| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to John F. Schnelle, PhD, JHA/UCLA Borun Center, 7150 Tampa Avenue, Reseda, CA 91335. E-mail: jschnell{at}ucla.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: MDS • Quality indicator • Physical restraint
Despite the lack of reliable information about these important quality issues, a prevalence of restraint quality indicator (QI) is calculated on the basis of Minimum Data Set (MDS) data generated by all NHs and provided to consumer groups, NHs, and federal and state survey teams under the assumption that differences in prevalence rates between facilities may reflect differences in quality of care (Nursing Home Compare, 2002). However, there has been no independent verification that this QI reflects different rates of restraint use or differences in quality of care in areas relevant to restraint use (e.g., exercise to prevent mobility decline). Studies have reported that the restraint QI is stable, and one recent study reported that it was valid (Morris et al., 2002; Zimmerman et al., 2002). However, this validity study did not report the measures that led to its conclusion, nor did it address issues relevant to the accuracy of MDS items that had been raised (U.S. General Accounting Office, 2002b).
This study fills this gap in knowledge by addressing two questions. First, are there differences in care processes or use of physical restraints (e.g., any device that potentially limits a resident's movement) between NHs that score high (in the upper 75th percentile) on the prevalence of restraint QI and those that score low (in the lower 25th percentile)? Second, are NHs that report a low prevalence of restraint use more likely to implement care processes that reflect better assessment and management of gait and balance problems or enhanced quality of care than NHs that report high prevalence of restraint use?
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
All residents not in transitional care (Medicare coverage) were eligible to participate. Consents were obtained from 144 residents (35% of eligible residents) in high-restraint homes and 269 residents (40%) in low-restraint homes. All homes were staffed according to industry standards, with approximately 7 to 9 residents to one aide on the 7 a.m. to 3 p.m. shift and 10 to 13 residents to one aide on the 3 p.m. to 11 p.m. shift. There were no differences between the two groups of homes in total nursing hours per resident per day reported to the State in 2000, the last year that such staffing data were available (3.4 hr vs. 3.2 hr for low and high homes, respectively; California Office of Statewide Health Planning and Development, 2002). Research staff collected data in all 14 NHs between July 2001 and May 2002 and were blinded as to the homes' designation as low- or high-restraint prevalence facilities.
Overview and Indicators
Trained research staff conducted medical record reviews, resident interviews, direct observations, and physical performance evaluations during three consecutive weekdays (12 hr per day, 7 a.m. to 7 p.m.) in each home and with all consented residents. Demographic information was retrieved from each participant's medical record. MDS information was retrieved from the most recently completed MDS assessment at the time of the study. Specific care processes related to restraint management, gait and balance problems, and validated by RAND in the Assessing Care of Vulnerable Elderly (ACOVE) project were scored on the basis of data from the medical record, resident interviews, and observational data and are shown in Table 1. In brief, in the ACOVE project, care processes that were empirically related to positive outcomes or recommended in practice guidelines were converted into indicators that were validated by expert consensus methodology as important quality measures. The methodology used to develop all indicators shown in Table 1 has been described elsewhere (ACOVE, 2001; Saliba et al., 2002; Shekelle, MacLean, Morton, & Wenger, 2001; Wenger & Shekelle, 2001).
|
Indicators 1–7 in Table 1 were scored on the basis of medical record data. The medical record review protocol was derived from a longer medical record review designed to measure NH quality of care (Saliba, Cadogan, & Roth, 2002). The first four indicators in Table 1 relate to either physical or chemical restraint management and assess whether restraints are being appropriately justified and whether their side effects (low physical activity) are being managed. Indicators 5–7 relate to the assessment or treatment of gait and mobility problems by primary care providers. The logic of scoring these indicators was that better assessment by primary care providers might lead to more appropriate preventive interventions for the gait and mobility problems that justify restraint use or more useful information for determining whether restraints are needed. Participants received two scores for Indicator 3 (repositioning for restrained residents): one score using thigh monitor data (discussed in following paragraphs) and one score using medical record data. Participants received a passing score on this indicator based on medical record data if the chart contained documentation that repositioning was provided every 2 hr.
