Reducing waiting lists for hospital admission:

community nutrition services reduce the need for hospital beds

Margaret Wynn and Arthur Wynn

9 View Road, Highgate, London N6 4DJ UK

Abstract

The number of hospital beds needed is greatly increased by the malnutrition of patients before admission and after discharge. Malnourished patients spend longer in hospital and are more likely to die following surgery or other treatment, Extensive studies show that low blood serum albumin, indicating protein deficiency, is a major risk factor for morbidity and mortality amon- hospital patients. Community food and nutrition services are extending throughout the developed world as part of public health policies. Such services can reduce costs by reducing the need for hospital beds, can reduce waiting lists and save the lives of many hospital patients. Preventive nutrition services can give financial benefits much exceeding costs.

Malnutrition of patients when admitted to hospital

"In Britain today about 40 per cent of adults admitted to hospital are undernourished". These are the opening words of a monograph by Davis and Bristow published by the Nuffield Trust in 1999 (1). Forty per cent of adults admitted to hospital per year is more than 3 million adults in the United Kingdom. This is not a problem only in the United Kingdom. An American paper, for example, says that "in eight studies of hospitalised adult patients 40 per cent to 55 per cent were found to be malnourished" (2). This prompts the question: How does nutritional deficiency increase the need for hospital beds?

Three ways in which demand for more hospital beds is increased by nutritional deficiency

Nutritional deficiency is called comorbidity in some papers and secondary malnutrition in others, on the assumption that it is superimposed on an existing illness. Nutritional deficiency before hospital admission can cause aggravation of most illnesses and so increase the demand for hospital beds. An increase in the number of patients needing hospital treatment is, therefore, the first way in which nutritional deficiency increases the need for more beds.

The second way, shown in many studies, is that patients admitted to hospital already sufferin- from nutritional deficiencies stay longer in hospital on average and so increase the demand for more beds.

The third way, again shown in many studies, is that patients who do not have an adequate diet after discharge from hospital are more likely to require readmission, which is another important way in which the demand for hospital beds is increased,

The cost to hospitals of malnourishment

A study from Ohio State University summarizes (2):

"Hospital charges range from 35 to 75 per cent higher in malnourished patients than in well-nourished patients, and the higher charges are directly attributable to increased length of stay and increased use of resources for treatment of complications associated with malnutrition."

The cost of 467 patients by their nutritional status from the Medical Center of the University of Illinois is shown in Table 1 (3):

Table 1

On admission 55 per cent of these patients were moderately or severely malnourished and 60 per cent at discharge.

A study from the University of Pennsylvania reported an average increase in hospital charges of $12, 542 per patient from infection following a surgical operation. A post-operative fever without wound infection cost an extra $91,145 per patient (4).

The association of malnutrition and infective illness of patients in hospital The US Surgical Risk Study reported in 1999 on 54,215 surgical cases and found that malnutrition before and after surgery greatly increased the risk of wound infection, urinary infection, systemic sepsis and pneumonia (5). These were not the only complications significantly associated with malnutrition in this study of surgical patients, but the risk of infective illness exceeded the risk of other complications. Many other studies show an increase in infection of malnourished patients in hospital with heart disease, cancer and other disorders not involving sureery (6).

The conclusion from the literature is that the risk of morbidity and mortality from malnutrition among hospital patients is so high that it 'ustifies measures to ensure that wherever possible patients have good normal nutritional status before admission to hospital and especially before admission for surgery.

Serum albumin, the primary index and benchmark of malnutrition

As long ago as 1968 the World Health Organisation published a monograph on the assessment of nutritional status of a community and concluded that blood serum albumin level should be used as the primary index of protein calorie malnutrition in patients both in and out of hospital ( PCM) (7). This recommendation has stood the test of time. The results of preoperative serum albumin tests on 54,215 patients in the US National Surgical Risk Study were reported in 1999 in Archives of Surgery and concluded (5) :

"Serum albumin concentration is a better predictor of surgical outcomes than many other preoperative patient characteristics. It is a relatively low cost test that should be used more often as a prognostic tool to detect malnutrition and the risk of adverse surgical outcomes."

