Nutrition education for care staff and possible effects on nutritional ...

2 downloads 0 Views 114KB Size Report
Jun 8, 2005 - sheltered accommodation, that is service flats (SF), nursing homes, old peoples homes and group living for the demented. Chronic disease and ...
European Journal of Clinical Nutrition (2005) 59, 947–954

& 2005 Nature Publishing Group All rights reserved 0954-3007/05 $30.00 www.nature.com/ejcn

ORIGINAL COMMUNICATION Nutrition education for care staff and possible effects on nutritional status in residents of sheltered accomodation G Faxe´n-Irving1,2*, B Andre´n-Olsson3, A Geijerstam4, H Basun5 and T Cederholm1,2 1 Department of Neurotec, Karolinska Institutet, Karolinska University Hospital, Huddinge, Stockholm, Sweden; 2Department of Geriatric Medicine, Karolinska University Hospital, Huddinge, Stockholm, Sweden; 3Department of Clinical Nutrition, Sahlgrenska University Hospital, Gothenburg, Sweden; 4Department of Oncology, Karolinska University Hospital, Solna, Sweden; and 5Clinical Science, AstraZeneca, So¨derta¨lje, Sweden

Objective: We investigated the nutritional, cognitive and functional status in residents of two service-flat (SF) complexes and the effects of a nutrition education programme for care staff. Design: Controlled nonrandomised study. Setting: Two SF complexes, that is community-assisted accommodation. Subjects: Of 115 eligible SF residents, 80 subjects participated (age 8377 y, 70% women). Intervention: The nutritional status was assessed using body mass index (BMI, kg/m2), subjective global assessment (SGA), serum concentrations of albumin, insulin-like growth factor-I (IGF-I) and vitamin B12. Cognitive and functional status were evaluated using the Mini Mental State Examination (MMSE, 0–30 points, o24 points indicates impaired cognition) and the Katz activities of daily living (ADL) index, respectively. Two assessments were made with a 5-month interval. At the start, a 12-h education programme was given to the staff at one of the SF complexes. Results: At baseline, the means of BMI and the biochemical nutritional indices were normal, whereas one-third had BMI o22 kg/m2 and one-fourth had lost Z10% of previous weight. According to SGA, 30% demonstrated possible or serious malnutrition. The median MMSE was 23 points (19.5–26.5, 25–75th percentile). Nearly 70% were ADL-independent. At the 5month follow-up there were no differences in the nutritional and cognitive status of the residents. The nutritional knowledge of the staff improved slightly (Po0.05) at both SF complexes (NS between groups). Conclusions: Around one-third of SF residents appeared to be at nutritional risk. Five months after a 12-h staff nutrition education programme, no objective changes were seen in the nutritional status of the SF residents. Sponsorship: Supported by grants from the Swedish National Board of Health and Welfare and by the Swedish Research Council.

European Journal of Clinical Nutrition (2005) 59, 947–954. doi:10.1038/sj.ejcn.1602163; published online 8 June 2005 Keywords: elderly; nutritional status; cognitive status; functional status; service flats; education

Introduction *Correspondence: G Faxe´n-Irving, Department of Geriatric Medicine, Karolinska University Hospital Huddinge, SE-141 86, Stockholm, Sweden. E-mail: [email protected] Guarantor: T Cederholm. Contributors: All authors have been engaged in the planning of the study. GF-I, BA-O and AG performed the examinations at the SF complexes. GF-I, BA-O, AG and TC were engaged in the education programme. GF-I performed the data analysis. GF-I, TC and HB completed the analyses, and took part in the writing process and final compilation of the manuscript. Received 8 June 2004; revised 4 February 2005; accepted 29 April 2005; published online 8 June 2005

Owing to demographic changes and a continuing reduction in the number of beds in Swedish hospitals, increasing numbers of chronically ill and frail elderly subjects live in sheltered accommodation, that is service flats (SF), nursing homes, old peoples homes and group living for the demented. Chronic disease and old age are risk factors for protein-energy malnutrition (PEM) (Cederholm et al, 1995; Pennington, 1998), which occurs in 25–40% of elderly subjects admitted to hospital (Unosson et al, 1991; Larsson et al, 1994; McWhirter & Pennington, 1994) and in about 5% of elderly subjects living at home (Thorslund et al, 1990;

