Abstract
Background/objectives
Routine nutrition screening is recommended for all older patients admitted to hospital however data on the prevalence of malnutrition in rehabilitation settings is sparse. This study assessed the nutritional status of older patients admitted to rehabilitation hospitals over a 5 year period and described the association between nutritional status and length of hospital stay (LOS) in this context. The usefulness of a recently revised version of the shortened MNA (MNA-SF) was also investigated.
Methods
A retrospective analysis was conducted of patients aged 65 + y admitted to two rehabilitation hospitals in New South Wales, Australia between 1st March 2003–30th June 2004, and 11th January 2005–10th December 2008. Nutritional status was determined on admission by trained dieitians using the full MNA instrument and the MNA-SF. Information on diagnosis-related grouping and length of stay (LOS) was obtained.
Results
Data was available for 2076 patients with a mean age of 80.6 (27.7) y. Thirty-three percent and 51.5 % of patients were classified as malnourished and at nutritional risk, respectively. Controlling for date of admission and diagnosis related grouping, LOS was higher in malnourished and at risk groups compared to their well nourished peers (P<0.001) by 18.5 and 12.4 days, respectively. MNA-SF demonstrated high sensitivity but relatively low specificity against the full MNA.
Conclusion
The majority of older patients in the rehabilitation setting are nutritionally compromised which adversely influences LOS. In order to encourage more widespread screening, the MNA-SF may be able to replace the full MNA.
Key words: Nutrition screening, elderly, rehabilitation, MNA, malnutrition
Introduction
Malnutrition is a serious and commonly reported problem amongst hospitalised elderly patients. It is well documented that those who are either malnourished at admission or become malnourished during their hospital stay experience increased surgical complications, greater morbidity and increased length of hospital stay (1, 2, 3) and have poorer quality of life and higher rates of mortality at 12 months (4, 5, 6). Data from the UK suggests that malnutrition-related costs are in excess of €9.2 billion per year, and are mostly attributed to the treatment of malnourished patients in hospital and in long-term care facilities such as nursing homes (7).
In recognition of the adverse outcomes associated with malnutrition, many organizations recommend routine screening of older adults admitted to acute hospital settings (8, 9). The combination of poor monitoring of nutritional status, inadequate nutrient intake and increased nutritional requirements during acute hospital admissions (10) means that by the time patients are discharged to rehabilitation facilities (also known as sub-acute care hospitals) they are often at higher risk of malnutrition. The heterogeneity of the rehabilitation population, together with the use of different assessment tools, the lack of an accepted “gold standard” for malnutrition screening in these patients, and debate within the medical profession as to how nutritional issues should best be managed, has resulted in the extent of malnutrition within the rehabilitation setting having been poorly identified. We located fifteen published studies in this patient group, in which the prevalence of malnutrition ranged between 29 and 49% (10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24).
The objectives of this study were to describe the nutritional status of older patients, according to diagnosis and sociodemographic characteristics, admitted to Australian rehabilitation hospitals in the Illawarra region over a 5 year period and to describe the association between nutritional status and length of hospital stay (LOS) in the rehabilitation context. A secondary objective was to assess the usefulness of the shortened version of the MNA (MNA-SF) against the full version.
Methods
Study Design
A retrospective analysis was conducted of all patients aged 65 years and older who were admitted to two rehabilitation hospitals in the South East Sydney and Illawarra Area Health Service region, New South Wales, Australia between the periods of 1st March 2003 to 30th June 2004, and again from 11th January 2005 to 10th December 2008. To be eligible for admission to these facilities (predominantly from acute care hospitals in the surrounding local area), patients needed to meet the 2009 Australasian Faculty of Rehabilitation Medicine Guidelines (25), namely: (i) have defined rehabilitation goals; or (ii) be willing to participate actively in therapy to maximise functional abilities; or (iii) not require palliative or maintenance care. A total of 2 514 patients were identified as eligible for inclusion from the database; of these 438 had incomplete data, resulting in a total sample of 2076 patients.
Data collection
Nutritional status was determined on admission using the 18-item Mini Nutritional Assessment (MNA) (26), scored as follows: malnourished (score < 17); at risk of malnutrition (17 – 23.9); and well nourished (= 24) (27). The 6-item shortened version of the MNA (MNA–SF) (28), was scored according to the following recently revised categories (29): malnourished (0 – 7); at risk (8 – 11); well nourished (12 – 14). Scoring and nutritional risk classification of the MNA-SF was compared against the full MNA instrument, to assess sensitivity, specificity, and positive and negative predictive values of the shortened instrument.
Dietitians working within the two rehabilitation hospitals sampled received standardised training in the use of the MNA by senior dietitians and routinely administered the MNA within 72 hours of admission. Patient demographic details, admission dates and location, MNA assessment scores, weight, BMI and Diagnosis-Related Groups (DRGs) were entered into the departmental computerised patient database. Hard copies of patients' files were reviewed to obtain missing information. Date of hospital discharge was obtained from the computerized patient information system (iPm).
Ethical Considerations
Ethical approval was obtained from the University of Wollongong Human Research Ethics Committee and the South Eastern Sydney Illawarra Area Health Service.
Statistics
Statistical analysis was conducted using Microsoft Office Excel (1985-2003, Microsoft Corporation) and SPSS statistical program (V17.0: 2006, SPSS, Inc., Chicago, II, USA). Descriptive statistics were performed and categorical variables expressed as proportions. Data was assessed for normality using the Sharpiro-Wilk test. Between-group comparisons were made using Mann-Whitney U Test or Kruskal-Wallis Test. Spearman correlation was used to test for associations between variables of interest. Proportions were compared using Chi-squared test. A general linear model procedure was used to show if there was a difference in LOS between MNA categories when co-varying for DRG and the two periods of patient admission (2003/04 and 2005 – 2008). Sensitivity, specificity and positive and negative predictive values were calculated for categories of nutritional risk using the full MNA and the MNA-SF, while level of agreement between the two instruments was assessed using the kappa statistic. Statistical significance was defined as P=<0.05.