Serum albumin was found in this study to be the strongest predictor out of 62 preoperative patient risk variables of both mortality and morbidity for surgery as a whole and for a range of sub-specialities. Serum albumin level was also the best predictor of infective illness following an operation.

About 66 per cent of all serum protein in healthy adults is albumin. The mean serum albumin is about 46 g/l (8). The US National Surgery Fisk Study found that as serum albumin fell from 46 g/l to less than 21 g/l there was an exponential increase in mortality from 1 to 29 per cent and in morbidity from 10 to 65 per cent (5). Many smaller studies have shown similar results. A study from the Boston Center for Clinical Computing of 15,511 patients found serum albumin on admission to hospital to be the best predictor of length of hospital stay and said that " serum albumin levels should always be measured." (9)

Serum albumin level is a useful general indicator of protein status, adequacy or deficiency. It does not follow that a patient with a low serum albumin level will benefit from albumin supplementation by injection. The circumstances in which albumin supplementation by injection may be desirable have been the subject of a long controversy, which is not within the scope pf the present paper.

If serum albumin level is used as the basic indicator of malnutrition are not some other important deficiencies missed? Studies have shown that if serum albumin level was treated as a confounding variable then other nutritional risk factors ceased to be significantly correlated with morbidity and mortality (10). This is mainly explained by the correlation in all human diets of protein intake with other nutrients such as the B vitamins, zinc, magnesium or phosphorus, often called protein's fellow travellers. Serum albumin levels are associated with intake of protein rich foods.

Nutritional deficiencies in diet other than protein deficiency can cause depression of serum albumin levels. For example iodine deficiency has been found to depress serum albumin (11). Iodine deficiency causes hypothyroidism. Depressed thyroxine levels depress albumin synthesis. If serum albumin levels are found to be depressed a test for a deficiency of nutrients other than protein may be merited. It is then necessary to decide at what serum albumin level or threshold further action is desirable. A serum albumin level of 40 g/l is in use as the threshold above which no action is indicated (12) (13) (14), The 40 g/l threshold is in use at all stages both before admission, during hospital stay and after discharge. Some studies consider 40 g/l is rather too low a threshold and some Japanese papers, for example, use 41 g/l as a threshold (15). A serum albumin level below 40 or 41 g/l indicates that further examination of nutritional status is desirable and that the diet of the patient should be designed to remedy any deficiencies.

The second lower albumin threshold

An extensive literature then suggests that a second threshold is needed below which urgent action should be taken to improve nutrition. The Boston study of 15,511 patients found that at serum levels below 34 g/l on admission 14 percent of patients died while above this threshold only 4 per cent died (9). Protein catabolism, or decrease in body protein, continued for too long is a major cause of death. Serum albumin level of 34 or 35 g/l is widely accepted as the second threshold.

The influence of nutritional status on clinical outcome after acute stroke is reported in a study from Birmingham and Leicester which found that low serum albumin level (< 34 g/l) was "a strong and independent" predictor of mortality at 3 months and also a "good predictor" of the degree of disability and handicap at the time of admission and other times during the hospital stay (16). The majority of patients had a serum albumin level below 40 g/l on admission.

A Norwegian study, for example, reported that 26 per cent of all patients on admission to hospital were suffering from heart disease or stroke and 48 per cent of these patients with vascular disorders were undernourished on admission (17). This study also compared patients admitted to hospital with a sample of home living population matched for age and sex and found that of the hospital patients 38.3 per cent of men and 38.0 per cent of women had albumin levels below 35 g/l whereas only 3.8 of men and 8.0 per cent of the women in the home population had levels below 35 g/l. The conclusion of this Norwegian study was that aggravation of vascular diseases by malnutrition, resulting in low body protein or protein catabolism, is an important cause of patients needing hospital admission for heart disease or stroke and a cause of mortality,

It is, however, a limitation of the serum albumin test that serum concentration changes only slowly in response to changes in a patient's health, nutrition or treatment. This is partly a consequence of the fact that the considerable mass of blood albumin exceeds that of all other blood proteins taken together and of the rate at which albumin can be synthesized by the liver. Other tests are needed to follow a patient's progress or decline.

