| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Peter S. Reed, PhD, MPH, Director, Care Services, Alzheimer's Association, National Office, 225 N. Michigan Ave, FL. 17, Chicago, IL 60601. E-mail: peter.reed{at}alz.org
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Cognitive impairment • Nursing homes • Assisted living • Malnutrition • Dehydration
The observed proportion of food and fluid consumed (of that served) is commonly used to identify problematic eating and drinking (Holben et al., 1999; Amella, 2002). Regulations reflected in the Minimum Data Set define clinically significant problem eating as the consumption of less than 75% of one's meal (Simmons & Reuben, 2000). Similarly, studies evaluate fluid consumption to identify intake deficiencies, with a daily minimum of 1,500 to 2,000 mL of fluid considered protective against dehydration among long-term care residents (Holben et al.). Assuming that each resident's meal has been served in accordance with a dietary service plan, this proportionate definition allows comparisons across residents and settings.
Resident characteristics contributing to food and fluid intake include cognitive status (Young, Binns & Greenwood, 2001), ability to eat independently (Kayser-Jones & Schell, 1997), and physical limitations, such as difficulty swallowing (dysphagia; Steele, Greenwood, Ens, Robertson, & Seidman-Carlson, 1997). Care provision also contributes to intake, with up to half of residents requiring assistance (Priefer & Robbins, 1997), including monitoring, verbal encouragement, and physical assistance (Van Ort & Phillips, 1995; Kayser-Jones & Schell). Environmental characteristics contributing to adequate intake include food quality, absence of environmental distractions (e.g., noise), and noninstitutional features (e.g., tablecloths), as well as social interactions (Kayser-Jones & Schell; McDaniel, Hunt, Hackes, & Pope, 2001).
Despite what is known about correlates of adequate food and fluid intake among cognitively impaired residents of long-term care, no single study has examined these problems in residential care/assisted living (RC/AL; nonnursing home settings that provide room, board, assistance with activities of daily living [ADLs], and 24-hour oversight) or comprehensively considered these factors among a large sample. This study fills this gap by describing the multiple factors present during mealtime that are associated with low intake. Understanding these factors may help form strategies to increase food and fluid intake and reduce the incidence of malnutrition and dehydration in long-term care. While the primary purpose of this study is to assess resident, staff, and environmental characteristics associated with low food and fluid intake to further contextualize these factors, data also provide a description across different facility settings, including both nursing homes and RC/AL facilities.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Measures
The Structured Meal Observation (SMO)
The SMO was developed for the Dementia Care Study to provide a research tool capable of capturing resident experience during mealtime. The SMO draws from the Meal Assistance Screening Tool (MAST; Steele et al., 1997), and incorporates other items relevant to resident need, staff assistance, environmental context, and mealtime outcomes. The SMO, as used in these analyses, consists of 28 items and is administered by observing up to five residents during the course of a single meal. It was developed using an iterative process of literature review and consensus building among study investigators, yielding an instrument with good content validity and high interrater reliability (all items in these analyses had 
70% agreement with only 4 [14%] having a kappa or intraclass correlation below 0.6, based on a sample of 48–56 paired observations). Structured meal observations were conducted by 10 data collectors, each of whom was a member of the research team and underwent training and interrater reliability testing on the SMO instrument.
In these analyses, the two outcomes of interest were amount of food and fluid consumed during a single meal, with low intake being 
75% of food and 8oz. of fluid consumed. Other resident characteristics observed using the SMO included alertness (low/high), utensil use (low/high), and postural stability and movement (assessed dichotomously as per presence or absence of at least one of six types of movement/instability). The SMO also assessed several staff assistance variables and environmental features: whether staff provide monitoring, talk to the resident, and offer physical assistance; the number of staff that provide assistance relative to the number of residents in the room; whether residents eat in a dining area; noise level; food texture; if fluids are thickened; and the number of noninstitutional features present (e.g., not eating off a tray). Social environment items included the number of people grouped with the resident and presence of the resident's family.
Other resident characteristics
Care supervisors reported whether or not a resident had at least one of 13 different mouth and throat problems, including dysphagia. Facility staff also provided data to determine residents' affect, behavior, activity involvement, and pain (Zimmerman et al., 2005). Cognitive status was assessed using either the resident's Mini-Mental State Exam score (Folstein, Folstein, & McHugh, 1975) or the supervisor-reported Minimum Data Set Cognition Scale (Hartmaier, Sloane, Guess, & Koch, 1994). Resident mobility was assessed through direct observation (Williams et al., 2005, this issue).
Facility characteristics and care
The conceptual model underlying this study (Zimmerman et al., 2005) understands a resident's unmet need as the product of resident and care factors. Care factors under study include assessment of residents' difficulty eating and drinking, staff's perceived current status of residents' eating and drinking difficulties, treatment strategies employed by the staff to address eating and drinking difficulties, and the staff's perception of the success of the treatment provided. Additional facility characteristics reported by administrators include the proportion of direct care and supervisory staff with formal training in care of resident nutrition and hydration problems, facility ownership, and facility type.