Nutritional status in residents of sheltered accomodation G Faxe´n-Irving et al

948 ¨ m, 1992). In sheltered accommodaCederholm & Hellstro tion, an average of one-third are assessed as being malnourished (Saletti et al, 2000). Malnutrition causes fatigue, apathy and depression (Akner & Cederholm, 2001). Furthermore, muscle mass and muscle strength decrease (Cederholm et al, 1993) and the immune system is weakened (Chandra, 2002). Of a total of 135 000 elderly Swedish subjects living in community-funded sheltered accommodation, around 60 000 reside in SF. An SF is a type of accommodation where the residents live in their own needs—adapted flats, located in SF complexes, where they have access to care staff if needed. The residents usually have access to a restaurant in the SF complex where they can eat lunch. District nurses give medical care and support. Care staff working in SF primarily take care of frail residents in poor condition due to disease and poor nutrition. In general, such staff have no specific training or education about nutrition in the elderly. Education in the form of lectures is often available, but the effects of such efforts are seldom evaluated. We have previously assessed the nutritional status of subjects living in one SF complex and evaluated the effect of nutritional education for the SF staff in an uncontrolled 6-month follow-up study (Faxe´n-Irving et al, 1999). We observed that the nutritional status of the investigated group did not deteriorate during the observation period. The aims of this investigation were to include a larger group of residents, to relate nutritional status to cognitive and functional status, and to study the effects of a nutritional education programme for care staff under more controlled forms.

Materials and methods SF residents The residents who participated in the study lived in two different SF complexes, with similar staffing ratio, in two southern Stockholm suburbs with similar demographic and socio-economic structures. In one SF complex, a staff training intervention (SF-I) took place, while the other SF complex served as a control (SF-C). Altogether, 115 subjects resided in the two SF complexes. Of 53 eligible residents in SF-I, 37 subjects (73% female subjects, mean age 8477 y) agreed to participate, while the corresponding numbers in SF-C were 43 out of 62 individuals (67% female subjects, 8277 y). The residents of SF-I and SF-C had lived in the SF for 5 (2–8) and 3 (2–6) (median, 25–75th percentile) y, respectively.

Study protocol At the start, all participating residents underwent an assessment of their nutritional status, and cognitive and functional ability (see below). They were interviewed about their eating habits, where they had their main meal European Journal of Clinical Nutrition

and if they were able to prepare and purchase food themselves. The 25 staff subjects at SF-I took part in a 12-h educational programme about nutrition and diet for the elderly (see below). The educational programme took place during a 4-months-period. Each session lasted 3 h and was repeated three times to give all staff possibility to attend. The first three sessions took place within 4 weeks and the last session was 3 months later. After another month, the questionnaire was answered a second time. The 20 staff subjects at SF-C were not included in the education programme. After an average of 5 months, a follow-up examination of the residents was done using the same protocol as was used at the start. It was possible to re-examine 30 residents in SF-I and 41 residents in SF-C. In SF-I, one of the subjects had died, two had moved to a nursing home, one abused alcohol and three refrained from the follow-up examination. In SF-C, one of the participants had died and one had moved to a nursing home at the time of the follow-up. An identical questionnaire (see below) was answered at baseline and after 5 months by the SF care staff. In all, 17 staff members at SF-I and 13 at SF-C returned the questionnaire on both occasions. The dropout rate was due to changes of work place, longterm sick leave and abstaining from answering the questionnaire.

Nutritional status The subjects were weighed with a chair scale. Standing height was measured using a horizontal headboard. Eight bed-bound residents were measured in a stretched position lying in bed with an unstretchable measuring tape. Body mass index (BMI, weight (kg)/height (m)2) was calculated. Moreover, the nutritional status of the residents was assessed using a version of the subjective global assessment (SGA) (Detsky et al, 1987) that has been modified for Swedish conditions (Ulander et al, 1993). We have also made further adaptations for the assessment of older subjects (Faxe´nIrving et al, 1999). Muscle strength in the dominant hand was measured using a Harpenden Grip Strength DynamoMeters. In addition, the residents classified their appetites using a visual analogue scale (VAS, 0–10), where 1 ¼ poor and 10 ¼ very good (Faxe´n-Irving et al, 1999; Parker et al, 2004). Two trained dieticians (BAO, AafG) did the measurements, one at each SF. Serum concentrations of albumin, transferrin, vitamin B12 and haemoglobin (Hb) were analysed as biochemical indicators of nutritional status. The routine methods and reference ranges of the Laboratory of Clinical Chemistry at the hospital were used. C-reactive protein (CRP, Hitachi 917, Boehringer-Mannheim, Germany) was measured to determine any active inflammatory process, reference value o10 g/l. Furthermore, the anabolic mediator serum insulin-like growth factor-I (IGF-I) was analysed with a commercial radioimmunoassay after acid ethanol extraction (Nichols Products Corp., San Juan, Capistrano, CA, USA).