Prealbumin and other tests of nutritional status

Serum albumin tests meet the important need for tests that permit comparison between different populations,f or example of hospital catchment areas or electoral wards. However,, in clinical practice for the individual patient, either at home or in hospital, it Is necessary to use other tests in addition which record patient response within two or three days to a change in circumstances or treatment. Many other tests have been studied and are in use. A study from the Gastroenterology Research Foundation, Indianapolis, for example, first of 100 patients and later of 150 patients, concluded that "the best three indicators of malnutrition were serum albumin level, lymophocyte count and percentage weight loss."(18)

A number of studies use anthropometric indices for assessing nutritional status. Anthropometric measurements avoid the need for blood samples and blood analysis. However, low body mass index (BMI) may be a consequence of low body fat or low body protein, which have quite different physiological and medical consequences. Major surveys, each involving more than a million participants, in Europe and the USA have shown increasing mortality as BMI falls below 22 kg/m2 Low BMI is, therefore a valuable and established risk indicator, but if BMI is below 22 kcr/M2 biochemical measurements are desirable to find out which nutrient deficiencies are causing the low body weight, Particular individuals may have a low serum albumin level indicating protein deficiency with a BMI above 22kg/m2. Some studies use a BMI below 20kg/m2 as a definition of malnutrition without, however, giving an explanation for the choice of this figure or any discussion of its limitations (21). Moreover, a French paper notes (22):

"In older patients weight cannot be easily and reliably measured due to immobilization and illness. The high prevalence of vertebral osteoporosis and spinal deformities impair stature measurement."

This French comparative study concluded that mid-arm muscle circumference was the best anthropometric indicator and was reliable, inexpensive and non-invasive and gave the most statistically significant correlation with death rate of the anthropometric indicators. While the anthropometric indicators may be useful they do not respond quickly to a change of circumstances or treatment and there has been much further study of blood serum components, particularly blood proteins, and there are studies using retinol-binding protein (23), transferrin (24), haemoglobin and hematocrit (25) and prealbumin which is also called transthyretin (26 and see Note 1).

The prealbumin test is preferred by the largest number of published studies as a useful and inexpensive test which responds rapidly to changes in a patient's nutritional status (27) (28). The median prealbumin level in blood serum is about 300 mg/l and there is no general catabolism of body protein until the value falls to below about 190 to 200 mg/l (29). Thresholds of malnutrition are recorded using prealbumin levels of 160, 170 or 180 mg/l but the rate of rise or fall of prealbumin levels is generally of greater clinical interest than the absolute level (30 ) (31) (32) (33) (34) (35). Whether or not a patient's body protein is decreasing can be determined by a series of prealbumin tests on successive days and the tests will contribute to the evidence as to whether a patient is recovering or deteriorating.

Intervention to remedy the malnutrition of hospitalised patients

The Patients Association published a statement entitled Cateringfor Patients in Hospital in 1993. The excellent recommendations do not include any reference to the special requirements of hospital patients for protein. The diet desirable for hospital patients is essentially different from that of normal, healthy people of the same age and sex.

There are several important reasons why hospital diets need to be different. The first reason is that hospital patients spend much time in bed or in a chair in a warm ward and have low energy requirements because of inactivity. Energy requirements of both men and women asleep or inactive are about 30 per cent below their average requirement when living their ordinary lives and 50 per cent or more below that of men or women engaged in manual labour. When in hospital patients do not generally eat more than is necessary to meet their energy requirements. If the nutrient content of hospital food is the same as that of the average citizen, or is based on recommendations for most people in health, the hospital patient will consume 20 or 30 per cent less not only of energy but of protein and other nutrients. However, most illness increases the requirement for protein and some other nutrients.