Analyses
Bivariate linear and logistic regression models, with facility type as the independent variable, were estimated to assess differences between facility settings (i.e., RC/AL facility vs nursing home). Logistic regression models were used to estimate odds ratios and 95% confidence intervals for the association between resident and facility characteristics and each intake outcome. Multivariable models provided estimates adjusted for gender, race, age, cognitive status, comorbidities, and impairments in ADLs. All analyses were adjusted for facility clustering effects using generalized estimating equations (GEE), and an exchangeable correlation structure (Stokes, Davis, & Koch, 2000). Interactions of predictors with facility setting also were tested to confirm that factors' associations with food and fluid intake did not differ between nursing homes and RC/AL.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
The prevalence of low intake observed in this study was 54.1% (food, 75% consumed) and 51.3% (fluid, 8oz. consumed). As shown in Table 1, a significantly lower proportion of RC/AL residents had low food and fluid intake relative to those living in nursing homes (50.4% vs 61.8%, 45.8% vs 63.4%, respectively, p <.05). Several components of assessment and resident status differed by setting. RC/AL residents were significantly less likely to be assessed for eating and drinking difficulties, with assessment twice as prevalent in nursing homes. Treatment varied across settings as well. RC/AL residents were less likely to receive treatment for eating difficulty, either formally, for example by a professional (7.1% vs 30.2%, p =.007), or informally (15.6% vs 34.9%, p =.048). Further, staff were observed providing physical assistance to a higher proportion of residents in nursing homes (44.3%) versus RC/AL facilities (26.7%; p =.045). In addition, the number of residents per staff member providing care was lower in nursing homes, with an average of 4.7 versus 8.2 residents per staff member (p =.005).
|
Table 2 displays risk factors for low intake, limited to a smaller sample (n = 326–335) of residents with complete data on factors used as covariates. The characteristics of this smaller sample did not differ in respect to age, gender, race, or cognitive status. After adjustment, only one resident characteristic was associated with low food intake (i.e., being nonalert), and none were associated with low fluid intake. Facility type was associated with intake, with residents of small RC/AL facilities less likely to have low food intake (OR = 0.26; 95% CI =.10,.65) and residents of new-model RC/AL facilities less likely to have low fluid intake (OR = 0.46; 95% CI =.27,.79). Ownership status also was related to both outcomes: Residents of for-profit facilities were less likely to have low food intake (OR = 0.29; 95% CI =.15,.57) and low fluid intake (OR = 0.34; 95% CI =.22,.53).
|
Two environmental features were significantly associated with both food and fluid intake. Residents having meals in the facility dining area rather than in their bedrooms were less likely to have low food intake (OR = 0.17; 95% CI =.04,.73) and low fluid intake (OR = 0.18; 95% CI =.06,.63). Similarly, residents in dining areas with more noninstitutional features were less likely to have low food intake (OR = 0.84; 95% CI =.72,.97) and low fluid intake (OR = 0.65; 95% CI =.55,.77). No differences were discovered in analyses of the interactions of these characteristics with facility type (RC/AL vs nursing home).
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
While nursing homes are more likely to assess and treat residents, residents in small RC/AL facilities have better food outcomes, and those in new-model RC/AL facilities have better fluid outcomes. Differences in staff assistance during meals and in the mealtime environment may contribute to disparate outcomes across facility settings. Residents monitored by staff during mealtimes are significantly less likely to have low food and fluid intake. Similarly, even after adjustment, residents having their meals in public dining areas are much less likely to have low intake relative to those in their bedrooms. Also, residents in dining areas with more noninstitutional features are less likely to have low food and fluid intake. Each of these beneficial staff and environmental conditions are more common in RC/AL settings.
Two aspects of this research should be noted. First, this is a study of food and fluid intake, not nutritional content. Thus, no nutritional information was used to evaluate food quality or food appropriateness relative to residents' nutritional needs. However, assessing the quantity consumed does provide an indicator that can be extrapolated into a measure of potential undernutrition and dehydration. Second, food and fluid intake during a single meal was recorded instead of overall resident intake levels throughout the day.
This study shows that across RC/AL facilities and nursing homes, there is a high prevalence of low food and fluid intake among cognitively impaired residents and a discrepancy between observed low intake and that reported by staff. While nursing homes report more assessment and treatment, outcomes do not relate to these, but instead are improved for RC/AL residents, who are more likely to be monitored during the meal, be in a public dining area, and be in a non-institutional-like setting. This research uncovers a potentially problematic care area in long-term care and highlights modifiable conditions that could be addressed through intervention. Staff who are more vigilant to low intake and facilities that attend to resident need and attempt to enhance the mealtime experience may improve nutrition and hydration among residents with dementia.
|
Footnotes |
|---|
1 Alzheimer's Association, National Office, Chicago, IL.
2 Cecil G. Sheps Center for Health Services Research, The University of North Carolina at Chapel Hill.
3 School of Social Work, The University of North Carolina at Chapel Hill.
4 Department of Family Medicine, The University of North Carolina at Chapel Hill.
5 Department of Epidemiology, School of Public Health, The University of North Carolina at Chapel Hill.
6 Regenstrief Institute, Inc. and Indiana University Center for Aging Research, Indianapolis, IN.
Decision Editor: Richard Schulz, PhD
Received for publication July 9, 2004. Accepted for publication October 14, 2004.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. L. Gruber-Baldini, S. Zimmerman, M. Boustani, L. C. Watson, C. S. Williams, and P. S. Reed Characteristics Associated With Depression in Long-Term Care Residents With Dementia Gerontologist, October 1, 2005; 45(suppl_1): 50 - 55. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Boustani, S. Zimmerman, C. S. Williams, A. L. Gruber-Baldini, L. Watson, P. S. Reed, and P. D. Sloane Characteristics Associated With Behavioral Symptoms Related to Dementia in Long-Term Care Residents Gerontologist, October 1, 2005; 45(suppl_1): 56 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Dobbs, J. Munn, S. Zimmerman, M. Boustani, C. S. Williams, P. D. Sloane, and P. S. Reed Characteristics Associated With Lower Activity Involvement in Long-Term Care Residents With Dementia Gerontologist, October 1, 2005; 45(suppl_1): 81 - 86. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