Results
Patient demographics and anthropometry
The mean age of participants was 80.6 years (SD=27.7, range 65 – 104 y). No differences were found between men and women for LOS, MNA score or BMI (Table 1). Exploratory analyses were performed to identify whether any differences were evident according to the two periods of data collection. No difference in age or proportion of men in the sample was found between the 2003/04 and 2005-08 cohorts, however MNA score was significantly higher in the 2005/08 group (19.4 (SD=5.1) vs 17.1 (4.6); P<0.001)) and a lower proportion of patients were classified as malnourished in the later group (27 %; 54 % and 19 % for malnourished, at risk and well nourished, respectively compared to 47 %, 41 % and 11 % for the 2003/04 group; P<0.001). Due to differences in recording between the two time periods, information on diagnosis-related classification groupings (DRGs) and LOS was not available for 52 % (n = 238) and 48 % (n = 221), respectively, of the 2003/04 cohort. For this reason, the “Unknown” DRG classification was excluded from the general linear model analyses to investigate LOS. The remaining diagnosis-related classification groupings (DRGs) were similar between the two cohorts (data not shown).
Table 1.
The nutritional and anthropometric characteristics of 2076 older Australian patients admitted to rehabilitation hospitals, 2003 - 2008
| Mean (SD) | Difference between genders (p-Value) † | |||
|---|---|---|---|---|
| Characteristic | Total | Male | Female | |
| n | 2076 | 838 | 1238 | |
| Age (years) | 80.6 (27.7) | 81.6 (7.1) | 79.1 (7.0) | <0.001∗∗ |
| Range | 65 - 103 | 65 - 96 | 65 - 103 | |
| Weight (kg) | 66.5 (16.0) | 72.5 (15.1) | 62.4 (15.3) | <0.001∗∗ |
| Range | 30 – 137 | 39 - 133 | 30 – 137 | |
| Body Mass Index (kg/m²) | 24.7 (5.3) | 24.5 (4.6) | 24.6 (5.6) | 0.68 |
| Range | 13 - 52 | 15 - 45 | 13 – 52 | |
| Total MNA Score | 18.7 (4.8) | 18.6 (4.9) | 18.8 (4.8) | 0.26 |
| Range | 5-29 | 5-29 | 5–29 | |
| Length of stay (days) | 32.4 (27.6) | 33.7 (29.9) | 31.5 (26.0) | 0.63 |
| Range | 1 - 242 | 1 - 214 | 2 – 242 | |
MNA=mini nutritional assessment tool, BMI=body mass index, LOS=length of stay in hospital; † Mann-Whitney Test; ∗∗P=<0.001; Missing data: MNA total score (n=5; 0.2%), LOS (n=341; 16.4%,), BMI (n=22;; 1.1%)
Classification of nutritional status and factors associated with poor nutritional status
Mean MNA score was 18.9 (SD=3.4) which fell in the “at risk of malnutrition” range. Classification of nutritional status, according to MNA score is shown in Table 2. Significant differences between sub-groups for MNA score were found for LOS, age and BMI.
Table 2.
Nutritional and anthropometric indices and length of stay of MNA nutritional risk groups
| Variables Mean (SD) (Range) | Malnourished (MNA score <17) | At risk of malnutrition (MNA score = 17 - 23.5) | Well-nourished (MNA score = 24 points) | Difference between MNA categories (P-value)† |
|---|---|---|---|---|
| n (%) | 680 (32.8%) | 1,066 (51.5%) | 325 (15.7%) | |
| Age (year) | 81.4 (7.4) | 80.7 (7.1) | 78.6 (6.6) | <0.001∗∗ |
| (65-104) | (65-100) | (65-99) | ||
| BMI (kg/m2) | 21.4 (4.4) | 25.5 (4.9) | 28.0 (4.7) | <0.001∗∗ |
| (13.0-48.0) | (14.5-46.8) | (20.1-52.2) | ||
| Length of stay | 36.8 (36.8) | 32.8 (27.8) | 23.6 (20.6) | <0.001∗∗ |
| (days) | (1-214) | (2-242) | (4-166) | |
| MNA total | 13.0 (2.9) | 20.3 (2.0) | 25.2 (1.1) | <0.001∗∗ |
| (4.0-16.5) | (17.0-23.5) | (24.0-29.0) |
MNA=mini nutritional assessment tool, BMI=body mass index, LOS=length of stay in hospital; † Kruskal-Wallis Test; ∗∗P=<0.001
The majority of patients had been admitted due to orthopaedic reasons (32.9%) or ‘other' reasons (27.4%) (Table 3). The ‘other' category included frailty (80 % of this category), cognitive decline (12 %) as well as surgical complications, spinal injuries, cellulitis and sepsis.
Table 3.