The second column of Table 1 identifies the specific medical record data source used to score each care process and the associated scoring rules developed in this project so that acceptable interrater reliability could be obtained. Interrater reliability for scoring the medical record-based indicators ranged from kappa values of.61 to.76 (p <.001) for all indicators except Indicator 4 (
=.48; p <.05).
Wireless Thigh Monitor Movement Observation Data
Observational data relevant to residents' physical movements were obtained from a wireless monitor worn on the thigh that measures horizontal and vertical orientation with respect to gravity every 4 s. In preliminary research for this study we determined that, when an individual is repositioned in bed, the monitor registers a minimum 40° move in the horizontal position of the thigh followed by maintenance of at least a 20° change in the horizontal position. When an individual is repositioned in a chair, the monitor registers at least two 40° vertical changes.
The monitor also enabled us to detect activities that involved sustained resident movement for at least 6 min and, thus, could possibly reflect exercise. Because we could not discriminate exercise care processes (e.g., walking assistance) from care processes that involved physical movement for other reasons (e.g., incontinence care), we characterized all resident movements that were sustained for at least 6 min as "activity episodes" possibly related to exercise. We used the thigh monitor technology because preliminary data indicated that any observational schedule feasible for a human observer to implement with more than three residents would underestimate care episodes such as walking and repositioning that occur infrequently (e.g., less than every 2 hr) and are relatively brief in duration.
We used thigh monitor movement data to score Indicator 3 and to provide data on activity and exercise. We calculated two movement measures from the graphs produced by the thigh monitor. First, we calculated the number of physically restrained participants who experienced at least one 4-hr or longer time interval with no movement. Participants meeting this measurement criterion failed to be released and repositioned on a 2-hr basis, as federal regulations stipulate for physically restrained residents (see Table 1, Indicator 3, thigh monitor scoring rule). We chose a 4-hr interval as opposed to a 3-hr period to allow NH staff greater flexibility. Participants passed Indicator 3 according to the thigh monitor if their longest time period without movement was shorter than 4 hr.
We also calculated the total number of movement patterns that lasted at least 6 min per hour for each resident in an effort to determine whether there were differences between high- and low-restraint homes in the provision of care processes that could be interpreted as exercise (activity episode). These data are presented as a continuous measure (average activity episode frequency per hour) as opposed to a discrete "pass–fail" measure. Participants who could stand and walk independently according to a performance evaluation (described later in this section) were not asked to wear the thigh monitor because we could not discriminate their self-initiated physical activities from activities or care episodes initiated by NH staff. Interrater agreement on interpretation of the wireless thigh monitor movement graphs produced kappa statistics of.61 for repositioning movements,.82 for activity episodes while in a chair, and.75 for activity episodes while in bed (all p <.001).
Direct Observation Data
Research staff attempted to observe all participants for 1 min each hour between 7 a.m. and 7 p.m. on a single day. The participant's location (bed or chair), engagement in a group activity or with another person, and all the devices that could potentially limit movement were recorded. We use the word potential because NH staff document that residents can release some restraining devices (e.g., seat belts) and, hence, these devices are not considered restraints. Given this, we evaluated each participant's ability to release the restraint device through a physical performance evaluation (described later in this section). We also noted whether residents were sitting in wheelchairs with the foot pedals elevated. In a recently completed exercise trial, we noted that many residents cannot move foot pedals to allow standing, transferring, or propelling the wheelchair with their feet; this is a common method of wheelchair locomotion in this population, who have poor upper body strength (Simmons, Schnelle, MacRae, & Ouslander, 1995). Therefore, we believe that wheelchairs with elevated foot pedals meet the MDS restraint definition of "a chair that limits movement." Pedals were considered "elevated" only if they were moved up from a 90° angle perpendicular to the floor. In the analysis section of this article, we report restraint data with and without elevated foot pedals because foot pedals may not be regarded as a conventional restraining device.
