The MAG guidelines and screening tool aim to improve patient outcome by early detection and treatment of undernutrition in the community. Dr Marinos Elia reports

Undernutrition is a serious public health problem. In the UK, more than 2 million individuals are underweight1 – defined as body mass index (BMI) in adults of <20kg/m2 by the Office of Population Censuses and Surveys.1 In addition, probably more than 2 million patients suffer from undernutrition according to the criteria used by the Malnutrition Advisory Group (MAG).2

In November 2000, Guidelines for the Detection and Management of Malnutrition were launched by MAG, under the auspices of the British Association for Parenteral and Enteral Nutrition (BAPEN). The guidelines focus on a new nutritional screening tool, linked to a care plan.

Why were a screening tool and guidelines needed?

Undernutrition can have avoidable adverse effects on physical, psychological, and behavioural function. It predisposes to disease, delays recovery from illness, and reduces wellbeing and ability to undertake work, with major economic consequences.

Much of the undernutrition goes unrecognised in hospital inpatients (up to 70%)3 and outpatients (45–100%),4 in nursing homes (almost 100% according to some studies),5 and in the community, where it affects patients with chronic diseases, the elderly, those recently admitted and discharged from hospital, and the poor and socially isolated.

Inadequate priority is given to undernutrition. A recent MORI survey6 commissioned by the MAG found that 74% of GPs received no undergraduate training in nutrition and 60% would welcome training for detecting and managing undernutrition. There is therefore a need to change practice.

Existing tools for use in the community were reviewed and found to have drawbacks. None of the 23 that were assessed met the combined recommendations for guideline development laid down by the Royal College of General Practitioners (RCGP), British Dietetic Association, Scottish Intercollegiate Guidelines Network (SIGN) and others.

Most of the existing tools had not been tested for reliability/validity and had not been developed using a multi-disciplinary input (deemed necessary because undernutrition crosses medical disciplines and involves a wide range of health professionals).

Virtually none of them appeared to have been assessed by independent external referees before piloting or submission for publication.

Furthermore, there are three practical problems with existing tools.

  • About half of the published tools that aim to detect undernutrition in the community are not linked to a care plan.
  • Some of the existing tools were developed for specific patient groups, and are probably not appro- priate for a wide range of patients.
  • Different tools for detecting undernutrition in the community have employed widely different criteria, often with inadequate justification. As a result, a widely different incidence of undernutrition is established, even when the tools are applied to the same patients.

New information about malnutrition and its treatment has become available, which should be incorporated into new guidelines.

Content of the guidelines

The guidelines make the distinction between nutritional screening and assessment:

  • Nutritional screening is a rapid, general procedure (taking approximately 2 minutes). It can be applied to various groups of patients differing in age, disease and disease severity, and can be undertaken by a variety of health professionals, who may not be specialists in nutrition (e.g. doctors, nurses, health visitors).
  • Nutritional assessment, which often follows screening, takes longer (l0–20 minutes) because it involves a more in-depth evaluation of nutritional status by a specialist in nutrition (e.g. a dietitian).

Nutritional assessment may vary considerably depending on age, disease and disease severity, and specific problems or disabilities affecting the individual.

The MAG has produced the only nutritional guidelines that follow the combined recommendations of the RCGP, British Dietetic Association, SIGN and others for guideline development.

The MAG guidelines recommend the use of a nutrition screening tool (Figures 1 & 2,below) and suggest roles for various healthcare professionals.

Figure 1: Detection and management of adults who are at risk of undernutrition in the community
detection, categorization and treatment of malnutrition
Figure 2: Screening tool for adults at risk of malnutrition*
screening tool for adults at risk of malnutrition
screening tool for adults at risk of malnutrition

Grading the evidence

The report provides more than 200 references to support the guidelines and the strength of recommendations, which are based on those of the Agency for Health Care Policy and Research (AHCPR) (see Table 1, below).7

Table 1: Grading of recommendations according to the Agency for Health Care Policy and Research7
Grade A Requires at least one randomised controlled trial as part of the body of literature of overall good quality and consistency addressing the specific recommendation.
Grade B Requires availability of well-conducted clinical studies but no randomised clinical trials on the topic of recommendation.
Grade C Required evidence from expert committee reports or opinions and/or clinical experience of respected authorities. Indicates absence of directly applicable studies of good quality.

There is strong evidence (grade A) that nutritional intervention is of benefit in some categories of patients, such as those with established malnutrition.

There is also strong evidence (grade A) of benefit following correction of specific identifiable trace element, mineral and vitamin deficiencies.

There is less confidence about the timing of administration of complete supplements or snacks (grade C), or food flavouring (grade B).

The guidelines also provide two additional grades:

  • Y for obvious common-sense procedures used in routine clinical practice, e.g. diagnosis and treatment of the underlying disease
  • X for generally accepted criteria, such as. BMI <18.5 to designate high risk of chronic protein-energy undernutrition.

These criteria (X & Y) are supported by a large body of evidence and by international committees, such as the World Health Organization. It is difficult to apply the grades recommended by the AHCPR to certain types of data, such as observational data used to establish normal intraindividual variation in weight over time.