When the body suffers damage or surgical shock the body loses protein. Even following a minor operation the loss may be 100 to 200g and may exceed 1,000g following a major operation. Many people have substantial protein reserves and can resist the effects of protein loss associated with illness for a substantial period. However, in the absence of such protein reserves, or when reserves are exhausted, the blood levels of protein, recorded by serum albumin measurements, fall to dangerous levels.

It is impairment of a patient's immune system which has the greatest impact on morbidity and mortality. A fall in serum albumin levels has been shown to be associated with a fall in important serum proteins of the immune system which are essential for resistance to infection and to many toxic substances (36). Low serum albumin has been noted above as the main deficiency in the blood associated with infection of patients in hospital. Low serum albumin is evidence of low levels of complement, a word used to describe a system of proteins found in blood (3 6). A deficiency of at least IO proteins of complement have been associated with serious disease. The highest concentrations of these proteins is complement protein C3 which is normally present in serum at 1,200 mg/l, and serum C, level has been found to be a usefijl clinical measurement C3 is one of the many proteins of the immune system which illustrate the body's complete dependence on protein synthesis for defence against infections and indeed for survival.

The efficacy of supplementary feeding of patients in hospital has been the subject of randomized controlled trials for about 20 years. In 1983 it was reported from in the British Medical Journal that supplementary feeding in a population of 744 elderly women with fractured neck of femur resulted in improved nutritional status as shown by serum protein tests followed by a decline in mortality and morbidity, and a shortened hospital stay (37). These results were repeated in 1990, also for elderly women with fractured femur, in a controlled trial in Switzerland (38).

A later study in 1999 describes the introduction in an American hospital of an integrated plan providing nutrition care to patients in a 395-bed acute care hospital (39). This paper reports in 1996 and 1998 the results of a study begun in 1993 comparing patients receiving nutritional supplementation with controls and found that average length of stay was reduced from 10.1 to 8.1 days. In 1998 the incidence of complications was reduced from 73.3 to only 17.5 per cent and the 30-day readmission rate from 16.5 to 7.1 per cent. These are examples of many studies showing the efficacy of nutritional intervention after hospital admission.

Intervention to improve the nutritional status of patients before elective admission A paper published in January 2000 in the Journal of the American Dietetic Association describes such intervention and claims (40):

"To our knowledge, this is the first published article on preadmission nutrition screening by hospital clinicians of patients who are scheduled for hospital admission."

This paper concludes:

"Our findings indicate that preadmission nutrition screening has potential to improve patient outcomes by increasing nutrient intake before their hospital admission and reducing hospitalization length, and enhancing patient satisfaction."

The authors recommend in particular screening before elective surgery and they suggest that surgery should be delayed until the patient's nutritional status is satisfactory. Because morbidity and mortality are increased by malnutrition it is irrational for a patient to have to wait weeks or months for admission for surgery without any action during that time to correct nutritional deficiencies.

The admission of large numbers of malnourished patients to hospital is part of the wider problem of the appropriateness of hospital admission. A study in 1996 by a Research Unit of the Royal College of Physicians found 31 per cent of in-patient admissions to a London hospital to be inappropriate (41). The admission of malnourished patients to hospital for elective surgery is an important part of this expensive problem.