We repeated hourly observations of a subsample of 128 participants on a second day in both low-restraint (n = 86) and high-restraint (n = 42) homes in order to assess the stability of the direct observation data. All residents who wore the thigh monitor (i.e., were physically restrained) and a random sample of others were observed on the 2 days. Pearson product–moment correlation coefficients for percentage of observations that a participant was either restrained, in bed, or engaged were.48,.79, and.40, respectively (p <.001), with no differences between the two groups of homes. We used kappa statistics to assess the stability of the proportion of residents who were restrained on Day 1 and Day 2. Kappa agreement statistics were better in the high-restraint homes for in-bed restraints, indicating a more consistent use of restraints. The kappa statistics for in-bed and chair restraint use in the high-prevalence homes was.64 and.65, respectively (p <.001). By comparison, low-prevalence homes showed a kappa statistic of.65 for chair restraints but only.32 for in-bed restraints. Although the influence of the base rate on kappa is well-known, a review of the data suggests that the lower kappa for the in-bed restraints observed in the low-prevalence homes reflects higher variability in the use of in-bed restraints because the number of residents observed in bed was relatively high, both in high- and low-restraint facilities. Although statistically significant, differences between high and low homes on the number of residents observed in bed should not seriously distort the kappa agreement statistic.
Direct observations were also completed during two separate mealtimes in order to document the amount of feeding assistance and the frequency of social interaction or verbal prompting residents received from staff. The initial reliability of the mealtime observational protocol was estimated with seven different observers and 94 residents in two NHs immediately prior to this project. The same personnel conducted the observations in this study, and reliability was monitored by conducting simultaneous observations with one or two residents per home. Initial reliability correlations ranged from.88 to.99, and periodic reliability checks were within this range. The mealtime observational protocol is more specifically described elsewhere (Simmons, Babineau, Garcia, & Schnelle, 2002).
Physical Performance Evaluation
We used a standardized physical performance evaluation to evaluate each participant's ability to stand, walk, and release restraints. The reliability and validity of this protocol are reported elsewhere (Schnelle et al., 2002). Briefly, research staff asked participants to stand and provided graduated levels of assistance ranging from no assistance (resident able to stand on command) to full physical lift. If a resident had any device that could potentially limit movement, he or she was asked to release this device prior to standing and given graduated prompts to do so. The performance evaluation was not completed on all residents because many residents were in bed during periods when the evaluation could occur (e.g., when residents were not at meals or being bathed).
Participant Interview Data
Participants with an MDS recall score of 2 or greater were targeted for interview because there is evidence that these residents can accurately describe the care processes that they receive (Simmons & Schnelle, 2001). We asked these residents how frequently they both received and preferred to receive walking assistance. A discrepancy score to estimate unmet need was calculated on the basis of residents' self-reports. For example, if a resident preferred three walking assists per day and received only two, a discrepancy score of –1 would reflect unmet need and the resident would be scored as failing Indictor 8, which pertains to physical activity. The usefulness of this procedure to describe unmet needs has been previously described (Simmons & Schnelle, 1999). As a way to assess stability, interviews were repeated on a subsample of 34 residents within 24 hr. The correlation between residents' reports of assistance received and preferred was.56 and.60, respectively (p <.001). The kappa agreement as to whether residents' walking assistance needs were met based on the discrepancy score was.72 (p <.001).
Data Analysis
We used chi-square analyses to compare the proportion of participants within the lower and upper quartile QI groups who received a "pass" score for each care process indicator. The standard error was adjusted for clustering by using survey data analysis (SUDAAN; Research Triangle Institute, 1996). SUDAAN multiplies the standard errors by design effects that are based on a comparison from the study's clustered sample with those from a simple random sample.
