of social engagement among nursing home residents. Journal of Gerontology: Psychological Sciences 50B:P1-P8. [Abstract]
• Morris J. N., Fries B. E., Mehr D. R., Hawes C., Phillips C. D., Mor V., et al. 1994. MDS cognitive performance scale. Journal of Gerontology: Medical Sciences 49:M174-M182.
• Morris J. N., Hawes C., Fries B. E., Phillips C. D., Mor V., Katz S., et al. 1990. Designing the national Residential Assessment Instrument for nursing homes. The Gerontologist 30:293-307. [Abstract]
• Nirenberg T. D., 1983. Relocation of institutionalized elderly. Journal of Consulting and Clinical Psychology 51:693-701. [Medline]
• Ogren E. H., Linn M. W., 1971. Male nursing home patients: Relocation and mortality. Journal of the American Geriatrics Society 19:229-239. [Medline]
• Petersen T., 1991. Time-aggregation bias in continuous-time hazard-rate models. Marsden P. V., , ed.Sociological methodology 263-290. American Sociological Association, Washington, DC.
• Phillips C. D., Chu C. W., Morris J. N., Hawes C., 1993. Effects of cognitive impairment on the reliability of geriatric assessments in nursing homes. Journal of the American Geriatrics Society 41:136-141. [Medline]
• Phillips C. D., Morris J. N., 1997. The potential for using administrative and clinical data to analyze outcomes for the cognitively impaired: An assessment of the Minimum Data Set for nursing homes. Alzheimer Disease and Associated Disorders 11:162-167.
• Phillips C. D., Sloane P. D., Hawes C., Koch G., Han J., Spry K., et al. 1997. Effects of residence in Alzheimer disease special care units on functional outcomes. Journal of the American Medical Association 278:1340-1344. [Abstract/Free Full Text]
• Pruchno R. A., Resch N. I., 1988. Intrainstitutional relocation: Mortality effects. The Gerontologist 28:311-317. [Medline]
• Reingold R., Werner R., 1994. The evolution of specialized dementia care: The relationship between research and program planning. Alzheimer Disease and Associated Disorders 8:S126-S132.
• Resnick N. M., Brandeis G. H., Baumann M. M., Morris J. N., 1996. Evaluating a national assessment strategy for urinary incontinence in nursing home residents: Reliability of the Minimum Data Set and validity of the resident assessment protocol. Neurology Urodynamics 15:583-598.
• Riter R. N., Fries B. E., 1992. Predictors of the placement of cognitively impaired residents on special care units. The Gerontologist 32:184-190. [Abstract]
• Robertson C., Warrington J., Eagles J. M., 1993. Relocation mortality in dementia: The effects of a new hospital. International Journal of Geriatric Psychiatry 8:521-525.
• Rogers J. C., Stuart M. R., Sheffield P., Swee D. E., Formica P., 1990. Functional health status of relocated nursing home residents. Journal of the American Board of Family Practice 3:157-172.
• Schnelle J. F., Wood S., Schnelle E. R., Simmons S. F., 2001. Measurement sensitivity and the Minimum Data Set depression quality indicator. The Gerontologist 41:401-405. [Abstract/Free Full Text]
• Schulz R., Brenner G., 1977. Relocation of the aged: A review and theoretical analysis. Journal of Gerontology 32:323-333.
• Singer J. D., Willett J. B., 1991. Modeling the days of our lives: Survival analysis when designing and analyzing longitudinal studies of duration and the timing of events. Psychological Bulletin 110:268-290.
• Teresi J. A., Holmes D., 1992. Should MDS data be used for research?. The Gerontologist 32:148-149. [Medline]
• Thorson J. A., Davis R. E., 2000. Relocation of the institutionalized aged. Journal of Clinical Psychology 56:131-138. [Medline]
• Tobin S. A., Lieberman M. A., 1976. Last home for the aged Jossey-Bass, San Francisco.
• van Dijk P. T., Dippel D. W., Habbema J. D., 1991. Survival of patients with dementia. Journal of the American Geriatrics Society 39:606-610.
• Zweig J. P., Csank J. Z., 1976. Mortality fluctuations among chronically ill medical geriatric patients as an indicator of stress before and after relocation. Journal of the American Geriatrics Society 24:264-277. [Medline]

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