Intervention after discharge can reduce hospital readmission

A study by a team from Pennsylvania State University of 270,000 patients of the US Medicare programme for the elderly found that 24 per cent of all expenditure on Medicare hospital in-patients was attributable to readmissions (42). It was decided to investigate the reasons for the readmissions in the hope that preventive intervention might make it possible to reduce the overall health care costs. It was found that a serum albumin test at one month after discharge predicted readmission with 90 per cent accuracy. Most nonelective readmissions were found to happen within 3 months and it was not, therefore expedient to delay the serum albumin test for longer than one month after discharge. The probability of readmission at different levels of serum albumin is shown in Table 2 (42):

Table 2

The discharge of patients to an environment in which they are not adequately nourished is the major cause of hospital readmission which, in at least the large study in Pennsylvania, was responsible for nearly a quarter of hospital in-patient costs

But this is not the whole cost to hospitals. The Research Unit of the Royal College of Physicians found that 66 per cent of the last days of stay of patients in one London hospital were "inappropriate". A major reason for failure to discharge patients was that many patients needed help from social services after discharge and "many of them were still in hospital because they were waiting for the service to be organised." Other patients were waiting for a non-acute" bed in a nursing home or long-stay hospital. This is another way in whicil inadequate food and nutrition services in the community increase the need for hospital beds.

A study from the teaching hospital in Adelaide describes how home-based improvement in diet after hospital discharge of patients with congestive heart failure reduced unplanned readmissions compared with controls, reduced total length of hospital stay, reduced mortality and reduced hospital-based costs by Aust$ 10,000 per patient (43). A paper from the Abbott Laboratories, Columbus, says (2):

" Hospital charges are reported to be from 35 to 75 per cent higher for malnourished patients than for well-nourished patients....... But it is not enough to assess and identify malnutrition. Outcomes are improved and costs are saved only when appropriate intervention follows."

A task for the new Care Trusts

The NHS Plan 2000 begins one chapter with this aim," if patients are to receive the best care, then the old divisions between health and social care need to be overcome" (44). In some parts of the country health and social services are already working closely together while in other places there is little contact. The Government intends to establish Care Trusts "which will be able to commission and deliver primary and community health care as well as social care." Where local health and social care organisations have failed to establish effective joint partnerships "the Government will take powers to establish integrated arrangements." The success of Care Trusts will be judged by how far they reduce the number of inappropriate admissions to acute hospitals, how far they reduce the number of patients kept longer in hospital because there is a delay in providing adequate care after discharge and how far they reduce cases of readmission to hospital because patients do not receive care after discharge. Success will depend on whether the Care Trusts improve the nutritional status of patients before admission, particularly patients to be admitted for elective surgery, and whether they arrange appropriate diet and care to prevent patients being readmitted after discharge.

The high cost to the NHS of patient malnutrition is only one reason for making nutrition a major concern of the new Care Trusts. The US National Surgical Risk Study of 54,215 patients established that preoperative serum albumin level is a major predictor of mortality and morbidity particularly from infectious illness. The conclusion of this study, repeated in many others, is that protein deficiency or protein-energy malnutrition is the most likely deficiency increasing the risk of complications and death of hospital patients. Protein-calorie malnutrition is a major problem for the new Care Trusts.

A patient has a right to information about his own chances of recovery from an operation or treatment. Lord Woolf said in the Court of Appeal in 1998 (45):

" If there is a significant risk which would affect the judgement of a reasonable patient, then in the normal course it is the responsibility of a doctor to inform the patient of that significant risk."

A deficiency of protein in diet and low sei7um albumin is a significant risk.

The size of the problem

In 1997-8 in England 8.459 million patients were admitted to hospital for inpatient care (46). Assuming 40 per cent of hospital patients are malnourished when admitted, as suggested by the Nuffield Trust (1) and by British (47) and American (48) papers, about 3.4 million of these patients were malnourished. The 3.4 million malnourished patients did not include people whose nutritional status was kept up by being provided with one daily meal, for example the 1.07 million disabled persons estimated in the survey Disability in Great Britain to have help from social services every day in "preparing a hot meal"(49), nor the 40,000 disabled or elderly people receiving "meals on wheels" from the Women's Royal Voluntary Service (50). People receiving these services are presumably not found to be malnourished on admission to hospital except in special cases.

