- ESPEN
European Society for Clinical Nutrition and Metabolism
- MUST
Malnutrition Universal Screening Tool
- NICE
National Institute for Health and Clinical Excellence
- ONS
oral nutritional supplement
- RCT
randomised controlled trial
Healthcare systems with budgetary constraints focus on the use of cost-effective evidence-based treatments, particularly during economic downturns. Therefore, in the current economic climate, it is becoming increasingly important to identify conditions promptly and to treat them in the most clinically and cost-effective way. This applies to the treatment of disease-related malnutrition, a condition that is prevalent in hospital inpatients (10–60% of hospital admissions are at risk of malnutrition), outpatients (15–30%) and in individuals living in the community (including free-living older people (aged 65 years and over), those visiting general practice and those living in care homes or sheltered housing, 14–44%) (see Fig. 1 for prevalence of disease-related malnutrition identified with the ‘Malnutrition Universal Screening Tool’ (‘MUST’), see www.bapen.org.uk for screening criteria)(Reference Elia1–Reference Rust, Cawood and Walters10). At any given point in time, more than 3 million people in the UK are malnourished with most (approximately 93%) living in the community(Reference Elia and Russell11). Disease-related malnutrition is detrimental physiologically and clinically, impairing the quality of life and delaying recovery from illness(Reference Stratton, Green and Elia12). Data suggest that disease-related malnutrition doubles the risk of mortality in hospital patients and triples mortality in older patients (aged 65 years and above) in hospital and after discharge(Reference Stratton and Elia5, Reference Stratton, King and Stroud13). Disease-related malnutrition increases the use of health care (hospitalisations, General Practitioner visits and support required after hospital discharge)(Reference Elia1, Reference Elia, Stratton and Russell3, Reference Stratton, Green and Elia12–Reference Elia16) and latest estimates suggest that this condition costs the UK more than £13×109 annually (2007 data; Fig. 2)(Reference Elia, Stratton, Elia and Russell17).
Prevalence of disease-related malnutrition (using ‘Malnutrition Universal Screening Tool’ (‘MUST’) in patients in hospital and community settings in the UK. G1, gastrointestinal; Obs & Gynae; obstetrics and gynaecology; COPD, chronic obstructive pulmonary disease; GP, general practitioner.
Estimated costs of disease-related malnutrition in the UK(Reference Elia, Stratton, Elia and Russell17).
Despite the enormous public expenditure on malnutrition, the expenditure on treatments which may be prescribed for the management of this condition remains low. In 2007, the total cost of prescriptions in England was £8·372×109 (£8372·7×106). Of this, 1·88% (£157×106) was for nutrition (British National Formulary Section 9.4.2), a category including adult and paediatric oral and tube feeds, specialised infant and other formulae and fibre supplements. Specifically, about £99×106 (about 1·2%) was for oral nutritional supplements (ONS) and approximately £56×106 (approximately 0·7%) was for enteral tube feeds(18). The costs of other forms of nutritional support, including the costs of dietetic time to manage malnourished patients and the costs of other interventions (food snacks, food fortification) remain unclear and need to be elucidated.
Considering the enormous costs of disease-related malnutrition, a condition that is largely treatable, prompt identification with screening, followed by the most appropriate, effective, evidence-based treatment is recommended(Reference Elia1, Reference Elia, Zellipour and Stratton19, 20). The National Institute for Health and Clinical Excellence (NICE), recently released cost saving guidance, within the top four of which was nutrition support (in the form of oral nutrition support, tube feeding and parenteral nutrition). Specifically, NICE suggested that improving the systematic screening, assessment and treatment of malnourished patients (NICE CG32 guideline) could lead to an estimated cost saving of £28 472 per 100 000. NICE suggests that ‘If this guidance (CG32) was fully implemented and resulted in better nourished patients then this would lead to reduced complications such as secondary chest infections, pressure ulcers, wound abscesses and cardiac failure. Conservative estimates of reduced admissions and reduced length of stay for admitted patients, as well as reduced demand for General Practitioner and outpatient appointments indicate that significant savings are possible’(21).