MNA nutritional risk groups and LOS, according to diagnosis-related code (DRG)
| Diagnosis-related code | % of total sample (N) | LOS† (Median (IQR), days) | MNA Category %(n)∗ | ||
|---|---|---|---|---|---|
| Malnourished | At risk for malnutrition | Well-nourished | |||
| Stroke | 14.7% | 30 (35) | 25.9% | 53.4% | 20.7% |
| (305) | (79) | (163) | (63) | ||
| Neurology | 3.9% | 24 (30) | 19.5% | 53.7% | 26.8% |
| (82) | (16) | (44) | (22) | ||
| Orthopaedic | 32.9% | 25 (23) | 26.1 % | 53.6% | 20.2% |
| (683) | (178) | (366) | (138) | ||
| Amputee | 2.4% | 43 (54) | 42.0% | 50.0% | 8.0% |
| (50) | (21) | (25) | (4) | ||
| Cardiac/ | 6.7% | 18 (16) | 37.4% | 48.2% | 14.4% |
| Respiratory | (139) | (52) | (67) | (20) | |
| Other∗∗ | 27.4% | 22 (26) ‡ | 35.5% | 51.3% | 12.5% |
| (569) | (202) | (292) | (71) | ||
| Unknown | 11.9% | 16 (14)¶ | 53.2% | 44.0% | 2.8% |
| (248) | (132) | (109) | (7) | ||
| Total | 2071 | 24 (27) | 32.8% | 51.5% | 15.7% |
| (680) | (1066) | (15.7) | |||
†Length of Stay (LOS) data available for n = 1735 (IQR: Inter-quartile range); ‡ Kruskall-Wallis test for differences in LOS between DRG categories; P<0.001 (excluding Unknown category); ¶n= 227/243 (93.4 %) missing data for LOS variable; ∗ X2 test for differences between MNA categories according to DRG; P<0.001; ∗∗ The “other” category includes the following: Frail (eg. deconditioning, decline in mobility, falls; 80%); Cognitive decline (eg. confusion, progressing dementia; 8 %); Other (pr bleed, cellulitis, sepsis etc; 12 %)
MNA score was negatively associated with LOS (Spearman r=-0.20, P <0.001) and positively associated with BMI (r=0.54, P <0.001). There was a negative correlation between age and BMI (r= -0.26; P=<0.001).
The effect of differences in DRG and date of admission on the difference in LOS between MNA status was adjusted for using a general linear model of the form: LOS = intercept + MNA + DRG + MNA∗DRG + Date. The overall model was significant (P<0.001) with a Bonferroni post hoc analysis showing a significant difference between MNA groups: “well nourished” vs “malnourished” (P<0.001; mean difference = 18.5 (SE=4.2) days); “well nourished” vs “at risk” (P=0.011; mean difference = 12.4 (SE=4.2) days); and a trend towards a difference between “at risk” and “, malnourished” groups (P=0.089; mean difference = 6.1 (SE=2.8) days).
Sensitivity, specificity and predictive value of MNA-SF
Complete data were available for both the full MNA and the MNA-SF in 1615 subjects. Using the MNA-SF, subjects were classified as follows: malnourished (39.5 %); at risk of malnutrition (52.4 %) and well nourished (8.0 %). Sensitivity, using an MNA-SF score of 0 – 8 to identify malnourished subjects, was 89.3 % (n = 465/521) while specificity of the MNA-SF (ie. Proportion of “true” well nourished that were identified) was 44.0 % (n=100 /227). Sensitivity of the MNA-SF to identify those “at risk” was 77 % (n = 669/869). The proportion of people screened by the MNA-SF that were correctly classified as malnourished (ie. positive predictive value) was 72.8 % (n = 465/639). The negative predictive value (NPV; proportion of people screened by MNA-SF that were correctly classified as well nourished) was 77.5 % (n = 100/129). The kappa statistic for level of agreement between classification using the MNA-SF and the full MNA was 0.532 (P<0.001). MNA-SF score was strongly correlated with total MNA score (Spearman r = 0.910 (P<0.01)) and positively correlated with BMI (r = 0.582; P<0.001).
When the MNA-SF was recalculated using calf circumference substituted for BMI (29), subjects (N = 1512) were classified as follows: malnourished (41.9 %); at risk of malnutrition (49.3 %) and well nourished (8.8 %). The MNA-SF-CC score was strongly correlated with total MNA score (Spearman r = 0.875; (P<0.001)) and positively correlated with BMI (Spearman r = 0.568; P<0.001). Kappa statistic between classification using the MNA-SF-CC and the full MNA was 0.475 (P<0.001).
Discussion
Our study provides convincing evidence that, within a rehabilitation (sub- acute) hospital setting, the majority of older patients admitted are either malnourished or at risk of malnutrition. The prevalence of malnutrition identified in the present study is higher then that reported from other similar settings in which rates of between 35% and 75% are documented (see Table 4) (10, 11, 12, 13, 14, 15) but consistent with the findings of Thomas et al. (17) and Compan et al. (20) who reported 91% and 85%, respectively, of patients in their samples to be at nutritional risk.
Table 4.