Nutritional status in residents of sheltered accomodation G Faxe´n-Irving et al

949 Cognitive status and functional status The cognitive function was assessed using the Mini Mental State Examination (MMSE), where 11 subtasks are assigned a total of 30 points (Folstein et al, 1975). Less than 24 points indicates impaired cognitive function. The assessments were made by two trained district nurses, one at each SF. The district nurse at SF-C shifted work place in the middle of the study; therefore another trained district nurse made the follow-up assessments. The participants were asked to estimate their mood according to a VAS 0–10, where 0 ¼ very sad and 10 ¼ very happy (Faxe´n-Irving et al, 1999; Kertzman et al, 2004). The participant’s ability to manage activities of daily living (ADL), that is bathing, dressing/undressing, using the toilet, movement, transfer, continence and eating, were assessed according to the Katz ADL index (Katz et al, 1963) and graded as: A ¼ independent in all functions, B ¼ dependent of help in one function, C–F ¼ dependent in 2–5 functions, and G ¼ dependent in all six functions. Medical records showing diagnoses of the residents were not available. The number of prescribed drugs was noted and averaged 2.5 (1–6) in SF-I and 3.5 (2–7) in SF-C residents (NS). The resident’s need of assistance from the care staff was regularly assessed by a senior local welfare official, using a scale ranging from 0 (no help at all) to 8 (total dependence). The median need of assistance did not differ between SF-I (3; 0–6 (25–75th percentile)) and SF-C (2; 0–5). In total, 19 (51%, SF-I) and 27 (63%, SF-C) residents received assistance once a week or more.

Staff education and questionnaire The education programme was compulsory for all staff at SF-I. Dietitians, physicians and external care staff gave lectures on malnutrition in the elderly, food and nutritional requirements, dental care, how to detect swallowing difficulties and how to change the consistency of food. Problems in eating connected with conditions like dementia, Parkinson’s disease and stroke were elucidated. Lectures were combined with practical exercises, such as calculating BMI, making and testing nutritional drinks, thickening or enriching drinks of various types, and trying foods with changed consistencies. The personnel at both SF complexes answered a questionnaire (Faxe´n-Irving et al, 1999) at the start and after 5 months. The questionnaire reflected whether the staff were aware of the residents’ weights, appetites, preference of food consistency and if the residents ate alone or in company. They were also asked if they thought they could influence the residents’ food purchases. The care staff stated their assessment of the importance of nutrition in various diseases using a VAS scale (0, low—10, high). Finally, the questionnaire contained a description of a fictitious patient case: a depressed, widowed woman living alone, with hemi-paresis, chewing and swallowing difficulties and weight loss after a stroke. In total, 12 questions were given and the task was to

suggest nutritional measures according to fixed alternatives. The test score could be 0–12 points.

Statistics and ethics Data are presented as mean7s.d. unless otherwise stated. To analyse the variations between the two groups of residents, the Student’s t-test, the Mann–Whitney U-test and w2-test were used in accordance with the type and distribution of the variables. Possible changes at follow up were evaluated by the paired t-test, Wilcoxon’s matched pair test and w2-test. For correlation analyses, Pearson’s and Spearman’s correlation coefficients were used. The Statisticas software package (Statsoft, OK, USA) was used for the statistical calculations. All investigated subjects received oral and written information before consenting to participate. The project was reviewed and approved by the local ethics committee and conformed to the Helsinki declaration.

Results Nutritional, cognitive and functional status at baseline The nutritional status of all participating subjects is shown in Table 1. The mean BMI was close to the upper traditional reference range (20–25 kg/m2) and did not differ between the two SF complexes. About one-third (n ¼ 27, 34%) of the participants had a BMI o22 kg/m2, and 13 (16%) had a BMI o20 kg/m2. About one in four of the subjects, possibly more common in SF-C (P ¼ 0.06), reported weight losses that exceeded 10%, when current weight was compared with the recalled highest weight previously in life. Serum albumin and serum IGF-I were subnormal in 15 (19%) and 27 individuals (34%), respectively. One-fourth of the SF population had CRP 4 10 g/l, indicating ongoing inflammatory activity. The residents assessed their appetite as being in the upper range. According to SGA, a total of 10 subjects (13%) were seriously malnourished and 17 (21%) were suspected to be malnourished. Around 50% of all subjects had MMSE o24, indicating a cognitive impairment and one-fourth had MMSE o20 points (Table 1). The residents assessed their mood as a 5 on a 0–10 VAS scale. The median Katz ADL index was A (independent), but 25 subjects (31%) were dependent in one or more functions. SF-I residents seemed to need more help. BMI correlated weakly to serum IGF-I (r ¼ 0.22, P ¼ 0.05). BMI also correlated to mood (r ¼ 0.24, P ¼ 0.03,) but not to MMSE (r ¼ 0.05). Hand-grip strength correlated to weight (r ¼ 0.47, Po0.0001), Hb (r ¼ 0.45, Po0.0001), albumin (r ¼ 0.31, P ¼ 0.006), MMSE (r ¼ 0.42, P ¼ 0.0001), and to ADL (r ¼ 0.32, P ¼ 0.004) and need of assistance (r ¼ 0.34, P ¼ 0.004). Serum levels of CRP correlated inversely to albumin (r ¼ 0.44, Po0.0001) and IGF-I correlated to transferrin (r ¼ 0.24, P ¼ 0.035). According to the residents’ self-assessments of their nutritional situation, around 70% were able to prepare and cook their own meals. One-third managed to purchase food independently. Less than 40% ate lunch in the SF restaurants European Journal of Clinical Nutrition