A pragmatic programme of screening implementation (using ‘MUST’) in care homes in Peterborough Primary Care Trust highlights some such benefits(Reference Cawood, Smith and Pickles22). A programme involving education and training on malnutrition, screening and treatment using the framework of ‘MUST’, locally agreed care plans and monitoring, improved the documentation of nutritional status, the proportion of residents screened and the use of appropriate care plans. After the implementation of the screening programme, significant reductions in the number and duration of hospital admissions were observed, associated with a significant cost saving of about £600 per resident over 3 months(Reference Cawood, Smith and Pickles22). This could be equivalent to approximately £1×109 annually if extrapolated to the entire population of care home residents in the UK, although the costs of training and screening would need to be taken into account in order to estimate the overall net cost saving. Similar improvements of nutritional care and outcome have been observed in other settings where screening has been implemented. Rypkema et al. (Reference Rypkema, Adang and Dicke23) showed improvements in nutritional status and fewer complications in older (>65 years) inpatients with a multi-disciplinary intervention involving a screening programme and Stratton and coworkers found an increase in the documentation of nutritional information, use of care plans and shorter hospital stay in neurology wards after ‘MUST’ implementation (RJ Stratton, M Collins and M Elia, unpublished results). In another study, early nutritional screening and the treatment of malnourished patients reduced the length of hospital stay in malnourished patients who had low handgrip strength (i.e. frail patients). The authors concluded that to shorten the mean length of hospital stay by 1 d for all malnourished patients, a mean investment of 76 euros (£69) in nutritional screening and treatment would be needed (only about one-quarter of the cost of one bed day)(Reference Kruizenga, Van Tulder and Seidell24).
Implementing a screening programme is an effective way to identify those who do and do not need nutritional support, such as help with meals, feeding or shopping or artificial nutrition. Most patients who are malnourished or at risk of malnutrition can be managed using the oral route, often with a variety of dietary approaches including dietary modification (fortification, extra snacks, etc.), counselling by a dietitian and/or ONS(Reference Stratton, Thomas and Bishop25). However, with the rise of evidence-based practice, there is a need to demonstrate the effectiveness of these different strategies while making sure that they are used appropriately and their efficacy is maximised.
Although the focus of this review is the use of ONS, it is important to note that there is little formal evidence to support many of the other strategies used in the management of malnutrition, such as food snacks, fortification and dietary counselling(Reference Baldwin and Weekes26). Although the clinical and cost effectiveness of such strategies appear to be undocumented, they are widely advocated across health care settings, especially in primary care, in an attempt to reduce the use and costs of products which may be prescribed. Policies that focus only on reducing costs by withholding or stopping treatments are considered unsatisfactory, since they do not assess the clinical and economic consequences of this reduction and do not represent cost effectiveness(Reference Silk, Russell and Stratton27). Malnutrition that is untreated or ineffectively treated is likely to lead to a poorer quality of life, more complications, greater health care use and greater total health care expenditure. There is an additional financial implication for patients if prescribed forms of nutritional support are withdrawn or withheld, as the individual becomes responsible for acquiring and paying for the extra food or over the counter products that are recommended. An additional risk is that patients may then seek unsuitable or unsafe alternative forms of over the counter supplements, some of which may be unbalanced and potentially dangerous for their condition. In such patients, the end result could be increased costs to deal with unforeseen complications. Therefore, the implementation of such policies needs to be carefully considered and the effects monitored frequently to assess their clinical and cost effectiveness in both the short and long term.