Before testing, the tool was reviewed by more than 30 independent referees, including members of the Royal College of Physicians, RCGP, Royal College of Paediatrics and Child Health, Royal College of Nursing, and the British Dietetic Association. It is considered to have face validity, content validity, and 'excellent' reliability (see Table 2, below).

Table 2: Agreement between health professionals in the categorisation of risk of undernutrition in three separate studies8
Study Health professionals undertaking assessment No. of patients Agreement no. (%) Kappa* Weighted kappa*
1 Nurse and healthcare assistant 44 42 (95.5) 0.888 0.932
2 Nurse and healthcare assistant 60 59 (98.3) 0.948 0.967
  Nurse and student nurse 60 60 (100.0) 1.000 1.000
  Student nurse and healthcare assistant 60 59 (98.3) 0.948 0.967
3 Doctor and nurse 50 50 (100.0) 1.000 1.000
* Kappa and weighted kappa are chance-corrected measures of agreement between two observers, who independently assess the same patients (0 = no agreement and 1.0 = perfect agreement). Unlike kappa, weighted kappa takes into account the degree of disagreement between observers. It has been suggested that values <0.4 represent poor agreement, 0.4 to 0.75 fair to good agreement, and >0.75 excellent agreement beyond chance. In separate studies (not shown) the agreement between the more objective components of the tool and the clinical impression formed by asking questions listed on the tool (see 'detection' on Figure 1) kappa was found to be 0.691 and weighted kappa 0.775.

How will the guidelines improve patient care?

Unrecognised malnutrition has no place in a modern, dependable health service. At present there is confusion about how to detect and manage malnutrition in the community, and uncertainty about the types of patients who are likely to benefit from nutritional intervention.

The MAG evidence-based guidelines address both issues. For example, patients with a BMI of <20kg/m2 are more likely to respond to complete liquid nutritional supplements than those with a BMI of >20kg/m2.

Furthermore, the use of the same tool to identify patients at risk of malnutrition in the community enables valid comparisons to be made about the effects of nutritional intervention.

The guidelines provide cut-off values for intraindividual weight changes. An unexplained unintentional weight loss of >5% over 3–6 months, which is continuing, suggests the presence of an underlying problem, which may need to be investigated.

An unintentional weight loss of >10% over 3–6 months is clinically significant.

Thus, routine measurement of weight change may alert a health professional to the presence of underlying disease. This in turn may lead to investigations to establish an early diagnosis and treatment.

Ultimately the guidelines and tool aim to reduce morbidity and improve wellbeing in malnourished patients or those at risk of developing malnutrition, by encouraging timely and appropriate nutritional intervention.

How will the guidelines and screening tool encourage best practice?

The screening tool has educational value for health professionals. Use of the tool will help detect malnutrition at an early stage and prevent the development of severe adverse effects.

The guidelines and screening tool are intimately linked to the overall care of the patient, which focuses on the underlying disease.

To make health professionals aware of the guidelines, relevant information will be widelyŒdistributed to primary and secondary healthcare professionals working in PCGs/trusts, and administrators in health authorities. The information is also being distributed to professional organisations, industry, the DoH, and members of Parliament.

Workshops for raising the profile of malnutrition and use of the screening tool are also planned.

  • Copies of Guidelines for Detection and Management of Malnutrition can be obtained from the British Association of Parenteral and Enteral Nutrition, PO Box 922, Maidenhead, Berks SL6 4SH (tel 01628 644160/ 644162; fax 01628 644105), price £10 in the UK (including p&p) and £13 overseas, ISBN 1 899 467 459. Screening Tool for Adults at Risk of Malnutrition and Explanatory Notes for the Screening Tool for Adults at Risk of Malnutrition can be obtained free of charge, either from the above address or from the website of the British Association for Parenteral and Enteral Nutrition (


  1. Breeze E, Maidment A, Bennett N, Flatley J, Carey S. Health Survey for England 1992. OPCS. London: HMSO, 1994.
  2. Guidelines for Detection and Management of Malnutrition. A report by the Malnutrition Advisory Group, a Standing Committee of BAPEN. November 2000.
  3. Kelly IE, Tessier S, Cahil A et al. Still hungry in hospital: identifying malnutrition in acute hospital admissions. Q J Med 2000; 93:93-8.
  4. Miller DK, Morley JE, Rubenstein LZ, Pietruszka FM, Stome LS. Formal geriatric assessment instruments and care of older general medical outpatients. J Am Geriatr Soc 1990; 38: 645-51.
  5. Abbasi A, Rudman D. Observations on the prevalence of protein-calorie malnutrition in older persons. J Am Geriatr Soc 1991; 39: 1089-92.
  6. MORI Survey, 14-25 September 1998.
  7. US Department of Health and Human Services Public Health Service. Acute Pain Management: Operative and Medical Procedures and Trauma. Agency for Health Care Policy and Research Publications, Rockville, MD (AHCPR Pub 92-0038) 1992.
  8. Guidelines for Detection and Management of Malnutrition. A report by the MAG, a Standing Committee of BAPEN. November 2000: Section 9.

Guidelines in Practice, February 2001, Volume 4(2)
© 2001 MGP Ltd
further information | subscribe