Original Article

Cut-off scores for the Minimal Eating Observation and Nutrition Form – Version II (MEONF-II) among hospital inpatients

Albert Westergren1*, Erika Norberg2, Christina Vallén2 and Peter Hagell1,3

1Department of Clinical Nursing Sciences, School of Health and Society, Kristianstad University, Kristianstad, Sweden; 2Departments of Nutrition and Health, Central Hospital, Kristianstad, Sweden; 3Department of Health Sciences, Lund University, Lund, Sweden

Abstract

Background and objective: The newly developed Minimal Eating Observation and Nutrition Form–Version II (MEONF-II) has shown promising sensitivity and specificity in relation to the Mini Nutritional Assessment (MNA). However, the suggested MEONF-II cut-off scores for deciding low/moderate and high risk for undernutrition (UN) (>2 and >4, respectively) have not been decided based on statistical criteria but on clinical reasoning. The objective of this study was to identify the optimal cut-off scores for the MEONF-II in relation to the well-established MNA based on statistical criteria.

Design: Cross-sectional study.

Methods: The study included 187 patients (mean age, 77.5 years) assessed for nutritional status with the MNA (full version), and screened with the MEONF-II. The MEONF-II includes assessments of involuntary weight loss, Body Mass Index (BMI) (or calf circumference), eating difficulties, and presence of clinical signs ofUN. MEONF-II data were analysed by Receiver Operating Characteristics (ROC) curves and the area under the curve (AUC); optimal cut-offs were identified by the Youden index (J=sensitivity+specificity–1).

Results: According to the MEONF-II, 41% were at moderate or high UN risk and according to the MNA, 50% were at risk or already undernourished. The suggested cut-off scores were supported by the Youden indices. The lower cut-off for MEONF-II, used to identify any level of risk for UN (>2; J=0.52) gave an overall accuracy of 76% and the AUC was 80%. The higher cut-off for identifying those with high risk for UN (>4; J=0.33) had an accuracy of 63% and the AUC was 70%.

Conclusions: The suggested MEONF-II cut-off scores were statistically supported. This improves the confidence of its clinical use.

Keywords: cut-off scores; MEOF; MNA; MEONF; nutritional screening; ROC-curve

Received: 16 May 2011; Revised: 30 June 2011; Accepted: 5 July 2011; Published: 29 July 2011

Citation: Food & Nutrition Research 2011, 55: 7289 - DOI: 10.3402/fnr.v55i0.7289

Food & Nutrition Research 2011. © 2011 Albert Westergren et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.


Introduction

Undernutrition (UN)is associated with poorer health, compromised ability to recover from medical conditions, and increased mortality (1). People at risk for or with manifest UN therefore need to be identified in order to initiate prevention or treatment. The most well-established tool for nutritional screening is the mini nutritional assessment (MNA), which consists of two parts: the first part (short form) is an initial screening tool (MNA-SF), whereas the full MNA provides a more detailed assessment (2, 3). Other examples of nutritional screening tools include the Nutritional Risk Screening (NRS) 2002 (4), Malnutrition Universal Screening Tool (MUST) (5), Nutritional Form for the Elderly (NUFFE) (6), and the recently developed minimal eating observation and nutrition form–version II (MEONF-II) (7, 8).

MEONF-II is based within an interdisciplinary nursing framework (7), including description of mealtime problems that are associated with needs for nutritional interventions and clinical outcome (9, 10). In addition to mealtime problems, classical signs of UN, i.e. low BMI and unintentional weight loss (11) are included in the instrument in order to facilitate detection of problems, take preventive actions, or initiate treatment (7). The rationale for developing the MEONF-II was that earlier instruments do not at all, or to a limited extent, take eating difficulties into account in assessing nutritional risk. Furthermore, as nutritional screening typically is conducted by nurses it is important that the instrument is experienced as user-friendly and relevant from the nurses’ perspective in order to achieve a successful clinical implementation (7, 8).