Oral nutritional supplements
Many individuals while acutely or chronically ill are unable to consume the energy and/or nutrients they need from food alone, for a wide variety of reasons(Reference Stratton, Green and Elia12). These are broadly divided into two main causes: a variety of disease-related factors and the inadequate availability and quality of food. Disease-related factors are many and include anorexia, nausea and vomiting, changes in taste and smell, painful or dry mouth, disorders of swallowing, breathlessness, difficulties in chewing, fatigue and other specific side-effects of surgery, drugs or inflammatory conditions. Disease-related psychosocial problems may also impair food intake, including depression, anxiety, social isolation and unappealing meal environments. For individuals at home, inadequate availability of food may be due to inadequate resources (finances and cooking facilities), poor access to shops or physical difficulties affecting food preparation. Within institutions, the quality, timing and presentation of food may limit intake. ONS offer a useful and effective strategy to meet this deficit in nutrient intake. ONS are typically multi-nutrient containing a mix of macronutrients (protein, carbohydrate and fat) and micronutrients (vitamins, minerals and trace elements). Ready-made, nutritionally complete supplements are also energy dense (mostly containing 6·3 kJ (1·5 kcal)/ml to 10·1 kJ (2·4 kcal)/ml, 1·26 MJ (about 300 kcal) per serving (125–220 ml)) and provide a good source of protein (10–20 g per 1·26 MJ (about 300 kcal) serving) and a balance of micronutrients. Most are liquids that are available ready-made and so are convenient for individuals to use at home when ill and for health care professionals to administer in busy community and hospital settings, such as care homes and hospital wards. Powder supplements (to be reconstituted with whole milk before consumption) are also available on prescription or over the counter, although currently there is limited evidence to support their use(Reference Stratton, Green and Elia12, 20). The current evidence base supporting supplements is almost entirely restricted to ready-made types(Reference Stratton, Green and Elia12, 20) and highlights their effectiveness for use in patients with or at a risk of malnutrition, particularly older individuals, those who are acutely ill and perioperative patients. In the UK, there are a number of prescribed indications for the use of ONS which extend beyond disease-related malnutrition (and vary according to the supplement type; see Table 1 for a full list).
Recommendations for use of oral nutritional supplements in clinical practice(Reference Stratton and Elia32)
Evidence-base for oral nutritional supplements
A systematic review, which may or may not include a meta-analysis, has traditionally been considered as the best way of assessing the evidence-base for an intervention, such as ONS (type I in the hierarchy of evidence(Reference Higgins and Green28)). This is particularly so when the review is undertaken by those with a good understanding of the clinical use of the treatment, in combination with statistical expertise. In the last decade, an increasing number of systematic reviews have been undertaken to examine the effectiveness of oral nutrition support strategies in the management of malnutrition. The majority of these systematic reviews have focused on the use of ONS, the strategy for which there is the greatest number of individual trials available, usually in comparison with routine care (no nutritional support)(Reference Stratton, Green and Elia12, 20). Such reviews have had a major influence on the guideline development by national and international organisations, including non-governmental and governmental bodies. Indeed, there are already a substantial number of guidelines and standards referring to the use of ONS. Within Europe, these include guidelines from British organisations, such as the British Association for Parenteral and Enteral Nutrition, NICE and the Scottish Intercollegiate Guidelines Network (SIGN) as well as the European Society for Clinical Nutrition and Metabolism (ESPEN) (summarised in Table 2).
A summary of guidelines from British Association for Parenteral and Enteral Nutrition, National Institute for Health and Clinical Excellence, Scottish Intercollegiate Guidelines Network and European Society for Clinical Nutrition and Metabolism referring to oral nutritional supplement (ONS) use
* Levels of evidence: 1++, high-quality meta-analyses, systematic reviews of randomised controlled trial (RCT) or RCT with a very low risk of bias; 1+, well-conducted meta-analyses, systematic reviews or RCT with a low risk of bias; 2++, high-quality systematic reviews of case-control or cohort studies, high-quality case-control or cohort studies with a very low risk of confounding bias; 2+, well-conducted case-control or cohort studies with a low risk of confounding bias. 3, case reports; 4, expert opinion.
† Grades of recommendations: A, at least one meta-analysis, systematic review or RCT rated as 1++, and directly applicable to the target population or a body of evidence consisting principally of studies rated as 1+ directly applicable to the target population and demonstrating an overall consistency of results; B, a body of evidence including studies rated as 2++, directly applicable to the target population and demonstrating good overall consistency of results or extrapolated evidence from studies rated as 1++ or 1+; C, a body of evidence including studies rated as 2+, directly applicable with consistent results or extrapolated from studies rated as 2++; D, evidence level 3 or 4, extrapolated evidence from studies rated as 2+ or formal consensus; D (GPP), a good practice point, a recommendation for best practice based on experience. For more information, refer to www.nice.org.uk or www.sign.ac.uk.
‡ Oral nutritional support; includes ONS, dietary advice, food fortification, etc.