Nutritional status in residents of sheltered accomodation G Faxe´n-Irving et al

950 Table 1 Nutritional, cognitive and functional status at baseline in elderly residents of two service flat (SF) complexes SF (all) (n ¼ 80)

SF-intervention (n ¼ 37)

Body mass index (kg/m2)

24.374.9

2575.5

Weight Men, median (range)a Womenb

63715.2 71.3 (51.3–112.5) 58.2712.5

64718 77.7 (67–112.5) 57.1713

62.1712.4 66.5 (51.3–83.5) 59.2712.1

26% (15/57)

13% (3/23)

35% (12/34)

30.379.1 15.776.9

27.278.9 11.675

32.578.9 19.276.4***

40.174.2 o10 (o10–11) 2.270.4 130716 3817284 105748 7.272.9

39.474.1 o10 (o10–11) 2.270.3 130715 3277170 101748 6.872.8

40.874.1 o10 (o10–o10) 2.370.5 130716 4277348 109748 7.573

Subjective global assessment (%)e A B C

52 (66) 17 (21) 10 (13)

20 (55) 9 (25) 7 (20)

32 (74) 8 (19) 3 (7)

Mood (VAS 1–10)f

5.372.5

5.672.1

5.172.7

NA

22.575.2

21.675.4

23.275.1

NA

A (A-B)

A (A-C)

Subjects with weight loss Z10%c Hand-grip strength (kg) Men Women Serum albumin (g/l) Serum C-reactive protein (g/l) (median, 25–75th perc.)d Serum transferrin (g/l) Serum haemoglobin (g/l) Serum vitamin B12 (pmol/l) Serum insulin-like growth factor-I (mg/l) Appetite (VAS 1-10)

MMSE, (0–30 points) ADL (A-G) (median, 25–75th perc.)

SF-control (n ¼ 43) 23.774.3

Ref. value NA

NA

35–46 o10 1.94–3.26 130–165 110–600 84–115d NA A

A (A-A)*

A

Mean7s.d. if not otherwise stated. a In total 24 men; 10 in SF-I, 14 in SF-C. Median is given due to the fact that two men in SF-I weighed 109 and 112.5 kg. b In total 56 women; 27 in SF-I, 29 in SF-C. c Weight at age 65 y compared to current weight. Results are given as percentage and relative number of subjects able to give reliable information. d Mean values in healthy elderly controls (Hilding et al., 1999). e A ¼ well-nourished; B ¼ possibly malnourished; C ¼ seriously malnourished. VAS ¼ Visual Analogue Scale (1–10). Appetite: 1 ¼ no appetite, 10 ¼ good appetite. f VAS ¼ Visual Analogue Scale, 1 ¼ sad, 10 ¼ happy. MMSE ¼ Mini Mental State Examination, ADL ¼ Activities of Daily Living: A-G ¼ independent to totally dependent. *Po0.05, ***Po0.001, refers to differences between the groups at the start. NA ¼ not applicable.

and eight subjects had their main meal routinely delivered as ready-cooked foods. Difficulties in eating were reported by half, and chewing and swallowing problems dominated. The residents (n ¼ 53) who were able to cook and prepare meals had significantly better hand-grip strength than the group that needed meal support. They also had higher MMSE scores, ADL function and serum values of haemoglobin and albumin (data not shown).

5-Month follow-up of nutritional, cognitive and functional status (Table 2) Seven and two residents in SF-I and SF-C, respectively, did not participate in the follow up. Mean BMI among the dropouts was 22.973.4. Four dropouts were classified as well nourished (one from SF-C), two were classified as at risk (one from SF-C) and two as malnourished at inclusion. European Journal of Clinical Nutrition

No change in the objective measures weight or BMI was observed in any of the two SF populations. A decrease in serum albumin in SF-C and a rise of serum transferrin was noticed in both groups. In contrast, SGA classification indicated improvement of nutritional status in seven subjects and deterioration in one subject (Po0.01) in SF-I. Corresponding findings in SF-C were improvement in two subjects and deterioration in four subjects (Po0.01 within the group, Po0.05 between groups). When we compared only those residents in need of assistance at least once a week at SF-I and SF-C, no differences were noticed at follow up, except a slight decrease in hand-grip strength in SF-I (data not shown). The resident’s appetites were fairly good and stable during the observation period according to VAS. No changes in MMSE or in assessed mood were seen, while the ADL status seemed to deteriorate in both groups.