In two previous studies, the sensitivity, specificity, and user-friendliness of the MEONF-II, MUST, and NRS-2002 in relation to the MNA were analysed among hospital inpatients (7, 8). Results suggested that the MEONF-II is easy and relatively quick to use, and its accuracy was favourable to that of the MUST (82 vs. 78%) (7) as well as to the NRS 2002 (68 vs. 55%). However, the MEONF-II cut-off scores used in the two studies (7, 8) were based on clinical reasoning, not on statistical criteria. Here we fill this gap by evaluating the MEONF-II cut-off scores in relation to the MNA classifications among hospital inpatients using statistical criteria. The primary aim was to identify the lower cut-off score for the MEONF-II; that is, the score that best separates those identified at risk for UN or being undernourished from those who are well-nourished according to the MNA. A secondary aim was to explore the higher cut-off score for the MEONF-II; that is, the MEONF-II cut-off for identifying UN (according to the MNA) among those at any risk according to the lower MEONF-II cut-off.

Present investigation

Materials and methods

The study included 187 hospital inpatients from two earlier studies (7, 8). Approval was obtained from the local ethics council and all participants provided informed consent. The first sample consisted of 100 orthopaedic, stroke, and cardiology inpatients aged 65 or older (7). The second sample consisted of 87 adult (18+ years old) inpatients (of whom 15 were <65 years old) receiving inpatient care at four hospital departments (stroke, surgery, orthopaedic, and geriatric medicine) (8).

Mini Nutritional Assessment (MNA)

The MNA was developed for use among elderly patients (≥65 years) (12). The full MNA consists of 18 items with a maximum possible total score of 30. The MNA score indicates three different levels of nutritional status: well-nourished (30–24 points), at risk for UN (23.5–17 points), and undernourished (<17 points) (13). The tool has been shown to have high sensitivity (96%), specificity (98%), and positive predictive value (PPV) (97%) when compared with extensive assessments of nutritional status (3) and has commonly been used as a comparator when testing other instruments (5–8, 14, 15) . Here we used the full 18-item MNA as the comparator for determination of cut-off scores for the MEONF-II.

Minimal Eating Observation and Nutrition Form–Version II (MEONF-II)

MEONF was based on the minimal eating observation form–version II (MEOF-II; available from www.hkr.se/meof) (10, 16) that includes three components of eating (food intake, swallowing/mouth, and energy/appetite). MEONF-II also includes unintentional weight loss, low BMI (<20 for 69 years or younger, or <22 for 70 years or older) (17), or calf circumference <31 cms (7, 18), and the presence or absence of clinical signs of UN (7). The full instrument is available online at www.hkr.se/meonf. All items are scored one except for unintentional weight loss and energy/appetite, which are scored two since such problems are strong predictors of UN (10). MEONF-II yields a total score ranging from zero to eight. Based on clinical reasoning, it has been suggested that a score of zero to two represents low risk for UN, a score of three to four is considered a moderate risk, and a score of five or more as high risk for UN (7). MEONF-II has shown a sensitivity of 0.73, specificity of 0.88, PPV of 0.81, negative predictive value (NPV) of 0.82, and an accuracy of 0.82 when compared with the MNA among elderly (65+ years old) hospital inpatients (7). Among adult (18+ years old) hospital inpatients the sensitivity was 0.61, specificity 0.79, PPV 0.82, NPV 0.57, and accuracy was 0.68 when compared with the MNA (8).

Procedure

Following written and oral information about the studies and the included assessment methods, 10 registered nurses conducted nutritional assessments during lunch or dinner. Nine of them had special responsibility for nutrition at their respective wards and one had overall responsibility for the data collection (7, 8). For more detailed information see Westergren and colleagues (7, 8).

Power

With an expected sensitivity/specificity of 0.75, and a 40% (±10%) expected prevalence (7), a sample size around n=180 would be adequate to estimate the sensitivity and specificity of a diagnostic test (19).

Analyses

Identification of optimal MEONF-II cut-off scores (for low vs. moderate/high risk for undernutrition and moderate vs. high risk for undernutrition, respectively) with the established MNA (full version) categorisations (well-nourished vs. at risk for undernutrition/undernutrition and well-nourished/at risk for undernutrition vs. undernutrition) as criteria was based on receiver operating characteristics (ROC) curves. In constructing a ROC-curve, sensitivity and specificity are determined for each possible cut-off point. The optimal cut-off point is where the overall number of errors (false-positives and false-negatives) are minimised (20). In addition, the area under the curve (AUC) was calculated for the ROC curves. The AUCs can range between 0 and 1; an AUC <0.5 indicates that a test performs worse than chance, whereas an AUC of 1 indicates perfect discrimination. For example, if the AUC is 0.8 and we randomly select two people, one who is undernourished and one who is not, the probability is 80% that the former will have higher score (here indicating higher risk) (20). The AUC should be ≥0.7 to be acceptable (21); AUCs between 0.7–0.9 and >0.9 are considered moderate and high, respectively (22). To select optimal cut-off scores, the Youden index (J=sensitivity+specificity-1) was calculated. The cut-off score associated with the highest J is considered to indicate the optimal cut-off point (23).