Clinical outcomes and oral nutritional supplements
Systematic reviews and meta-analyses(Reference Stratton, Green and Elia12, 20, Reference Stratton, Ek and Engfer29–Reference Milne, Avenell and Potter31) consistently suggest that ONS produce a range of clinical benefits including reduced mortality and fewer complications, such as wound and chest infections, pressure and leg ulcers. These are described in more detail in a ‘review of reviews’(Reference Stratton and Elia32) and a summary of these meta-analyses is shown in Table 3. In most trials, the daily reported intake of ONS was typically between 1·05 and 2·52 MJ (250–600 kcal), with one review(Reference Milne, Avenell and Potter31) suggesting that patients offered 1·68 MJ (>400 kcal)/d were more likely to benefit. The duration of supplementation varied depending on the patient group, from short periods in hospital (1 week) to much longer periods in the community (up to 2 years). The NICE(20) review suggested that the significant reduction in mortality and other improvements in clinical outcome seen with proprietary ONS in malnourished patients could not be demonstrated for other forms of oral nutritional support (e.g. food fortification and dietary advice), due to a lack of data. Similarly, a Cochrane review also highlighted the lack of data on the effects of dietary advice on clinical outcomes in both community and hospital settings(Reference Baldwin and Weekes26).
Effect of oral nutritional supplements (ONS) on mortality and complication rates (analyses of trials in a mix of patient groups; adapted from Stratton & Elia(Reference Stratton and Elia32))
RR, relative risk; RCT, randomised controlled trials.
* An OR or RR<1 indicates a lower incidence of mortality or complications in the supplemented patients than control patients. If the confidence interval does not pass 1·00, the result is statistically significant.
† In this meta-analysis the only complication assessed was pressure ulcers. In all other trials, a variety of complications were assessed (including wound infections, chest infections, pressure ulcers, etc.).
‡ Added since publication of the ‘review of reviews’(Reference Stratton and Elia32).
§ Data presented as absolute risk difference.
Systematic reviews and meta-analyses have highlighted significantly reduced mortality with ONS v. routine care (Table 3) in patients with a range of acute and chronic conditions, especially older patients(Reference Stratton, Green and Elia12, 20, Reference Potter, Langhorne and Roberts30–Reference Koretz, Avenell and Lipman33). One meta-analysis of studies across health care settings indicated that ONS produced a 30% reduction in mortality, predominantly due to effects in acutely ill, hospitalised patients(Reference Stratton, Green and Elia12, Reference Milne, Avenell and Potter31, Reference Koretz, Avenell and Lipman33). The effect of ONS on mortality may vary according to the nutritional status of patients. The NICE reviews only considered trials in malnourished patients, but both Stratton et al.(Reference Stratton, Green and Elia12) and Milne et al.(Reference Milne, Avenell and Potter31) indicated that ONS were more likely to reduce mortality in underweight/undernourished patients. One review indicated that the greatest reduction in mortality was observed in studies in which the mean BMI was <20 kg/m2(Reference Stratton, Green and Elia12). Meta-analyses of trials in the acute setting indicated significant reductions in mortality with ONS use in the undernourished (OR 0·66 (95% CI 0·49, 0·90), nine randomised controlled trials (RCT)) with less effect in the well-nourished (OR 0·99 (95% CI 0·81, 1·21), five RCT), although the results from one single trial of stroke patients, >90% of whom were well-nourished (FOOD trial(Reference Dennis, Lewis and Warlow34)), dominated this analysis(Reference Milne, Avenell and Potter31). However, it is difficult to interpret these results fully as individual studies defined malnutrition in different ways and some studies included a range of malnourished and well-nourished individuals. In addition, significant reductions in mortality have been reported in patients classified as well-nourished, for example by Stratton et al. (Reference Stratton, Green and Elia12) and by a small, specific meta-analysis of patients receiving ONS in long-term care (OR 0·46 (95% CI 0·25, 0·86), two RCT)(Reference Milne, Avenell and Potter31). Irrespective of nutritional status, the strongest and most consistent evidence for an effect of ONS on mortality is in older people and in the acutely ill(Reference Stratton and Elia32), where mortality is the highest. There was little evidence of an effect of ONS on mortality in free living individuals, where mortality rates are low. Studies with very large sample sizes and a prolonged duration of supplementation are needed to examine the existence of such effects as sufficiently powered trials are lacking. There is also a need to undertake high-quality studies to obtain a better understanding of the mechanisms by which ONS improve patient outcome, which may enable a more targeted use of ONS in specific health care settings and patient types. Although the methodology of individual trials (assessed in all reviews) was often judged to be poor, usually due to a lack of blinding, the effect of bias on unambiguous outcome measures, such as mortality, may be limited.