Studies measuring incidence/prevalence of malnutrition in patients in rehabilitation/subacute care
| Author, Country | Number and patient type Mean age, % Female Location | Assessment of malnutrition | Incidence/Prevalence of Malnutrition |
|---|---|---|---|
| Chevalier et al., 2008 Canada (11) | 182 ambulatory care rehabilitation patients 82.1 y, 66.5% F 2 geriatric hospitals | MNA | Malnourished = 3%At Risk of Malnutrition = 53%Well-nourished = 44% |
| Kaur et al., 2008 Australia (12) | 229 ambulatory care rehabilitation patients 7 2 y, 52 %F | MNA | Malnourished = 4.4%At risk of malnutrition = 56.8% |
| Brynneingsen et al., 2007 Denmark (13) | 89 patients with ischemic stroke 77.9 y, 43.3% FGeriatric stroke rehabilitation unit | Body weight BMI, MUAC, TSF and serum albumin and transferrin | Malnourished = 35% |
| Soderhamn et al., 2007 Sweden (14) | 147 patients 77.0 y, 52% F Rehabilitation ward | The Nutritional Form For the Elderly (NUFFE) | High risk of undernutrition = 14%Medium risk of undernutrition = 55%Low risk of undernutrition = 31% |
| Neumann et al., 2005 Australia (15) | 133 consecutive admissions to a rehabilitation hospital 81.0 y; 75% F | MNA, CAMA, BMI | Malnourished = 6%At risk of malnutrition = 47%Well-nourished = 47%At 90 days follow up:MNA>24; 51%MNA<24; 49% |
| Visvanathan et al., 2004 Australia (10) | 65 patients6 5 + y; 6 8 %FSubacute rehabilitation facility | MNA | Malnourished = 29.2%At risk of malnutrition = 24.6%Well-nourished = 56.9% |
| Donini et al., 2003 Italy (16) | 167 patients 6 0 – 9 5y; 75 %FGeriatric rehabilitation hospital | MNA | Malnourished = 67.7%At Risk of malnutrition = 29.9 %Well-nourished = 2.4% |
| Thomas et al. 2002 USA (17) | 837 patients76.0 y; 61% FAdmissions to subacute-care centre | MNA | Malnourished = 29%At Risk of malnutrition = 63%Well-nourished = 8% |
| Beck et al., 2001 Australia (18) | 344 patients Rehabilitation ward | SGA | SGA A= 42.7%SGA B= 28.5%SGA C = 28.8%Malnourished (B+C) = 49% |
| Aquilani et al., 1999 Italy (19) | 150 self feeding stoke patients 60.0 y; 43 % F Rehabiliation | Loss of >5% body weight and one of the following AMA <5th percentile; serum albumin, <3.5g/dl; or total lymphocytes <1500n/mm3 | Malnourished = 30% |
| Compan et al., 1999 (20) France | 196 patients 83.4 y; 62% F Subacute care | MNA | Malnourished = 32%At Risk of malnutrition = 55%Well-nourished = 12% |
| Sullivan et al., 1995 (21) USA | 350 patients1% FGeriatric Rehabilitation Unit in Veterans Hospital | Albumin <35; Chol <150mg/dl; BMI < 22 and IBW <90% | Albumin < 35 = 35%Cholesterol < 150mg = 26%BMI < 22 = 32%IBW < 90% = 26% |
| Finestone et al., 1996 (22) Canada | 49 patients 35% F Stroke rehabilitation unit | Malnutrition score comprising IBW < 90% or BMI < 20 and sum 4SF and MAMC < 5% and albumin < 3.5 g/dl and transferrin < 2g/L and TLC < 1800mm/3 | Malnutrition on admission = 49%34% at 1 mth19% at dischargeOr92% on admission80% at 1 mth if exclude MAMC and skinfold |
| Sullivan et al., 2005 (23) USA | 282 patients75.4y; 1 % FAdmissions to geriatric rehabilitation unit, Department of Veterans Affairs hospital. | Hypoalbuminaemia | Those with albumin concentrations < 35 g/L had a 2.6 times greater mortality than those with albumins= 40 g/L (relative risk = 2.6, 95% CI: 1.8–3.8). |
| Sullivan et al., 1989 (24) USA | 250 patients admitted to a Veterans Affair Hospital 71.3 y; All male | 5-10%, serum albumin <2.5g/dl, TLC<0.8 cells/mm3 unintentional weight loss >5-10% | 15% were at very high risk of malnutrition and 24% moderately at risk of malnutrition |
Many of the cited studies differed from the current study with regard to type of rehabilitation patient (ambulatory compared to inpatient (11,12) and predominantly stroke patients (19), or patient exclusion criteria (10,14) where traditionally “difficult” patient types such as those receiving specialised feeding regimens or with communication difficulties are not included. Other studies have shown these patient groups are most often at greater nutritional risk (30, 31, 32).
The present study used the Mini Nutritional Assessment (MNA), a validated screening tool that was first developed and validated in the early 1990s (26, 33) and has since been used extensively in older patients in community, hospital and nursing home settings in many countries around the world (34). A recent analysis of a large number of studies that included MNA assessments (N = 6,257 from 13 countries; mean age = 82.3y) found an overall prevalence of malnutrition of 22.8% but found considerable differences between settings (rehabilitation 50.5%, hospital 38.7%, nursing home 13.8%, and community 5.8%) (35). A further 46.2 % were classified as being in the “at risk” group, confirming that about two thirds of all older adults investigated were either at nutritional risk or malnourished. It is noteworthy that the international dataset included a total of only 384 participants from the rehabilitation setting. Our study therefore makes an important contribution as it provides one of the worlds' largest datasets in this patient group (n = 2076).
The difference in malnutrition rates between the two periods of data collection (2003/04 and 2005-08) may reflect a change in district hospital policy over this time period. For example, patients considered suitable for rehabilitation are now given preference for admission to the facilities included in the study sample, whereas in the past many patients awaiting residential care placements (with few/no rehabilitation goals) may also have been allocated beds at the study sites.
Nutrition screening is only the first step in the pathway to improved nutritional care, and screening alone is unlikely to result in beneficial patient outcomes (36). Ensuring an optimal nutritional status is not just about providing adequate food intake. There is ample evidence to show that rehabilitation patients are unable to meet their nutritional requirements, either through oral intake alone and/or supplements. An Australian study in three rehabilitation hospitals found, despite being provided with their energy and protein requirements, only 23% of patients were able to consume sufficient energy through oral intake, including supplement prescription (37).
Research on the cost-effectiveness and efficacy of interventions to prevent or treat malnutrition in various clinical settings and patient groups is sparse and the findings inconsistent (38). Examples of innovative and highly effective ways in which to improve patient dietary intake, such as incorporating feeding assistants onto a trauma ward, have been reported (39). In the nursing home setting, the positive impact of family-style dinners instead of a tray service has been demonstrated (40). Such interventions may be transferable to rehabilitation hospital settings, but would probably be prohibitively expensive and unsustainable in the long term. The difference in malnutrition rates and LOS across various diagnosis-related groupings identified in the present study may warrant prioritisation of nutrition interventions to specific groups. For example, we identified that that amputees had the longest hospital stay, as well as the highest proportion of malnutrition.