Finally, the identified cut-off scores were assessed regarding sensitivity, specificity, PPV, NPV, and accuracy. These indices provide values ranging from zero to one (or equivalently expressed as a percentage), where higher values are better (24, 25). The analyses were carried using PASW Statistics 18.0, MedCalc version 11.4.4.0, and GraphPad Instat version 3.06 for Windows.

Results

The mean age of the patients was 77.5 years and 57% were women. According to the previously suggested cut-off scores of the MEONF-II, 41% were at moderate/high UN risk; according to the MNA, 50% were at risk for/already undernourished (Table 1).


Table 1. Characteristics of the patients
  n=187
Age, mean (SD)a 77.5 (11.6)
Min–Max 23–98
Gender,%
 Male 43
 Female 57
Care setting,%
 Orthopaedic 30
 Stroke 29
 Cardiology 18
 Geriatric 13
 Surgery 10
Unintentional weight loss,% 35
Low BMI or calf circumference <31 cm,%b 29
Eating difficulties,%
 Food intakec 21
 Swallowing/mouthd 9
 Energy/appetitee 35
 Clinical signs,% 29
Undernutrition risk according to the MEONF-II,%f
 No/low risk 59
 Moderate risk 20
 High risk 21
Undernutrition according to the MNA,%g
 Well-nourished 50
 At risk for undernutrition 37
 Undernutrition 13
aFifteen patients were younger than 65 years.
bBMI <20 for <69 years or <22 for ≥70 years. Calf circumference replaced BMI in two patients.
cIncludes sitting position, manipulate food on plate, conveying food to mouth.
dIncludes chewing, coping with food in mouth, swallowing.
eIncludes amount food eaten, energy to complete a meal, appetite.
fMinimal Eating Observation and Nutrition Form–Version II. Internal attrition n=1.
gMini Nutritional Assessment. Internal attrition n=2.

According to the Youden index, the optimal cut-off point for identifying those at any nutritional risk (at risk for undernutrition/undernutrition according to the MNA) was >2 points, with an AUC of 0.80 (95% CI, 0.73–0.85) (Table 2). After excluding those at no/low risk according to the MEONF-II, the optimal cut-off point for identifying those with UN according to the MNA was >4 according to the Youden index, with an AUC of 0.70 (95% CI, 0.57–0.80) (Table 3).


Table 2. ROC-curve analysis including all cases identifying the optimal lower cut-off point (indicated with *) for the MEONF-II in comparison to the MNA
Criterion Sensitivity (95% CI) Specificity (95% CI) Youden index
>=0 1.0 (0.96–1.00) 0.00 (0.00–0.04) 0.00
>0 0.80 (0.71–0.88) 0.65 (0.55–0.75) 0.45
>1 0.77 (0.67–0.85) 0.72 (0.61–0.81) 0.49
>2* 0.67 (0.57–0.77) 0.85 (0.76–0.91) 0.52
>3 0.54 (0.44–0.65) 0.90 (0.82–0.95) 0.44
>4 0.37 (0.27–0.48) 0.93 (0.86–0.98) 0.30
>5 0.27 (0.18–0.37) 0.97 (0.91–0.99) 0.24
>6 0.13 (0.07–0.22) 0.99 (0.94–1.00) 0.12
>7 0.03 (0.01–0.09) 1.0 (0.96–1.00) 0.03
>8 0.0 (0.00–0.04) 1.0 (0.96–1.00) 0.00
Youden index=sensitivity+specificity–1.
MEONF-II=Minimal Eating Observation and Nutrition Form–Version II; MNA=Mini Nutritional Assessment.