The need for more local food and nutrition services

The new Care Trusts will need to ensure that there is a local service to whom doctors can refer prospective hospital patients. This service must be able to assess a patient's nutritional needs and provide appropriate meals together with any other help the patient needs. Men and women living by themselves are at greater risk of food insecurity (17) Patients may be unable to shop or need assistance with food preparation (51). The service will be particularly necessary when it is planned for the patient to be admitted to hospital. The service will require many more qualified nutritionists and dieticians than are presently in post The " major programme of training and development" proposed in The NHS Plan will need to include nutritionists.

Many illnesses reduce appetite and need a special diet. Diabetes is an example of a growing problem in much of the world where a special diet is necessary (52). General practitioners need the support of nutritionists in devising and monitoring special diets. Home delivered meals or "Meals on Wheels" need a choice of menus which the doctor or nutritionist can recommend to a patient. Meals need to be adequate in protein. Both serum albumin and serum prealbumin levels are correlated with a patient's protein consumption over a range from 3O to 90 g/d. The new Care Trusts will need to employ an adequate number of trained care workers to help patients to follow this advice. All nations have ill-fed minorities of men and women who are not well enough fed to have the best outcome of medical care. Norwegian studies show that a member of this illfed minority is 10 times more likely to become a hospital patient than a member of the rest of the community and Norway is one of the healthiest nations (17).

The extra cost of malnourished patients to NHS hospitals is of the order of £15 to £20 billions a year. An ample pay-back would reward an annual expenditure measured in Is billions by the Care Trusts in extending and improving the food and nutrition services in the community which are needed by people who are discharged from hospital or who may become, or will shortly be, hospital in-patients.

Note 1.

Prealbuniin is a word in general use for a mixture of two blood proteins, transthyretin and orosomucol, and is somewhat misleading as there is little biological connection with albumin.

References

1. Davis AM, Bristow,A. Managing nutrition in hospital- a recipe for quality. Nuffield Trust Series No 8, II 8p, London: Nuffield Trust.

2. Gallagher-Allred CR, Voss AC, Finn SC, McCamish MA. Malnutrition and clinical outcome; the case for medical nutritional therapy. J Am Diet Assoc 1996; 96: 361-9.

3. Braunschweig CA. Creating a clinical registry: prospects, problems and preliminary results. J Am Dietet Assoc 1999-99/4:70.

4. Shulkin DJ, Kinosian B, Glick H, Glen-Puschett C, Daly J, Eisenberg JM. The economic impact of infections An analysis of hospital costs and charges in surgical patients with cancer. Arch Surg 1993: 128/4: 449-52.t

5. Gibbs J, Cull W, Henderson W, Daley J, Hur K, Khurl SF. Preoperative serum albumin level as a predictor of operative mortality and morbidity. Arch Surg 1999;134:36-42.

6. Potter J, Klipstein K, Reilly JJ, Roberts M. The nutritional status and clinical course of acute admissions to a geriatric unit. Age Ageing 1995; 24/2: 131-6.

7. World Health Organisation. The assessment of the nutritional status of the community. WHO monograph 53. Geneva- WHO, 1968.

8.Lentner C, ed. Geigy Scientific Tables 1984- 3: 141.

9. Herrmann FR, Safran C, Levkoff SE, Minaker KL. Serum albumin level on admission as a predictor of death , length of stay and readmission. Arch Intern Med 1992; 152: 125-30.

10. Agarwal N, Acevedo F, Leighton LS, Cayten CG, Pitchmoni CS. Predictive ability of various nutritional variables for mortality in elderly people. Am J Clin Nutr 1988; 48: 1173-8.

11. Centanni M, Maiani G, Parkes A]3, N'Diaye AM, Ferro-Luzzi A, Lazar-us JH. Thyroid homeostasis and retinol circulating complex relationships in a severely iodine-deficient area of Senegal. J Endocrinol Invest 1995; 18/8 -.608-12.