Locally, a Community & Outpatient Nutrition Extended Care Team (CONECT) established in 1997 provides outpatient dietetic services to nutritionally at risk patients in the Illawarra area who have been discharged from hospital (including hospitals sampled in the present study). Services include face-to-face consultations at a number of facilities throughout the region, telephonic support and home visits. An assessment of the CONECT programme has demonstrated improvements in the nutritional status (assessed using the MNA instrument) associated with a median time on the programme of 127 days (range = 42 – 480) (41).
Regardless of the type of nutrition intervention strategy employed in rehabilitation settings, early identification of older adults most at nutritional risk appears warranted to ensure optimal outcomes and a reduced length of stay. Our data confirms the findings of others (42) in that the malnourished and at-risk groups had an average 18.5 days and 12.4 days, respectively, increased length of hospital stay compared to their well-nourished counterparts, taking into account differences in diagnosis-related groupings and the period of hospital admission. Assuming a cost of AU$690 per day for a rehabilitation inpatient, this equates to an additional cost burden of AU$12,765 and AU$8,556 per malnourished and at risk patient, respectively. Based on this, albeit crude, economic assessment in 2008 alone, the additional cost of treating nutritionally compromised elderly rehabilitation patients who were admitted to the sites in the present study is in excess of AU$8.8m. It has been estimated that the cost of treating a nutritionally-at-risk patient is 20% higher than the average for the respective diagnosis-related group (43).
Regarding staffing levels required to ensure routine nutrition screening in the rehabilitation setting, the present study was able to achieve rates of 90% of patients seen within 3 days of admissions using a dietetic staffing ratio of a 0.3 full time equivalent staff member per 10 beds, assuming an admission rate of up to 15 referrals /week and an average LOS of 28 days. In settings where this dietetic staffing cannot be met, it is recommended that nursing staff perform the assessment.
Whilst the full MNA instrument can be performed (by adequately trained personnel) in 15 to 20 minutes, this assessment would need to compete with a battery of other geriatric assessment tests being conducted on admission, such as mobility and autonomy in daily living (ADL (44, 45) and IADL (46), mood (Geriatric Depression Scale (47), cognitive function (Mini Mental State Examination (48). It is therefore of relevance that the MNA-Short Form (MNA-SF) has recently been found to be an independent stand-alone instrument to assess nutritional status and that calf circumference can be substituted for BMI if weight and height measurements are not readily available (29).
Previously the MNA-SF was recommended only as a firststep screen with two categories of classification. The 2009 revised version (available at http://www.mna-elderly.com/mna_forms.html) now provides three categories of risk to allow more targeted interventions. We found a moderate level of agreement (? = 0.41 – 0.60) between the short and long MNA forms in the present sample. Using this classification, we found that the MNA-SF had a sensitivity to detect those that were malnourished of 89.3 % (465/521) compared to the full MNA. All 56 of the malnourished patients that would have been missed using the MNA-SF alone were classified as at risk on the MNA-SF and would therefore have been flagged for ongoing monitoring over time. The revised MNA-SF identified 669 of the 869 individuals (77 %) classified “at risk,” thus being informative for identifying the large category of people who fall into this “buffer” zone.
The relatively low specificity of the MNA-SF (44 %) means that 56 % of the patients classified as well nourished on the full MNA (n = 127/227) would have been identified as being at risk (n=124) or malnourished (n=3) using the short form. This represents an over-estimation of under-nutrition which may have implications for limited clinical resources. Although the full MNA has been shown to have better predictive power in terms of patient outcomes then other nutritional assessment methods (5, 16, 49, 50, 51), further studies are required to demonstrate the performance of the MNA-SF in this regard.
Substituting calf circumference for BMI in the MNA-SF (29) in our dataset resulted in similar associations and level of agreement with the full MNA, further suggesting that either of these anthropometric measurements could be used in a rapid screen.
There are a number of limitations to our study. Firstly, the lack of data on indicators of cognitive and physical function limits interpretation of the impact of nutritional status on daily functional ability in this patient group. However, other studies have demonstrated that malnutrition in older rehabilitation patients is associated with impaired cognition, lower BMI and reduced muscle mass, as well as poorer physical function, assessed using handgrip strength and gait speed (11). Regarding clinical outcomes, the format of the dataset did not allow for assessment of the number of hospital readmissions during the period of routine screening but we are able to report on the association between baseline nutritional status and 18-month follow up outcomes, including mortality, in a subsample of this population (unpublished). Dietary intake was not assessed thus reasons for the high prevalence of malnutrition cannot be identified.
The strength of the study lies in its large numbers and wide coverage of patients with diverse underlying conditions admitted over a defined period of time, as well as the systematic approach to nutrition screening that allows comparison with other studies. Routine administration of the MNA instrument by trained dietitians ensured that observer and selection bias was minimised.
In conclusion, the present study provides convincing evidence that most elderly patients admitted to an inpatient rehabilitation setting are nutritionally compromised and that numbers are greater than in the acute hospital setting (34). Universal screening for malnutrition in the rehabilitation setting is strongly supported by our data and the replacement of the longer MNA instrument with the shortened version (MNA-SF), using recently revised score categories appears promising.
Acknowledgements: Dr Marijka Batterham is thanked for her assistance with statistical analyses.
Source of funding: The study was supported by a summer scholarship (Nichols) from the Smart Foods Centre, University of Wollongong.