It was not possible to directly assess the accuracy of the QI reports of restraint prevalence in this study because the indicator counts only daily use of restraints for the 7 days prior to the MDS assessment date. Direct observations were not conducted for 7 consecutive days or during the period immediately prior to the MDS assessment. In addition, MDS documentation of full bed rails, which is not considered in the restraint prevalence QI but is considered a restraint on the MDS, does not distinguish between daytime and nighttime use. In this study, observational data were collected between 7 a.m. and 7 p.m. on only two separate days. We provide two separate statistics to estimate the degree to which restraints were used in high- and low-prevalence homes according to the MDS-based QI. First, we report a "restraint prevalence statistic" by calculating the proportion of all participants who we observed to be restrained at least once during an average of 10 hr of direct observation per subject between 7 a.m. and 7 p.m. This overall prevalence statistic was also adjusted based on estimates of how many participants could release restraints, as documented by our performance test. In addition, we also report the percentage of all observations throughout the day that found participants to be restrained. This statistic combines the number of participants who were restrained with the amount of time that they were observed to be restrained. None of these measures directly assess the accuracy of the restraint prevalence QI, but all of them address the related question of whether restraints were more commonly used with participants in the upper quartile (high-prevalence) NHs compared with participants in the lower quartile (low-prevalence) NHs at one point in time.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
|
2 = 3.7; p <.05). A significant difference also occurred in the percentage of all observations showing participants to be restrained in the low- versus high-prevalence homes, 46% versus 57%, respectively (t = 2.7; p <.001). This difference was due to differences in how much time participants spent in bed and in the use of bed restraints between the two groups of homes.
We observed participants in bed on a greater number of observations in the high-restraint homes compared with the low-restraint homes (44% vs. 33%; t = 2.3 and p <.001). There was no significant difference in the proportion of participants who were observed in bed with full side rails between the high- and low-restraint homes (22% vs. 16%, respectively). However, there was a significant difference in the proportion of participants observed with any type of side rail (i.e., full and partial rails) while in bed (74% in high-prevalence homes vs. 64% in low-prevalence homes; 2 = 3.9 and p <.03). Comparisons were also made between high- and low-restraint homes on seven resident characteristics that might reflect acuity levels, which could explain why residents in high-restraint homes spent significantly more time in bed. The characteristics were MDS-derived measures of a resident's pressure ulcer risk (pressure ulcer trigger); level of assistance required for transfer, eating, and bed mobility; recall score; and presence of bowel or urinary incontinence. In addition, one comparison was made on the physical performance test results conducted by research staff (e.g., level of assistance that a resident required to stand). There were no significant differences between participants in the two groups of homes on any of these resident acuity measures, which, when combined with the demographic data presented in Table 2, supports the hypothesis that there were not significant differences between homes on multiple resident characteristics that could reflect acuity and explain differential in-bed times.
Prevalence of Restraints: Out-of-Bed Observations
The most recent MDS for the residents who participated in this project indicated that 6% of the residents in low-restraint homes were restrained when out of bed on a daily or less than daily basis versus 22% of the residents in the high-restraint homes. The observational data did not agree with these prevalence rates, and the extent of the disagreement depended on the definition of restraint use.
Observations showed that 23% of participants in high-prevalence homes and 25% of participants in low-prevalence homes used elevated foot pedals, which could potentially limit their movement. When elevated foot pedals were included in the overall restraint calculation, there was no difference in the proportion of residents observed to be restrained while out of bed between high- and low-prevalence homes (37% vs. 40%, respectively). When foot pedals were excluded as a potential restraining device, 12% of participants in both groups of homes were observed to be restrained while out of bed.
A sample of 49 participants in low-prevalence homes and 36 in high-prevalence homes who were wearing a potential restraining device while out of bed were asked to release their restraints. Seventy-eight percent of the residents in the low homes and 75% in high homes could not release these devices. If overall prevalence of restraints used when out of bed, including foot pedals, is adjusted for this number, then 27% of residents in low-prevalence homes and 31% of residents in high-prevalence homes were restrained when out of bed. These numbers are listed in Table 3 as a liberal estimate of restraint use because it included elevated foot pedals. If foot pedals are not considered, then 10% of participants in low-restraint homes were observed with a restraining device that they could not release versus 11% of participants in the high-restraint homes (see last row in Table 3, conservative estimate of out-of-bed restraints).
Care Process Indicator Scoring
The percentage of participants who were eligible for scoring and the percentage that passed each indicator are reported in Table 1. There were no significant differences between high- and low-restraint homes on the first four indicators, which pertain to restraint management. In general, all homes did better at documenting the problem that justified restraint use (Indicator 1) than they did at documenting that alternative management strategies were attempted prior to restraint use. Furthermore, though the number of eligible participants was relatively small, both groups of homes failed to consistently reposition participants every 2 hr according to thigh monitor data, despite chart documentation that repositioning occurred every 2 hr for most participants (Indicator 3).