Nutritional status in residents of sheltered accomodation G Faxe´n-Irving et al

951 Table 2 Anthropometric and biochemical variables, cognitive and functional status and mood in individuals living in SF at baseline and after 5 months. Baseline data given only for residents participating also at the follow-up examination SF-Intervention (n ¼ 30) 0

5 months

Weight

65.2718.8

65.5719.1

Body mass index (kg/m2) Mena Womenb

28.276.2 24.275.4

28.476.1 24.275.3

Hand-grip strength (kg) Men Women

26.779.3 11.975.1

24.579.1 10.576.3

Serum albumin (g/l) Serum transferrin (g/l) Serum hemoglobin (g/l) Serum vitamin B12 (pmol/l) Serum insulin-like growth factor-I (mg/l) Appetite (VAS 1-10)

38.974.1 2.270.3 130715 3097161 102751 6.972.7

38.673.7 2.270.3 129715 3297156 103751 6.572.7

17 (59) 8 (27) 4 (14) 5.971.9 21.275.5 A-B (A-D)

Subjective global assessment (%)c A B C Mood (VAS 1-10) MMSE ADL, (median, 25–75th perc.)

SF-Control (n ¼ 41) P1

0 62712.7

5 months

23.473.5 23.874.8

3379 18.976.3

32.878.5 19.774.3

40.974.2 2.370.5 130716 4027313 109749 7.872.8

39.774.2 2.470.5 130715 4197317 106752 7.772.6

20 (69) 8 (28) 1 (3)

31 (76) 7 (17) 3 (7)

30 (73) 7 (17) 4 (10)

5.972.4 21.376 B (A-E)

5.272.8 23.275.2 A (A-A)

5.272.0 24.175.3 A (A-B)

Po0.05

P2

62712.7

23.473.4 23.874.8

Po0.01 P ¼ 0.05

P1

Po0.05 Po0.05 Po0.05 Po0.05

Po0.05

Po0.05

Po0.001

P1 refers to changes over time within the groups, that is Student’s paired t-test, Wilcoxon’s matched pair test and w2-test according to the distribution of the variables. P2 refers to difference in change over time between the groups, that is Student’s t-test, Mann–Whitney U-test and w2-test according to the distribution of the variables. a In total 22 men; nine in SF-I, 13 in SF-C. b In total 49 women; 21 in SF-I, 28 in SF-C. MMSE ¼ Mini Mental State Examination. ADL ¼ activities of daily living. c n ¼ 29.

Attitudes and knowledge among staff at the start and after intervention In all, 70% of the daytime working staff attended all training courses. On a VAS scale, the staff at both SF complexes regarded nutrition as important in patients with diabetes mellitus, malignant diseases, heart failure, depression and gastrointestinal disorder (VAS 6–10) at the start. Nutrition was scored as less important (VAS 3–6) in patients with hip fracture, lung disease, pressure ulcers and rheumatoid arthritis, especially by the SF-I staff. According to the test results on the fictitious patient case, the knowledge appeared to improve (Po0.05) in the staff of both SF complexes, without a significant difference between the two (data not shown).

Discussion The SGA assessments indicated that one in three SF residents was possibly or seriously malnourished. Half of the residents demonstrated cognitive impairment and one-third were dependent in one or more functions. Most of the subjects managed to cook and prepare their food. Half reported

difficulties in eating. The staff educational programme on nutrition did not affect objective nutritional measures in the SF residents during a 5-month observation period. The major aim of this study was to evaluate whether a limited education effort could affect the SF staff’s awareness of and, knowledge about, nutrition for the elderly, and to see whether the staff training might improve the nutritional status of the residents. In a previous uncontrolled study, we noticed that a similar approach was associated with a stable nutritional status in SF residents (Faxe´n-Irving et al, 1999). We could not, however, deduct whether the stable status was an effect of the educational programme, as the observation period may have been too short for a decline to become evident. In the present study, we did not notice any objective improvements in the residents’ nutritional status at the 5month follow-up. There was a significant increase in number of residents, both within the group (SF-I) and between the groups (SF-I vs SF-C), who improved their SGA-classification, but these data must be interpreted cautiously, as they might be biased. The impact of an isolated staff education programme may not be strong enough to show an effect with the instruments for nutritional evaluation used in this European Journal of Clinical Nutrition