Systematic reviews of trials across different patient groups have highlighted substantial reductions in a range of complications with ONS(Reference Stratton, Green and Elia12, 20, Reference Stratton, Ek and Engfer29–Reference Milne, Avenell and Potter31, Reference Koretz, Avenell and Lipman33, Reference Cawood, Elia and Freeman35), including infections (wound, chest, urinary, etc.), incomplete wound healing, pressure ulcers and total complications (see Table 3 for meta-analysis results). One systematic review showed a significant reduction in the development of one specific complication (pressure ulcers) with ONS v. routine care (OR 0·75 (95% CI 0·62, 0·89), four RCT, n 1224) (Table 3)(Reference Stratton, Ek and Engfer29). In this meta-analysis, most studies used a liquid, high protein ONS for between 2 and 26 weeks, across hospital and long-term care. The other systematic reviews and meta-analyses mostly reported significant reductions in a range of complications with ONS use in hospital, community or combinations of health care settings(Reference Stratton, Green and Elia12, 20, Reference Milne, Avenell and Potter31, Reference Koretz, Avenell and Lipman33, Reference Cawood, Elia and Freeman36). In most cases, liquid, ready-made, multi-nutrient ONS were used, with reported intakes of 250–600 kcal (1·05–2·52 MJ)/d. In one systematic review including only high-protein ONS (containing >20% energy from protein), marked reductions in complications were also observed(Reference Cawood, Elia and Freeman36). Systematic reviews of trials in patients in the peri-operative period have consistently indicated fewer complications with ONS use before, during and after hospitalisation, including total complications, infectious complications, major complications and intra-abdominal/thoracic complications(Reference Stratton, Green and Elia12, Reference Stratton and Elia32, Reference Koretz, Avenell and Lipman33, Reference Stratton and Elia37). Similarly, in hip fracture patients, a Cochrane review(Reference Avenell and Handoll38) suggests significantly fewer patients with an unfavourable outcome (mortality and complications) (relative risk 0·52 (95% CI 0·32, 0·84), three RCT, n 139) with ONS v. routine care.
The reduction in complications with ONS does not appear to differ between studies in which the mean BMI is <20 from those in which it is >20 kg/m2.(Reference Stratton, Green and Elia12) Indeed, reductions in complications are observed in surgical patients (e.g. gastrointestinal surgery(Reference Stratton and Elia32, Reference Koretz, Avenell and Lipman33, Reference Stratton and Elia37)), who are not obviously thin but in whom poor nutritional intake in the post-operative period may have contributed to the development of complications.
Functional outcomes and oral nutritional supplements
Due to the wide range of functional outcomes measured in trials of ONS across different patient groups, it can be difficult to synthesise the evidence into a single meta-analysis of RCT. One systematic review that considered the impact of ONS on function in detail was Stratton et al.(Reference Stratton, Green and Elia12). Within this systematic review, a number of individual studies in both hospital and community patients were found to show significant improvements in functional measures with ONS in groups such as older people, patients with liver disease and those undergoing surgery. The functional improvements included muscle strength, quality of life, immune function, walking distances and activities of daily life. This review also reported that in chronically ill patients in the community, functional benefits were more likely to occur in underweight individuals (BMI<20 kg/m2) who gained weight (>2 kg) with ONS. Another review that attempted to integrate a functional outcome (muscle strength) from only five trials in older people into a meta-analysis reported no significant effects of ONS(Reference Milne, Avenell and Potter31). Further research is warranted to determine the effects of ONS on important functional measures, especially quality of life, and other ‘Patient reported outcome measures’, which are now incorporated into policies within the National Health Service in England.