The thigh monitor activity episode (e.g., potential exercise) frequency data showed no difference on any activity frequency measure between high- and low-restraint homes for restrained participants. The average total number of activities per hour was.40 and.38, respectively, for restrained participants in low and high homes. These data confirm the Indicator 3 conclusion that participants were moved less frequently than every 2 hr. There also were no differences in total activity episodes per hour when all participants, including those not restrained, were considered (.44 and.48, respectively, in high- and low-restraint homes). Finally, no significant difference occurred in total activity frequency per hour between participants who were observed to be restrained and those who were observed to be unrestrained but who could not walk independently. The average total number of activities per hour was.40 and.46 for restrained and unrestrained dependent participants, respectively, with no differences between homes.
Indicators 5, 6, and 7 relate to the quality of assessment for gait and mobility problems performed by primary care providers; there were no significant differences between high- and low-restraint prevalence homes on these indicators. Indicator 8, scored using interview data, assessed whether a resident's needs for physical activity were met in the form of walking assistance. Consistent with the thigh monitor data, participants reported a low number of walking assists per day in both groups of homes, with no significant difference between homes. Ninety-seven participants in the high-restraint homes reported receiving an average of less than one (.65) walking assist per day and preferring 1.32 assists per day. Participants in low-restraint homes reported receiving.88 assists per day and preferring 1.24. The proportion of participants receiving the number of assists that they preferred and, hence, passing this indicator was not significantly different between the two groups of homes (see Table 1, Indicator 8).
Observational Measures of Social Activity, Feeding Assistance, and Daytime Sleep
We previously reported that participants in high-restraint homes spent significantly more time in bed compared with participants in low-restraint homes (44% vs. 33%, respectively); as a consequence, participants in low-restraint homes were observed to be eating a higher proportion of their meals in the dining room (53% vs. 24%; p <.001). Related to this, there was a significant difference between homes in the amount of feeding assistance that participants received during meals. Participants in low-restraint homes received an average of 4.1 min of assistance during 454 observed meals, versus an average of 2.7 min for 220 observed meals for residents in high- restraint homes (p <.001). There were no differences between high- and low-restraint homes on the frequency of social interaction during meals or any measure of engagement or social interaction with staff, groups, or other individuals outside of mealtimes.
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
The answer is a conditional yes, though there was a surprising lack of difference between homes in the use of restraining devices for residents who were out of bed. The most significant differences between high- and low-restraint homes were the percentage of observations that found residents with any type of restraining device and the proportion of residents who used partial bed rails. The fact that residents in high-restraint facilities spent significantly more time in bed than those in low-restraint facilities should not be overlooked as an important care process that limits movement. Given that all participants were observed to be in bed by 7 p.m., we estimate that the typical resident in a high-restraint home spent 19 to 20 hr per day in bed (approximately 1 hr more per day than low-restraint homes), even though there was no obvious acuity difference between residents in the two groups of homes that could explain these in-bed differences. In the absence of resident acuity differences, the in-bed difference between high- and low-restraint homes is most likely due to care decisions made by staff, even though a case can be made that residents in all homes spend too much time in bed. One negative side effect of these care patterns is that fewer residents in the high-restraint homes eat in the dining room, and they receive less feeding assistance. Also, the most invasive types of restraint (e.g., bed ties or full vest restraints) were not observed in any home, suggesting that restraint reduction efforts have succeeded in elevating NH awareness about restraint issues. One limitation of this study is that we did not conduct a performance test to determine a resident's ability to remove rails and, hence, we cannot estimate how much the bed rails limit freedom of movement as we did with out-of-bed restraints.
Compared with NHs that score low on the restraint prevalence QI, do NHs with high scores provide qualitatively different care related to the management of restraints, exercise, or gait and mobility problems?