Nutritional status in residents of sheltered accomodation G Faxe´n-Irving et al

952 study. Moreover, the study population was small. The data indicated that the knowledge about nutrition improved in both SF care staff groups, a possible secondary effect from the study in the SF-C staff subjects as well. However, these data need to be cautiously interpreted due to the few staff members that completed the questionnaire. About half of the residents received no, or very little, assistance and they would probably not benefit from education targeted at the staff. An interesting issue for future studies would be to explore the possible effects of a nutritional education programme offered to the residents, with or without relatives, focusing on strategies to prevent weight loss (Rivie`re et al, 2001). In a systematic review over studies between 1990 and 2003 on nutrition education for older adults, the authors concluded that age was not a limiting factor for gaining knowledge, and that nutrition education to older adults may prevent or delay the spiral toward ill health and disability (Sahyoun et al, 2004). The staff of the SF restaurants would be another possible target for adapted nutritional education. Although the restaurant employees at SF-I were invited to take part in the courses, they had no opportunity to do so. Consequently, it was difficult to influence the food that was served in the restaurant, as well as in the delivered ready-cooked foods that were prepared in the SF restaurant. In the interviews, one-fourth of the residents expressed a desire for food with adapted consistency, but the SF restaurants did not provide other options than ordinary food. Alternatively, a combination of nutritional supplements and an education programme may have been more effective in this population group. For example, in another study, 5 months after such combined intervention in residents of group living for the demented, a weight gain of 3.5 kg was observed (Faxe´nIrving et al, 2002). Although not corroborated by the present data, it is still reasonable to believe that nutritional education for care staff may, especially in sheltered accommodation with more dependent residents, reduce the risk of malnutrition in the residents. Nutritional education courses given to nursing staff in rehabilitation and long-term care resulted in more flexibility and individual care of the patients (Sidenvall, 1997). It is suggested by Perry (1997) that when changing nursing practice in nutritional care, increasing knowledge only is not sufficient. Insight into the influence of group norms and identification of individual and situational characteristics are important and must also be addressed. The other aim of this study was to characterise the nutritional status of SF residents in relation to cognitive and functional status. The mean BMI did not indicate a general occurrence of underweight, but one-third of the residents had a BMI o22 kg/m2, which may be a sign of nutritional depletion (Dorn et al, 1997; Beck & Ovesen, 1998; Flodin et al, 2000). About the same proportion of residents reported weight loss, had subnormal serum IGF-1 and an SGA indicating possible or serious malnutrition. In comparison with our previous report on 28 SF residents, the European Journal of Clinical Nutrition

residents in the current study had a lower BMI and more subjects were classified as malnourished or at risk for malnutrition. In a study including 349 Swedish SF residents, the mean BMI was 24 (Saletti et al, 2000), which is in line with what we found in the present study. Concurrently, 21% were classified as malnourished and 49% as at risk for malnutrition according to the Mini Nutritional Assessment (MNA), which is a much higher prevalence than we found. Previous reports indicate that MNA has a higher sensitivity for malnutrition in elderly people than SGA (Christensson et al, 2002; Persson et al, 2002), even though the methods correlate well (Persson et al, 2002). MNA has a more global approach, covering not only nutrition but also cognition, mobility and morbidity, whereas SGA focuses more on nutrition issues. Thus, SGA seems to be more useful in detecting elderly residents with established malnutrition, while MNA identifies subjects who need preventive nutritional measures. Still, both methods are used and validated in the assessments of older people (Sacks et al, 2000; Vellas et al, 2000). In accordance with several previous reports, hand-grip strength correlated with nutritional markers (Chilima & Ismail, 2001; Humphreys et al, 2002; Pieterse et al, 2002). It also correlated to MMSE and ADL. Therefore, hand-grip strength measurements seem useful in the nutritional and functional assessment of elderly people. Unexpectedly, as many as one-fourth of the residents had a CRP410 g/l, indicating the presence of a subclinical inflammation of unclear origin. Somewhat more expectedly, serum CRP correlated negatively with serum albumin, underlining the observation that low serum albumin concentration is a marker of on-going inflammation, and thus also of on-going catabolism (Fleck, 1988; Minuto et al, 1989; Cederholm et al, 1993, 1995). The MMSE assessments suggested that more than half of the investigated residents were cognitively impaired. It should be pointed out that MMSE is a useful screening tool for cognitive impairment, but not a diagnostic test for dementia. A Swedish study of unselected 85 y-old-subjects reported dementia to be present in close to one-third of the subjects (Skoog et al, 1993). Given the fact that mainly the frail elderly requiring extra support move to communityfunded sheltered housing, it is likely that the prevalence of dementia will be high. In contrast to previous findings, MMSE did not correlate to BMI in the present group of elderly SF residents (Faxe´n-Irving et al, 1999). However, MMSE correlated to other nutrition-related variables like hand-grip strength. Limitations of the study may include that the SGA and interviews were performed by two different dieticians, one at each SF complex. However, the SGA classification technique has demonstrated a high reproducibility among numerous assessors (Ek et al, 1996; Duerksen et al, 2000; Sacks et al, 2000). Moreover, at SF-I, two different nurses performed the assessment of ADL and MMSE, since one district nurse changed jobs in the middle of the study. MMSE is considered