Conflict of Interest: Karen Charlton is a member of the International MNA Revision Group and attended a workshop in Switzerland in October 2008 that was fully funded by Nestle, Switzerland. None of the other authors have any conflict of interest to declare.
References
- 1.Isabel T.D., Waitzberg D.L. The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr. 2003;22:235–239. doi: 10.1016/s0261-5614(02)00215-7. 10.1016/S0261-5614(02)00215-7 [DOI] [PubMed] [Google Scholar]
- 2.Bruun L., Bosaeus I., Bergstad L., Nygaard K. Prevalence of malnutrition in surgical patients: evaluation of nutritional support and documentation. Clin Nutr. 1999;18:141–147. doi: 10.1016/s0261-5614(99)80003-x. 10.1016/S0261-5614(99)80003-X 10451474. [DOI] [PubMed] [Google Scholar]
- 3.Akner G., Cederholm T. Treatment of protein-energy malnutrition in chronic non malignant disorders. Am J Clin Nutr. 2001;74:6–24. doi: 10.1093/ajcn/74.1.6. 11451713. [DOI] [PubMed] [Google Scholar]
- 4.Visvanathan R., Macintosh C., Callary M., Penhall R., Horowitz M., Chapman I. The Nutritional Status of 250 Older Australian Recipients of Domiciliary Care Services and Its Association with Outcomes at 12 Months. J Am Geriatr Soc. 2003;51:1007–1011. doi: 10.1046/j.1365-2389.2003.51317.x. 10.1046/j.1365-2389.2003.51317.x 12834523. [DOI] [PubMed] [Google Scholar]
- 5.Kagansky N., Berner Y., Koren-Morag N., Perelman L., Knobler H., Levy S. Poor nutritional habits are predictors of poor outcome in very old hospitalized patients. Am J Clin Nutr. 2005;82:784–791. doi: 10.1093/ajcn/82.4.784. 16210707. [DOI] [PubMed] [Google Scholar]
- 6.Brantervik A.M., Jacobsson I.E., Grimby A., Wallen T.C.E., Bosaeusz I.G. Older hospitalised patients at risk of malnutrition: correlation with quality of life, aid from the social welfare system and length of stay? Age Ageing. 2005;34:444–449. doi: 10.1093/ageing/afi125. 10.1093/ageing/afi125 15955751. [DOI] [PubMed] [Google Scholar]
- 7.Elia M. Nutrition and health economics. Nutrition. 2006;22:576–578. doi: 10.1016/j.nut.2006.01.005. 10.1016/j.nut.2006.01.005 16600820. [DOI] [PubMed] [Google Scholar]
- 8.Council of Europe. Food & Safety and Consumer Health. Council of Europe; Strasbourg: 2003. Food and Nutritional Care in Hospitals: How to prevent undernutrition. Nutrition Programmes in Hospitals Group for the Committee of Experts on Nutrition. [Google Scholar]
- 9.National Collaborating Centre for Acute Care. NICE Clinical Guideline 32, February. National Collaborating Centre for Acute Care; London: 2006. Nutrition Support in Adults: Oral Nutrition Support, Enteral Tube Feeding and Parenteral Nutrition. [PubMed] [Google Scholar]
- 10.Visvanathan R., Penhall R., Chapman I. Nutritional screening of older people in a subcute care facility in Australia and its relation to discharge outcomes. Age Ageing. 2004;33:260–265. doi: 10.1093/ageing/afh078. 10.1093/ageing/afh078 15082431. [DOI] [PubMed] [Google Scholar]
- 11.Chevalier S., Saoud F., Gray-Donald K., Morais J.A. The physical functional capacity of frail elderly persons undergoing ambulatory rehabilitation is related to their nutritional status. J Nutr Health & Aging. 2008;12:721–726. doi: 10.1007/BF03028620. [DOI] [PubMed] [Google Scholar]
- 12.Kaur S., Miller M., Halbert J., Gile L., Crotty M. Nutritional status of adults participating in ambulatory rehabilitation. Asia Pac J Nutr. 2008;17:1207–1999. [PubMed] [Google Scholar]
- 13.Brynningsen P.K., Damsgaard E.M., Husted S.E. Improved Nutritional Status in Elderly Patients 6 Months After Stroke. J Nutr Health & Aging. 2007;11:75–79. [PubMed] [Google Scholar]
- 14.Soderhamn U., Bachrach-Lindstrom M., Ek A.-C. Nutritional screening and perceived health in a group of geriatric rehabilitation patients. Journal of Clinical Nursing. 2007;16:1997–2006. doi: 10.1111/j.1365-2702.2006.01805.x. 10.1111/j.1365-2702.2006.01805.x 17331091. [DOI] [PubMed] [Google Scholar]
- 15.Neumann S.A., Miller M., Daniels L., Crotty M. Nutritional status and clinical outcomes of older patients in rehabilitation. J Hum Nutr Diet. 2005;18:129–136. doi: 10.1111/j.1365-277X.2005.00596.x. 10.1111/j.1365-277X.2005.00596.x 15788022. [DOI] [PubMed] [Google Scholar]
- 16.Donini L.M., Savina C., Rosano A., De Felice M.R., Tassi L., De Bernardini L., Pinto A., Giusti A.M., Cannella C. MNA predictive value in the follow-up of geriatric patients. J Nutr Health & Aging. 2003;7:282–293. [PubMed] [Google Scholar]
- 17.Thomas D., Zdrowski C., Wilson M., Conright K., Lewis C., Tariq S. Malnutrition in subacute care. Am J Clin Nutr. 2002;75:308–313. doi: 10.1093/ajcn/75.2.308. 11815323. [DOI] [PubMed] [Google Scholar]