The answer is clearly no. There were no significant differences between homes, nor even a trend for one group of facilities to score better on any of the eight indicators or continuous observation measures related to restraint management and the assessment or treatment of gait and balance problems. In general, all facilities provided care to residents, restrained or unrestrained, less than once every 2 hr, despite chart documentation in most cases that restraint release occurred every 2 hr. This restraint release either must not have occurred or occurred without repositioning. In addition, residents who could accurately describe their care reported infrequent walking assistance (less than one assist per day) in all facilities, and our continuous observation measures of resident movement supported reports from residents that they received little exercise that might prevent mobility decline. These resident reports and observations stand in contrast to consistent chart documentation that exercise was frequently offered (see Indicator 7). Moreover, restrained residents did not significantly differ from unrestrained residents on measures of physical or social activity.
Conclusion and Implications
Differences between homes that scored high and low on the MDS-generated prevalence of restraints QI are reflected in measures of how often residents are in bed with side rail devices in use. This study did not find differences between homes in measures of how often residents are observed to be out of bed with restraining devices in use, though this is what the restraint prevalence QI purports to measure. Either this QI, which elicits staff ratings of daily restraint use for a 7-day period prior to an MDS assessment, is not scored accurately by NH staff or it has little relationship to direct observation measures of restraint use during a 12-hr period. We believe that the best two explanations for the discrepancy between the observations and MDS restraint data are as follows: variability in how NH staff define physical restraints, and lack of nurse aide awareness about which residents have orders for restraints. The argument that nurses may not be completing the MDS restraint items accurately is supported by other studies documenting accuracy problems in completing many MDS items, and one would suspect that the restraint items are particularly difficult to complete because of relatively vague definitions of what constitutes a "device that limits movement" (U.S. General Accounting Office, 2002a). However, we have also published data to suggest that the daily care provided by aides may not be related to what is documented in care plans or the MDS. In this case, there was no relationship between an MDS notation that a resident was on a scheduled toileting program and resident reports about how often the resident received toileting assistance (Schnelle et al., in press). It is very plausible to hypothesize that a similar disconnect between chart documentation about restraint use and what aides do on a daily basis may be present. In any case, the best way to resolve this issue is to replicate this study in a larger group of non-Southern California NHs and to use longer periods of observation. Such replication is needed to correct for the fact that the current study is limited only to Southern California NHs. However, the fact that NHs reporting a high prevalence of restraint use also keep residents in bed for longer periods of time during the day should not be overlooked as an important care difference that may have the same long-term negative effects on physical and emotional functioning that are attributed to high restraint usage. There is ample data (see, e.g., Harper & Cyles, 1988) that excessive in-bed time has numerous negative sequelae, and we specifically documented one such negative side effect in this article (lower feeding assistance).
National efforts to promote restraint reduction have increased NH awareness about the use of restraints and should now focus on less obvious factors that limit resident freedom of movement, particularly wheelchair design and excessive time spent in bed with or without bed rails. In addition, the low physical activity levels of both restrained and unrestrained residents and the lack of difference in physical activity between these two groups suggest that staff care patterns related to exercise and walking assistance should be evaluated in a different fashion than is currently the case with the MDS-generated restraint or bedfast QIs. We make this point because the high- and low-restraint homes in this study reported that only 8% and 5% of the residents were bedfast, respectively (see Table 2). These low bedfast prevalence rates are similar to the restraint prevalence data in that there is discrimination between homes on direct observational measures of the amount of time residents spend in bed. However, these low bedfast rates cannot be interpreted as reflecting good care, because many residents in all homes were spending 16+ hr in bed. Even if many of these residents do not meet the MDS QI definition of "bedfast" (22 hr or more in bed or reclining chair), the care patterns that lead to these prevalence rates and the low levels of physical activity are potentially problematic.
New measures are needed to more accurately assess residents' physical activity and how much time they spend in bed. One would be hesitant, however, to suggest that the new indicators be generated from data generated in the current MDS format. Alternatively, the data reported in this study support new initiatives designed to balance the NH quality measurement process with more frequent use of direct resident interview and observational protocols (Schnelle, 2003). For example, a recent article by Kane and colleagues (2003) has demonstrated that many residents are capable of providing information about quality of life. It would be feasible in a quality-of-life interview to include questions about a resident's physical activity and preference for being out of bed and to determine if those needs are being met. In addition, the amount of time that residents spend in bed can be estimated more easily with direct observational protocols than many other conditions. In this case, external surveyors, NH managers, or families could easily utilize the observational protocol described in this study to determine the number of residents in bed at two critical time periods, such as 10 a.m. and 4 p.m., for example. If the majority of residents were in bed during both of these time periods, there is an excellent chance that these same residents would also be in bed more than 16 hr a day.