Table 3. ROC-curve analysis identifying the higher cut-off (indicated with *) for MEONF-II, including only patients at moderate/high risk for undernutrition according to the MEONF-II, in comparison to MNA (well nourished/at risk for undernutrition vs. undernutrition) (n=76)
Criterion Sensitivity (95% CI) Specificity (95% CI) Youden index
>=3 1.00 (0.86–1.0) 0.00 (0.0–0.07) 0.00
>3 0.83 (0.63–0.95) 0.25 (0.14–0.39) 0.08
>4* 0.75 (0.53–0.90) 0.58 (0.43–0.71) 0.33
>5 0.58 (0.37–0.78) 0.73 (0.59–0.84) 0.31
>6 0.37 (0.19–0.59) 0.92 (0.81–0.98) 0.29
>7 0.12 (0.03–0.32) 1.00 (0.93–1.00) 0.12
>8 0.00 (0.0–0.14) 1.00 (0.93–1.00) 0.00
Youden index=sensitivity+specificity–1.
MEONF-II=Minimal Eating Observation and Nutrition Form–Version II; MNA=Mini Nutritional Assessment.

The lower and higher cut-off points had sensitivities of 67 and 75%, specificities of 85 and 58%, and accuracies of 76 and 63%, respectively (Table 4).


Table 4. Diagnostic performance of the MEONF-II lower cut-off and higher cut-off compared to the MNA
Number of patients
A B C D SENSa (95% CI) SPECb (95% CI) PPVc (95% CI) NPVd (95% CI) Accuracye
MEONF-II lower cut-off (>2 points) in relation to MNA (any risk), n=184 62 14 30 78 0.67 (0.57–0.77) 0.85 (0.76–0.91) 0.82 (0.71–0.89) 0.72 (0.63–0.80) 0.76
MEONF-II higher cut-off (>4 points) in relation to MNA, n=76f 18 22 6 30 0.75 (0.53–0.90) 0.58 (0.43–0.71) 0.45 (0.29–0.62) 0.83 (0.67–0.94) 0.63
MNA
Screening (MEONF-II) Positive Negative
Positive A B
Negative C D
aSENSitivity=A/(A+C).
bSPECificity=D/(B+D).
cPositive Predictive Value (PPV)=A/(A+B).
dNegative Predictive Value (NPV)=D/(C+D).
eAccuracy=A+D/(A+B+C+D).
fAfter exclusion of patients with no/low risk for undernutrition according to the MEONF-II.
MNA=Mini Nutritional Assessment; CI=Confidence Interval; MEONF-II=Minimal Eating Observation and Nutrition Form–version II; UN=undernutrition.

Discussion

This study sought to determine optimal cut-off scores for the MEONF-II and found that previously suggested cut-off points based on clinical reasoning were statistically supported when using the MNA as the comparator criterion. The accuracies and AUCs found here were acceptable and indicate that the MEONF-II has sufficient ability to discriminate between patients with low, moderate, or high risk forUN. These observations provide additional support for the usefulness of the MEONF-II in research and clinical practice as an interpretable and user friendly interdisciplinary tool for identifying patients at risk forUN while also taking eating difficulties into account.

The sample size was calculated to achieve enough power in the statistical analyses. However, it should be noticed that the sample was insufficient for deciding the higher cut-off with enough power (19), due to the exclusion of patients at no/low nutritional risk according to the MEONF-II. The analysis for the higher cut-off, therefore, has to be considered as explorative and suggestive of the appropriate higher cut-off point. However, the fact that the identified higher cut-off coincides with that suggested based on clinical reasoning supports its usefulness. In addition, it should be remembered that the main purpose with screening is to identify people at risk and not to decide whether it is a low or high risk and that any case being at risk needs a more detailed assessment. From that perspective, the most important cut-off point is the lower one, which identifies any case at risk of UN irrespective of how severe the risk is.

The occurrence of moderate/high UN risk in this study, 41% according to MEONF-II, is somewhat higher than what has been found in other studies. For instance, in one study (n=2170) the point prevalence of moderate/high UN risk was 34, 26, and 22% in large-, middle-, and small-sized hospitals, respectively (26). One explanation for this could be that the mean age was lower (66–70 years) in that study compared to the sample studied here (77 years). In another large study, including 12 countries, the prevalence of UN among 1,384 older hospital inpatients (mean age: men, 81.2 years; women, 82.9 years) was 38.7% according to the MNA, while another 47.3% were at risk for undernutrition and 14.0% were well-nourished (27). The prevalence found in that study is higher than found here (50% at risk or already undernourished according to the MNA). Thus, through this perspective, the sample studied here appears representative of, at least, older in-hospital patients in general but perhaps not the oldest-old.