Nutritional status in residents of sheltered accomodation G Faxe´n-Irving et al

953 to be reliable even when performed by multiple examiners (Folstein et al, 1975). Finally, it cannot be ruled out that filling out questionnaires, as well as the presence of dieticians and nurses performing nutritional and functional resident assessments, influenced the staff at SF-C to become more aware of their residents’ nutritional situations. In conclusion, a majority of older SF residents had a satisfactory nutritional status. However, about one-third had a low BMI, reported weight loss, subnormal IGF-I and an SGA indicating evident or risk for malnutrition, which, taken together, highlights the need for undertaking nutritional measures within community care. This study, with its limitations, could not demonstrate that a short nutritional education programme raised the knowledge in the care staff or improved the nutritional status of the SF residents 5 months later. Further studies are needed to address the possible effects of the combining staff education with education of the residents, the SF restaurant staff, or combining of educational programmes with nutritional support.

Acknowledgements This work was supported by grants from the Stockholm South-West Health Care Region, from the Swedish National Board of Health and Welfare and from the Swedish Research Council. The work was carried out in co-operation with the ¨ m, Bra¨nnkyrka District Council. We thank Mrs Eyra Granstro ¨ ran Henriksson, Mrs Maj-Lis Styrwolt, Mrs Marianne Mr Go Popovski, Mrs Eeva Eriksson and Mrs Sonia Mattsson for their committed involvement in the study.

References Akner G & Cederholm T (2001): Treatment of protein-energy malnutrition in chronic nonmalignant disorders. Am. J. Clin. Nutr. 74, 6–24. Beck AM & Ovesen L (1998): At which body mass index and degree of weight loss should hospitalized elderly patients be considered at nutritional risk? Clin. Nutr. 17, 195–198. ¨ m K (1992): Nutritional status in recently Cederholm T & Hellstro hospitalized and freeliving elderly subjects. Gerontology 38, 105–110. ¨ m K (1993): Nutritional status and Cederholm T, Ja¨gre´n C & Hellstro performance capacity in internal medical patients. Clin. Nutr. 12, 8–14. ¨ m K (1995): Outcome of proteinCederholm T, Ja¨gre´n C & Hellstro energy malnutrition in elderly medical patients. Am. J. Med. 98, 67–74. Chandra RK (2002): Nutrition and the immune system from birth to old age. Eur. J. Clin. Nutr. 56 (Suppl 3), S73–S76. Christensson L, Unosson M & Ek A-C (2002): Evaluation of nutritional assessment techniques in elderly people newly admitted to municipal care. Eur. J. Clin. Nutr. 56, 810–818. Chilima DM & Ismail SJ (2001): Nutrition and handgrip strength of older adults in rural Malawi. Public Health Nutr. 4, 11–17. Detsky AS, McLaughlin JR, Baker JP, Johnston N, Whittaker S, Mendelson RA & Jeejeebhoy KN (1987): What is subjective global assessment of nutritional status? J. Parenteral Enteral Nutr. 11, 8–13. Dorn J, Schisterman E F, Winkelstein W & Trevisan M (1997): Body mass index and mortality in a general population sample of