- 18.Beck E., Patch C., Milosavljevic M., Mason S., White C., Carrie M., Lambert K. Implementation of malnutrition screening and assessment by dietitians: malnutrition exists in acute and rehabiliation settings. Austr. Nutr Diet. 2001;58:92–97. [Google Scholar]
- 19.Aquilani R., Galli M., Guarnaschelli C., Fugazza G., Lorenzoni M., Varalda E., Arrigoni N., Zelaschi G., Crespi M., Baladi P., Mariani P. Prevalence of malnutrition and Inadequate Food Intake in Self-feeding Rehabiliation Patients with Stroke Europa Medicophysica. 1999;35:75–80. [Google Scholar]
- 20.Compan B., Di Castri A., Plaze J., Arnaud-Battandier F. Epidemiological study of malnutrition in elderly patients in acute, subacute and long-term care using the MNA. J Nutr Health & Aging. 1999;3:146–151. [PubMed] [Google Scholar]
- 21.Sullivan D.H., Walls R.C., Bopp M.M. Protein energy undernutrition and the risk of mortality within one year of hospital discharge: a follow-up study. J Am Geriatr Soc. 1995;43:507–512. doi: 10.1111/j.1532-5415.1995.tb06097.x. 7730532. [DOI] [PubMed] [Google Scholar]
- 22.Finestone H.M., Geene-Finestone L.S., Wilson E.S., Teasell R.W. Prolonged length of stay and reduced functional improvement rate in malnourished stroke rehabilitation. Arch Phys Med Rehab. 1996;77:340–345. doi: 10.1016/s0003-9993(96)90081-7. 10.1016/S0003-9993(96)90081-7 [DOI] [PubMed] [Google Scholar]
- 23.Sullivan D.H., Roberson P.K., Bopp M.M. Hypoalbuminemia 3 months after hospital discharge: significance for long-term survival. J Am Geriatr Soc. 2005;53:1222–1226. doi: 10.1111/j.1532-5415.2005.53369.x. 10.1111/j.1532-5415.2005.53369.x 16108943. [DOI] [PubMed] [Google Scholar]
- 24.Sullivan D.H., Moriarty M.S., Chernoff R., Lipschitz D.A. Patterns of care: an analysis of the quality of nutritional care routinely provided to elderly hospitalized veterans. J Parent Ent Nutr. 1989;13:249–254. doi: 10.1177/0148607189013003249. 10.1177/0148607189013003249 [DOI] [PubMed] [Google Scholar]
- 25.Australasian Faculty of Rehabilitation Medicine (AFRM). AFRM Position Statement, 2009. Available: http://afrm.racp.edu.au/index.cfm?objectid=5F2AF087-A0F4-3BED-840CA01A9B41863B. Accessed 7/8/09.
- 26.Guigoz Y., Vellas B.J., Garry P.J. Mini Nutritional Assessment: a practical assessment tool for grading the nutritional state of elderly patients. Facts Res Gerontol. 1994;4(suppl2):15–59. [Google Scholar]
- 27.Vellas B., Garry P.J., Guigoz Y., editors. Mini Nutritional Assessment (MNA): research and practice in the elderly. Nestle Nutrition Workshop Series, Clinical and Performance Programme, Volume 1. Nestle Nutrition Services; Basel: 1999. [DOI] [PubMed] [Google Scholar]
- 28.Rubenstein L.Z., Harker J.O., Salva A., Guigoz Y., Vellas B. Screening for undernutrition in geriatric practice. J Gerontol Series A: Biol Sci Med Sci. 2001;56:M366–M372. doi: 10.1093/gerona/56.6.m366. [DOI] [PubMed] [Google Scholar]
- 29.Kaiser M.J., Bauer J.M., Rämasch C., et al. Validation of the Mini Nutritional Assessment Short-Form (MNA®-SF): A practical tool for identification of nutritional status. J Nutr Health Aging. 2009;13:782–788. doi: 10.1007/s12603-009-0214-7. 10.1007/s12603-009-0214-7 19812868. [DOI] [PubMed] [Google Scholar]
- 30.Feldblum I.L., German L., Castel H., Harman-Boehm I., Bilenko N., Eisinger M., Fraser D., Shahar D.R. Characteristics of undernourished older medical patients and the identification of predictors for undernutrition status. Nutrition Journal. 2007;6:37. doi: 10.1186/1475-2891-6-37. 10.1186/1475-2891-6-37 17980023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zekry D., Herrmann F.R., Grandjean R., Meynet M.P., Michel J.P., Gold G., Krause K.H. Demented versus non-demented very old inpatients: the same comorbidities but poorer functional and nutritional status. Age Ageing. 2008;37:83–89. doi: 10.1093/ageing/afm132. 10.1093/ageing/afm132 17971391. [DOI] [PubMed] [Google Scholar]
- 32.Riccio D., Solinas A., Astara G., Mantovani G. Comprehensive geriatric assessment in female elderly patients with Alzheimer disease and other types of dementia. Arch Gerontol Geriatr. 2007;44(Suppl1):343–353. doi: 10.1016/j.archger.2007.01.047. 10.1016/j.archger.2007.01.047 17317473. [DOI] [PubMed] [Google Scholar]
- 33.Guigoz Y., Vellas B., Garry P. Assessing the nutritional status of the older person: the Mini Nutrtuional Assessment as part of the geriatric evaluation. Nutr Rev. 1996;54:S59–S65. doi: 10.1111/j.1753-4887.1996.tb03793.x. 10.1111/j.1753-4887.1996.tb03793.x 8919685. [DOI] [PubMed] [Google Scholar]