The feasibility of using MNA as a ‘gold standard’ can be discussed even though it is commonly used as the comparator for nutritional assessment (5–8, 14, 15). The full MNA as used in this study represents a detailed and in-depth assessment that agrees strongly with extensive investigations of nutritional status (2, 3). A strength with the MNA is that it detects risk of malnutrition at an early stage, i.e. when albumin levels and BMI are still normal (12), which increases the possibility to take preventive actions. A potential disadvantage with the MNA is that it was developed for people 65 years or older (12). However, a previous study found no relevant differences in sensitivity and specificity of the MEONF-II in comparison to the MNA when including the full sample or when excluding the younger subsample (8).

An advantage of using the MEONF-II compared to other screening tools is that it also identifies actual problems for which immediate actions can be taken. Eating difficulties in the screening is especially important as such difficulties are predictors of the need for nutritional intervention as well as weight loss, length of hospital stay, and need for higher level of care after discharge from hospital (9, 10, 28, 29). The MEONF-II may thus facilitate the identification of people at risk for undernutrition while simultaneously identifying potential underpinning problems and subsequent interventions. For example, if low energy levels/poor appetite is identified, the provision of protein- and energy-enriched food and food supplements might be important and in cases of swallowing/mouth problems, adaptation of food consistency is an important intervention (10). By identifying eating difficulties, nurses can also initiate consultation with other professionals, depending on the type of problems, such as physiotherapists (adjustment of sitting position), occupational therapists (adapting cutlery), dietician (increasing energy intake/appetite), speech therapists (improve swallowing), or dental hygienist (if problems with chewing).

Conclusion

The cut-off scores based on clinical reasoning were confirmed statistically and validated the current version of the MEONF-II. Together with previous studies of the MEONF-II, these observations support the usefulness of the MEONF-II as an interpretable and user friendly tool for identifying people at risk for undernutrition while also providing a means for targeted interventions.

Acknowledgements

The authors wish to thank all participating patients and nurses for their cooperation. The study was supported by the Swedish Research Council and the Skåne County Council's Research and Development Foundation. The study was conducted within the Clinical Assessment Research and Education (PRO-CARE) group, Kristianstad University.