12. Salyoun NR, Jacques PF, Dallal G, Russell @. Use of albumin as a predictor of mortality in community dwelline, and institutionallsed elderly populations. Clin Epidemiol 1996; 49/9: 981-8.

13. Haupt W, Holzheimer RG, Riese J, Klein P, Hohenberger W. Association of low preoperative albumin concentrations and the acute phase response. Eur J Surg 1999- 165/4: 307-13.

14. Del Savio GC, Zelicof SB, Wexler LM, Byrne DW, Reddy PD, Fish D, Ende KA. Preoperative nutritional status and outcome of elective hip replacement. Clin Orthop 1996; 326: 152-61.

15. Ikematsu H, Nabeshima A, Yamaga S et al. Serum albumin level as a predictor of incidence of febrile episodes and mortality in hospitalized geriatric patients. Kansenshogaku Zasshi 1996-1 70/12-. 1259-65.

16. Gariballa SE, Parker SG, Taub N, Castleden CM. Influence of nutritional status on clinical outcome after acute stroke. Am J Clin Nutr 1998; 68: 275-8. 1.

17. Mowe M, Bohmer T, Kitidt E. Reduced nutritional status in an elderly population (>70y) is probable before disease and possibly contributes to the development of disease. Am J Clin Nutr 1994, 59-. '117-24.

18. Elmore NIF, Wagner DR, Knoll et al. Developing an effective adult nutrition screening tool for a community hospital. J Am Diet Assoc 1994; 94/10: 113-8, 119-21.

19. Waaler HT. Height, welaht and mortality: the Norwegian experience. Act Med Scand 1983 Supp] 679.

20. Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW. Body mass index and mortality in a prospective cohort of US adults. New Eng J Med 1999; 341/15: 1079-1105.

21. Eddington J, Boorman J, Durrant ER et al. Prevalence of malnutrition on admission to four hospitals in En-land. Clin Nutr 2000; 19/3: 191-5.

22. Constans T, BacqY, Brechot JF, Guilmot JL, Choutet P Lamisse F. Protein-energy malnutrition in elderly medical patients. J Am Geriatr Soc 1992, 40-1263-8.

23. Anderson CF, Moxness K, Meister K, Burritt MIF. The sensitivity and specificity of nutrition-related variables in relationship to the duration of hospital stay and the rate of complications. Mayo Clin Proc 1994-59/7- 477-83.

24. Johnson AM. Low levels of plasma- malnutrition or inflammation? Clin Chem Lab Med 1999;37- 91-6.

25. Carson JL, Duff A, Poses RM, Berlin JA, Spence RK et al. Effect of anaem' ia and cardiovascular disease on surgical mortality and disease. Lancet 1996; 348: 1055-66.

26. Ferguson RP, O'Connor, Crabtree B. Serum albumin and prealbumin as predictors of clinical outcomes of hospitalized elderly nursing home residents. Am J Geriatr Soc 1993; 41: 545-9.

27. Bernstein LH, Leukhardt-Fairfield CJ, Plebom W, Rudolph R. Usefulness of data on albumin and prealbumin concentrations in determining effectiveness of nutritional support. Clin Chem 1989; 3 5 - 271-4.

28. Mears E. Outcomes of continuous process improvement of a nutritional care program integrating serum prealbumin measurements. Nutrition 1996; 12: 479-84.

29. Tutin.MM, Wogt S, Dasse F, Leider Z. Utilization of prealbumin as a nutritional parameter. J Parent Ent Nutr 1985- 9- 709-11..

30. Erstad BL, Campbell DT, Rollins CJ, Rappaport VvD. Albumin and prealbumin concentrations in patients receiving preoperative parental nutrition. Pharmacotherapy 1994-1 14: 458-62.