- 34.Bauer J.M., Kaiser M.J., Anthony P., Guigoz Y., Sieber C. The Mini Nutritional Assessment — its history, today's practice and future perspectives. Nutr Clin Prac. 2008;23:388–396. doi: 10.1177/0884533608321132. 10.1177/0884533608321132 [DOI] [PubMed] [Google Scholar]
- 35.Kaiser. MJ, Bauer JM, Rämsch C, Uter, W, Guigoz Y, Cederholm T. et al., for the MNA International Group (2009). The world-wide frequency of malnutrition in elderly populations: results from a pooled analysis using the Mini Nutritional Assessment. J Am Geriatr Soc. In press. [DOI] [PubMed]
- 36.Weekes C.E., Spiro A., Baldwin C., Whelan K., Thomas E., Parkin D., Emery P.W. A review of the evidence for the impact of improving nutritional care on nutritional and clinical outcomes and cost. J Hum Nutr Diet. 2009;22:324–335. doi: 10.1111/j.1365-277X.2009.00971.x. 10.1111/j.1365-277X.2009.00971.x 19624401. [DOI] [PubMed] [Google Scholar]
- 37.Walton K.L., Williams P., Tapsell L.C., Batterham M. Rehabilitation inpatients are not meeting their energy and protein needs. e-SPEN the European e-Journal of Clinical Nutrition and Metabolism. 2007;2:e120–e126. 10.1016/j.eclnm.2007.09.001 [Google Scholar]
- 38.Baldwin C. & Weekes CE. Dietary advice for illness-related malnutrition in adults. Cochrane Database of Systematic Reviews 2008: (1). [DOI] [PubMed]
- 39.Duncan D.G., Beck S.J., Hood K., Johansen A. Using dietetic assistants to improve the outcome of hip fracture: an RCT of nutritional support in an acute trauma ward. Age Ageing. 2006;35:148–153. doi: 10.1093/ageing/afj011. 10.1093/ageing/afj011 16354710. [DOI] [PubMed] [Google Scholar]
- 40.Nijs K.A.N.D., de Graaf C., Kok F.J., van Staveren W.A. Effect of family-style mealtimes on quality of life, physical performance, and body weight of nursing home residents: cluster randomised controlled trial. Br Med J. 2006;332:1180–1183. doi: 10.1136/bmj.38825.401181.7C. 10.1136/bmj.38825.401181.7C [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Nichols C. Improvement in the nutritional status of older outpatients at risk of malnutrition: A retrospective evaluation of the CONECT initiative. Major project, Master of Science (Nutrition and Dietetics), University of Wollongong (unpublished), 2008.
- 42.Kyle U.G., Genton L., Pichard C. Hospital length of stay and nutritional status. Curr Opin Clin Nutr Meta. Care. 2005;8:397–402. doi: 10.1097/01.mco.0000172579.94513.db. 10.1097/01.mco.0000172579.94513.db [DOI] [PubMed] [Google Scholar]
- 43.Amaral T.F., Matos L.C., Tavares M.M., Subtil A., Martins R., Nazaré M., Sousa Pereira N. The economic impact of disease-related malnutrition at hospital admission. Clin Nutr. 2007;26:778–784. doi: 10.1016/j.clnu.2007.08.002. 10.1016/j.clnu.2007.08.002 17936442. [DOI] [PubMed] [Google Scholar]
- 44.Katz S.C., Stroud M.W. Functional assessment in geriatrics: a review of progress and directions. J Am Geriatr Soc. 1989;37:267–271. doi: 10.1111/j.1532-5415.1989.tb06820.x. 2645355. [DOI] [PubMed] [Google Scholar]
- 45.Mahoney F.I., Barthel D.W. Functional evaluation: the Barthel Index. Md State Med J. 1965;14:61–65. 14258950. [PubMed] [Google Scholar]
- 46.Lawton M.P., Brody E.M. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9:179–186. 5349366. [PubMed] [Google Scholar]
- 47.Yesavage J.A., Brink T.L., Rose T.L., Lum O., Huang V., Adey M.B., Leirer V.O. Development and validation of a geriatric depression screening scale: A preliminary report. J Psych Res. 1983;17:37–49. doi: 10.1016/0022-3956(82)90033-4. 10.1016/0022-3956(82)90033-4 [DOI] [PubMed] [Google Scholar]
- 48.Folstein M.F., Folstein S.E. Mini-Mental State. A practical method for grading the cognitive state of patients for the clinician. J Psych Res. 1975;12:189–198. doi: 10.1016/0022-3956(75)90026-6. 10.1016/0022-3956(75)90026-6 [DOI] [PubMed] [Google Scholar]
- 49.Bauer J.M., Vogl T., Wicklein S., Trogner J., Muhlberg W., Sieber C.C. Comparison of the Mini Nutritional Assessment, Subjective Global Assessment, and Nutritional Risk Screening (NRS 2002) for nutritional screening and assessment in geriatric hospital patients. Zeitschrift fur Gerontologie und Geriatrie. 2005;38:322–327. doi: 10.1007/s00391-005-0331-9. 10.1007/s00391-005-0331-9 16244816. [DOI] [PubMed] [Google Scholar]
- 50.Guigoz Y. The Mini Nutritional Assessment (MNA) review of the literature—What does it tell us? J Nutr Health & Aging. 2006;10:466–485. [PubMed] [Google Scholar]
- 51.Persson M.D., Brismar K.E., Katzarski K.S., Nordenstrom J., Cederholm T.E. Nutritional status using mini nutritional assessment and subjective global assessment predict mortality in geriatric patients. J Am Geriatr Soc. 2002;50:1996–2002. doi: 10.1046/j.1532-5415.2002.50611.x. 10.1046/j.1532-5415.2002.50611.x 12473011. [DOI] [PubMed] [Google Scholar]