Conflict of interest and funding

The authors have declare no conflicts of interest

References

  1. Dey DK, Rothenberg E, Sundh V, Bosaeus I, Steen B. Body mass index, weight change and mortality in the elderly. A 15 y longitudinal population study of 70 y olds. Eur J Clin Nutr 2001; 55: 482–92. [Crossref]
  2. Guigoz Y, Vellas B, Garry PJ. Assessing the nutritional status of the elderly: the Mini Nutritional Assessment as part of the geriatric evaluation. Nutr Rev 1996; 54: S59–S65. [Crossref]
  3. Vellas B, Guigoz Y, Garry PJ, Nourhashemi F, Bennahum D, Lauque S, et al. The Mini Nutritional Assessment (MNA) and its use in grading the nutritional state of elderly patients. Nutrition 1999; 15: 116–22. [Crossref]
  4. Kondrup J, Rasmussen HH, Hamberg O, Stanga Z. Nutritional risk screening (NRS ): a new method based on an analysis of controlled clinical trials. Clin Nutr; 22 2002; 2003: 321–36.
  5. Stratton RJ, Hackston A, Longmore D, Dixon R, Price S, Stroud M, et al. Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of use of the ‘malnutrition universal screening tool’ (‘MUST’) for adults. Br J Nutr 2004; 92: 799–808. [Crossref]
  6. Soderhamn U, Flateland S, Jessen L, Soderhamn O. Norwegian version of the Nutritional Form for the Elderly: sufficient psychometric properties for performing institutional screening of elderly patients. Nutr Res 2009; 29: 761–7. [Crossref]
  7. Vallén C, Hagell P, Westergren A. Validity and user-friendliness of the the Minimal Eating Observation Form-Version II (MEONF-II) for undernutrition risk screening. Food Nutr Res 2011; 55: 5801.
  8. Westergren A, Norberg E, Hagell P. Diagnostic performance of the Minimal Eating Observation and Nutrition Form–Version II (MEONF-II) and Nutritional Risk Screening 2002 (NRS 2002) among hospital inpatients–a cross sectional study. Submitted 2011.
  9. Westergren A, Ohlsson O, Hallberg IR. Eating difficulties in relation to gender, length of stay, and discharge to institutional care, among patients in stroke rehabilitation. Disabil Rehabil 2002; 24: 523–33. [Crossref]
  10. Westergren A, Lindholm C, Mattsson A, Ulander K. Minimal eating observation form: reliability and validity. J Nutr Health Aging 2009; 13: 6–12. [Crossref]
  11. Locher JL, Roth DL, Ritchie CS, Cox K, Sawyer P, Bodner EV, et al. Body mass index, weight loss, and mortality in community-dwelling older adults. J Gerontol A Biol Sci Med Sci 2007; 62: 1389–92.
  12. Guigoz Y, Lauque S, Vellas BJ. Identifying the elderly at risk for malnutrition. The Mini Nutritional Assessment. Clin Geriatr Med 2002; 18: 737–57. [Crossref]
  13. Bauer JM, Kaiser MJ, Anthony P, Guigoz Y, Sieber CC. The Mini Nutritional Assessment–its history, today's practice, and future perspectives. Nutr Clin Pract 2008; 23: 388–96. [Crossref]
  14. Bauer JM, Vogl T, Wicklein S, Trogner J, Muhlberg W, Sieber CC. Comparison of the Mini Nutritional Assessment, Subjective Global Assessment, and Nutritional Risk Screening (NRS 2002) for nutritional screening and assessment in geriatric hospital patients. Z Gerontol Geriatr 2005; 38: 322–7. [Crossref]
  15. Charlton KE, Kolbe-Alexander TL, Nel JH. Development of a novel nutrition screening tool for use in elderly South Africans. Public Health Nutr 2005; 8: 468–79. [Crossref]
  16. Hansen T, Kjaersqaard A, Faber J. Measuring elderly dysphagic patients’ performance in eating–a review. Disabil Rehabil 2011 (Epub ahead of print).
  17. SWESPEN. Nutritionsbehandling i sjukvård och omsorg [Nutritional treatment in care and service]. Swedish Society for Clinical Nutrition and Metabolism; 2006.
  18. Kaiser MJ, Bauer JM, Ramsch C, Uter W, Guigoz Y, Cederholm T, 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: 782–8. [Crossref]
  19. Malhotra RK, Indrayan A. A simple nomogram for sample size for estimating sensitivity and specificity of medical tests. Indian J Ophthalmol 2010; 58: 519–22. [Crossref]
  20. Streiner DL, Norman GR. Health measurement scales: a practical guide to their development and use. 4th ed. Oxford: Oxford University Press; 2008.
  21. Terwee CB, Bot SD, de Boer MR, van der Windt DA, Knol DL, Dekker J, et al. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol 2007; 60: 34–42. [Crossref]
  22. Fischer JE, Bachmann LM, Jaeschke R. A readers’ guide to the interpretation of diagnostic test properties: clinical example of sepsis. Intensive Care Med 2003; 29: 1043–51. [Crossref]
  23. Bewick V, Cheek L, Ball J. Statistics review 13: receiver operating characteristic curves. Crit Care 2004; 8: 508–12. [Crossref]
  24. Altman DG. Practical statistics for medical research. 2nd ed. London: Taylor & Francis; 2009.
  25. Loong TW. Understanding sensitivity and specificity with the right side of the brain. BMJ 2003; 327: 716–19. [Crossref]
  26. Westergren A, Wann-Hansson C, Borgdal EB, Sjolander J, Stromblad R, Klevsgard R, et al. Malnutrition prevalence and precision in nutritional care differed in relation to hospital volume–a cross-sectional survey. Nutr J 2009; 8: 20. Available from: http://www.nutritionj.com/content/8/1/20
  27. Kaiser MJ, Bauer JM, Ramsch C, Uter W, Guigoz Y, Cederholm T, et al. Frequency of malnutrition in older adults: a multinational perspective using the Mini Nutritional Assessment. J Am Geriatr Soc 2010; 58: 1734–8. [Crossref]
  28. Westergren A, Karlsson S, Andersson P, Ohlsson O, Hallberg IR. Eating difficulties, need for assisted eating, nutritional status and pressure ulcers in patients admitted for stroke rehabilitation. J Clin Nurs 2001; 10: 257–69. [Crossref]
  29. Westergren A, Unosson M, Ohlsson O, Lorefalt B, Hallberg IR. Eating difficulties, assisted eating and nutritional status in elderly (> or=65 years) patients in hospital rehabilitation. Int J Nurs Stud 2002; 39: 341–51. [Crossref]

*Albert Westergren
School of Health and Society
Kristianstad University
SE-291 88 Kristianstad
Sweden
Tel: +46 (44) 208550
Email: Albert.Westergren@hkr.se