Finally, consensus data should be developed to help NH care providers make informed decisions about getting people out of bed and providing assisted exercises. These decisions often require difficult judgments to be made among resident preferences, health issues, and the amount of time that staff have to provide care. However, we are aware of no guidelines to assist providers in making decisions about how to manage a resident's in-bed time or exercise frequency. For example, most would agree that it is inappropriate to allow a depressed resident to spend most of the day in bed even if that is the resident's preference. However, when a resident states a preference for staying in bed or for not exercising and there is no evidence of depression, the decision becomes more difficult. In general, there is no clear understanding about what physical or medical conditions (if any) justify leaving a resident in bed for most of the day or to what degree a resident's preference should be considered in making these decisions. The quality of social activity programming and systematic prompting protocols to encourage all residents to be out of bed should also be evaluated. Finally, providers and staff should systematically implement observational protocols to estimate the amount of time residents are spending in bed. This study demonstrates that such observational protocols can potentially produce information more useful for managing resident activity than information currently available through the MDS or medical record.
|
Footnotes |
|---|
1 Department of Medicine and the Borun Center for Gerontological Research, University of California–Los Angeles, Reseda, CA.
2 Geriatric Research Education and Clinical Center, VA Greater Los Angeles Healthcare System, Los Angeles, CA.
3 School of Nursing and the Borun Center for Gerontological Research, University of California–Los Angeles, Reseda, CA.
Decision Editor: Linda S. Noelker, PhD
Received for publication September 30, 2002. Accepted for publication March 6, 2003.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Engberg, N. G. Castle, and D. McCaffrey Physical Restraint Initiation in Nursing Homes and Subsequent Resident Health Gerontologist, August 1, 2008; 48(4): 442 - 452. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D. Phillips, M. Chen, and M. Sherman To What Degree Does Provider Performance Affect a Quality Indicator? The Case of Nursing Homes and ADL Change Gerontologist, June 1, 2008; 48(3): 330 - 337. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Miller and V. Mor Balancing Regulatory Controls and Incentives: Toward Smarter and More Transparent Oversight in Long-Term Care Journal of Health Politics Policy and Law, January 1, 2008; 33(2): 249 - 279. [Abstract] [PDF]
|
|
|
|
 |
|
 V. Mor, J. Zinn, P. Gozalo, Z. Feng, O. Intrator, and D. C. Grabowski Prospects For Transferring Nursing Home Residents To The Community Health Aff., November 1, 2007; 26(6): 1762 - 1771. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. G. Castle, J. Engberg, and A. Men Nursing Home Staff Turnover: Impact on Nursing Home Compare Quality Measures Gerontologist, October 1, 2007; 47(5): 650 - 661. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Kash, C. Hawes, and C. D. Phillips Comparing Staffing Levels in the Online Survey Certification and Reporting (OSCAR) System With the Medicaid Cost Report Data: Are Differences Systematic? Gerontologist, August 1, 2007; 47(4): 480 - 489. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. G Castle, J. Engberg, and D. Liu Have Nursing Home Compare quality measure scores changed over time in response to competition? Qual. Saf. Health Care, June 1, 2007; 16(3): 185 - 191. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. M. Bates-Jensen, S. F. Simmons, J. F. Schnelle, and C. Alessi Evaluating the Accuracy of Minimum Data Set Bed-Mobility Ratings Against Independent Performance Assessments: Systematic Error and Directions for Improvement Gerontologist, December 1, 2005; 45(6): 731 - 738. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. G. Castle, H. Degenholtz, and J. Engberg State Variability in Indicators of Quality of Care in Nursing Facilities J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2005; 60(9): 1173 - 1179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Arling, R. L. Kane, T. Lewis, and C. Mueller Future Development of Nursing Home Quality Indicators Gerontologist, April 1, 2005; 45(2): 147 - 156. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