31. Ferguson RP, O'Connor P, Crabtree B, Batchelor A, Mitchell J, Coppela D. Serum albumin and prealbumin as predictors of clinical outcomes of hospitalized elderly nursing home residents. J Am Geriati- Soc 1993: 41; 545-9.

32. Kuvibidila S, Warrier RP, Ode D, Yu L. Serum transferrin receptor concentrations in women with mild malnutrition. Am J Clin Nutr 1996; 63/4: 5 96-60 1.

33. Padillo FJ, Rodriguez M, Gallardo M et al. Changes in the visceral protein concentrations after internal binary drainage in patients with obstructive jaundice. Eur J Surgery 1999; 165/6: 550-5.

34. Potter MA, Luxton G. Prealbumin as a screening tool for protein-calorie malnutrition in emergency hospital admissions; a pilot study. Clin Invest Med 1999; 22/2: 44-52.

35. Sutor P. Prealbumin for screening and assessment. A quick, efficient and reliable tool. Health Care Food Nutr Focus 1998; 15/1: 6-7.

36. Suskind RM, ed. Malnutrition and the immune response. In: The complement system. Raven Press, 1977, p295-337.

37. Bastow NfD, Rawlings J, Allison SP. Benefits of supplementary tube feeding after fractured neck of femur; a randomised controlled trial. Br Med J 1982; 287: 1589-92.

38. Delmi, M, Rapin CH, Bengoa JM, Delmas PD, Vasey H, Bonj'our JP. Dietary supplementation in elderly patients with fractured neck of femur. Lancet 1990; 335: 1013-6.

39. Brugler L, DiPrinzio MJ, Bernstein L. The five-year evolution of a nutrition treatment program in a community hospital. J Qual Improve 1999; 25/4: 191-206.

40. Schwartz DB, Gudzin D. Preadmission nutrition screening; expanding hospital-based nutrition services by implementing earlier nutrition intervention. J Am Diet Assoc 2000;100/1 @81-7.

41. Houghton A, Bowling A, Jones 1, Clarke K. Appropriateness of admission and the last 24 hours of hospital care in medical wards in an east London teaching hospital. Int J Qual Health Care 1996; 8/6; 543-53

42. Friedmann JM, Jensen GL, Smiciklas-Wright H, McCamish MA. Predicting early nonelective hospital readmission in nutritionally compromised older adults. Am J Clin Nutr 1997; 65: 1714-20.

43. Stewart S, Vanderbroek AJ, Pearson S, Horowitz JD. Prolonged beneficial effects of a home-based intervention in unplanned readmissions and mortality among patients with congestive heart failure. Arch Intern Med 1999; 159/3: 257-6 1.

44 Sedley S The right to know. London Review of Books 200-22/15-15

45. Department of Health. The NHS Plan, July 2000.Cm 4818 - 1

46. UK 2000 in Ficures: Health England. London: Government Statistical Service.

47. McWhirter JP; Pennington CR. Incidence and recognition of malnutrition in hospital. Br Med J 1994; 308: 945-8.

48. Kamath SK, Lawler M, Smith AE, Kalat T, Olson R. Hospital malnutrition; a 33 hospital screening study. J Am Diet Assoc 1996- 86/2: 203-6.

49. Grundy E, Ahlburg D. Ali M, Breeze E, Sloggett A. Disability in Great Britain Department of Social Security Research Report No 94, London: Stationery Office, 1999.

50. Women's Royal Voluntary Service. Meals on Wheels. Abingdon, Oxon: VVRVS.

51. Webber CB, Splett PL. Nutrition risk factors in a home health population. Home Care Services Quarterly 1995; 15/3; 97-1 1 0.

52. Cheta DM. Preventing diabetes, theory, practice and new approaches. New York- Wiley 1999 187 pp.

53. Morgan DB, Newton UMV, Schorah CJ, Jewitt MA, Hancock MR, Hullin RP. Abnormal indices of nutrition in the elderly- a study of different clinical groups. Age Ageing 1986; 15:65-76.