men and women. The Buffalo Health Study. Am. J. Epidemiol 146, 919–931. Duerksen DR, Yeo TA, Siemens JL & O’Connor MP (2000): The validity and reproducibility of clinical assessment of nutritional status in the elderly. Nutrition 16, 740–744. Ek AC, Unosson M, Larsson J, Ganowiak W & Bjurulf P (1996): Interrater variability and validity in subjective nutritional assessment of elderly patients. Scand. J. Caring Sci. 10, 163–168. Faxe´n-Irving G, Andre´n-Olsson B & Cederholm T (1999): Nutritional and cognitive status in elderly subjects living in service flats and the effect of nutrition education on personnel. Gerontology 45, 187–194. Faxe´n-Irving G, Andre´n-Olsson B, Geijerstam A, Basun H & Cederholm T (2002): The effect of nutritional intervention in elderly subjects residing in group-living for the demented. Eur. J. Clin. Nutr. 56, 221–227. Fleck A (1988): Acute phase response: implications for nutrition and recovery. Nutrition 4, 109–117. Flodin L, Svensson S & Cederholm T (2000): Body mass index as a predictor of 1-year mortality in geriatric patients. Clin. Nutr. 19, 121–125. Folstein MF, Folstein SE & McHugh PR (1975): ‘Mini Mental State’. A practical method for grading the cognitive state of patients for the clinician. J. Psychiatry Res. 12, 189–198. Hilding A, Hall K, Wivall-Hilleryd IL, Sa¨a¨f M, Melin A-L & Thore´n M (1999): Serum levels of insulin-like growth factor 1 (IGF-1) in 152 patients with growth hormone (GH) deficiency aged 19–82 years in relation to healthy subjects. J. Clin. Endochrinol. Metab. 84, 2013–2019. Humphreys J, de la Maza P, Hirsch S, Barrera G, Gattas V & Bunout D (2002): Muscle strength as a predictor of loss of functional status in hospitalized patients. Nutrition 18, 616–620. Katz S, Ford AB, Moskowitz RW, Jackson BA & Jaffe MW (1963): Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychological function. JAMA 185, 94–99. Kertzman S, Aladjem Z, Milo R, Ben-Nahum Z, Birger M, Grinspan H, Weizman A & Kotler M (2004): The utility of the Visual Analogue Scale for the assessment of depressive mood in cognitively impaired patients. Int. J. Geriatr. Psychiatry 19, 789–796. ¨ f N & Bark T (1994): Underna¨ring Larsson J, Andersson M, Askelo vanligt vid svenska sjukhus (in Swedish, abstract in English). La¨kartidningen 91, 2410–2413. McWhirter JP & Pennington CR (1994): Incidence and recognition of malnutrition in hospital. Br. Med. J. 308, 945–948. Minuto F, Barreca A, Adami GF, Fortini P, Del Monte P, Cella F, Scopinaro N & Giordano G (1989): Insulin-like Growth Factor-1 in human malnutrition: Relationship with some body composition and nutritional parameters. J. Parenteral Enteral Nutr. 13, 392–396. Parker BA, Sturm K, MacIntosh CG, Feinle C, Horowitz M & Chapman IM (2004): Relation between food intake and visual analogue scale ratings of appetite and other sensations in healthy older and young subjects. Eur. J. Clin. Nutr. 58, 212–218. Pennington CR (1998): Disease-associated malnutrition in the year 2000. Postgrad Med. J. 74, 65–71. Perry L (1997): Nutrition: a hard nut to crack. An exploration of the knowledge, attitudes and activities of qualified nurses in relation to nutritional nursing care. J. Clin. Nurs. 6, 315–324. ¨ m J & Cederholm T Persson M, Brismar K, Katzarski K, Nordenstro (2002): Nutritional status using mini nutritional assessment and subjective global assessment predict mortality in geriatric patients. J. Am. Geriatr. Soc. 50, 1996–2002. Pieterse S, Manandhar M & Ismail S (2002): The association between nutritional status and handgrip strength in older Rwandan refugees. Eur. J. Clin. Nutr. 56, 933–939. Rivie`re S, Gillette-Guyonnet S, Voisin T, Reynish E, Andrieu S, Lauque S, Salva A, Frisoni G, Nourhashemi F, Micas M & Vellas B (2001): A nutritional education program could prevent weight loss and slow cognitive decline in Alzheimer’s disease. J. Nutr. Health Aging 5, 295–299.

European Journal of Clinical Nutrition

Nutritional status in residents of sheltered accomodation G Faxe´n-Irving et al

954 Sacks GS, Dearman K, Replogle WH, Cora WL, Meeks M & Canade T (2000): Use of subjective global assessment to identify nutritionassociated complications and death in geriatric long-term care facility residents. J. Am. Coll. Nutr. 19, 570–577. Sahyoun NR, Pratt CA & Anderson A (2004): Evaluation of nutrition education interventions for older adults. A proposed framework. J. Am. Diet Assoc. 104, 58–69. Saletti A, Lindgren E, Johansson L & Cederholm T (2000): Nutritional status according to Mini Nutritional Assessment in an institutionalized elderly population in Sweden. Gerontology 46, 139–145. Sidenvall B (1997): Altered values concerning meals procedures among caregivers in elderly care. Health Care in Later Life 2, 187–196. Skoog I, Nilsson L, Palmertz B, Andreasson LA & Svanborg A (1993): A population-based study of dementia in 85-year-olds. N. Eng. J. Med. 328, 153–158.

European Journal of Clinical Nutrition

Thorslund S, Toss G, Nilsson I, Schenk H, Symreng T & Zetterqvist H (1990): Prevalence of proteinenergy malnutrition in a large population of elderly people at home. Scand J. Prim. Health Care 8, 243–248. Ulander K, Grahn G & Jeppsson B (1993): Subjective assessment of nutritional status—validity and reliability of a modified Detsky index in a Swedish setting. Clin. Nutr. 12, 15–19. Unosson M, Ek A-C, Bjurulf P & Bjurulf P (1991): Demographical, sociomedical and physical characteristics in relation to malnutrition in geriatric patients. J. Adv.. Nurs. 16, 1406–1412. Vellas B, Guigoz Y, Baumgartner M, Garry PJ, Lauque S & Albarede JL (2000): Relationships between nutritional markers and the mininutritional assessment in 155 older persons. J. Am. Geriatr. Soc. 48, 1300–1309.