Nutritional outcomes and oral nutritional supplements
Nutritional status
Systematic reviews and meta-analyses that assess nutritional status consistently indicate significant improvements with ONS(Reference Stratton, Green and Elia12, Reference Milne, Avenell and Potter31, Reference Baldwin, Parsons and Logan39). The most common indicator of nutritional status assessed has been body weight. All meta-analyses that examined this outcome appeared to report a significant improvement with ONS relative to routine care in all settings (hospital, long-term care and community)(Reference Stratton, Green and Elia12, Reference Milne, Avenell and Potter31). Meta-analyses by Milne et al. (Reference Milne, Avenell and Potter31) indicated significant improvements in the percentage weight change with ONS relative to routine care in older patients in long-term care (weighted mean difference 2·51 (95% CI 1·73, 3·20)%), in the community (2·25 (95% CI 1·72, 2·70)%) and in the hospital (1·75 (95% CI 1·12, 2·30)%) with a wide range of conditions, although it was unclear whether they were malnourished or not(Reference Milne, Avenell and Potter31). Another meta-analysis indicated a significant improvement in weight with ONS relative to dietary advice (weighted mean difference 1·09 (95% CI 0·29, 1·90) kg)(Reference Baldwin, Parsons and Logan39). In general, ONS use aided weight gain in chronically ill patients in the community setting and attenuated weight loss in studies in acutely ill patients in the hospital(Reference Stratton, Green and Elia12). Improvements in weight with ONS were seen more frequently in community trials in which the mean BMI was <20 kg/m2 (v. >20 kg/m2), but it was not possible to analyse the impact of ONS on patients who were losing weight, irrespective of BMI(Reference Stratton, Green and Elia12). The composition of weight gain achieved with ONS (e.g. lean tissue and fat mass) was either not assessed or unclear in many studies, although a few RCT indicated significant improvements in lean tissue or body fat. In most trials, upper-arm anthropometry was used, possibly as an easier and more practical method to use clinically than other more sophisticated body composition techniques(Reference Stratton40). The review by Milne et al. (Reference Milne, Avenell and Potter31) showed a significant increase in the mid-arm muscle circumference with ONS use in older hospital patients (weighted mean difference 1·41% (95% CI 0·46, 2·35%), six RCT). Irrespective of the composition of weight change, the increases in weight observed with ONS were linked to functional benefits, particularly in older people and in patients with chronic obstructive pulmonary disease(Reference Stratton, Green and Elia12). In studies in the acute setting, improvements in clinical outcome were often associated with very small changes in body weight (<1–2 kg) as the periods of supplementation were sometimes as brief as 1 week. It is possible that the improvements in outcome observed with ONS, particularly in the acutely ill, occur via mechanisms that are largely independent of changes in body mass. One or more nutrients (macronutrients or micronutrients) provided by liquid, multi-nutrient ONS, could influence immune/inflammatory responses, and hence clinical outcome irrespective of changes in body weight or lean tissue mass(Reference Stratton, Green and Elia12, Reference Stratton40).
Nutritional intake
Systematic reviews that address the impact of ONS on nutritional intake consistently show improvements in total energy intakes in acutely and chronically ill patients in hospital and community settings(Reference Stratton, Green and Elia12, 20, Reference Stratton, Ek and Engfer29, Reference Baldwin, Parsons and Logan39). Some indicate significant improvements in the intakes of protein and micronutrients(Reference Stratton, Green and Elia12, 20, Reference Stratton, Ek and Engfer29). Although often not assessed in trials, it is likely that total nutrient intakes (including micronutrient intakes) are improved with ONS. This is partly because liquid supplements contain a range of macronutrients and micronutrients and also because liquid ONS do not appear to substantially suppress voluntary food intake or appetite(Reference Stratton, Green and Elia12, Reference Stratton and Elia41). This may be particularly important in older patients and those suffering from anorexia, in whom liquid ONS have been shown to improve energy, protein and micronutrient intakes to a greater degree than isoenergetic food snacks in both the hospital and community environment(Reference Stratton, Bowyer and Elia42, Reference Turic, Gordon and Craig43). Similarly, the Cochrane review of Baldwin et al. found significantly greater energy intakes were achieved with ONS than dietary advice (weighted mean difference 91 (95% CI 23, 159) kcal) (0·38 MJ (95% CI 0·10, 0·67 MJ)), four RCT (n 138)(Reference Baldwin, Parsons and Logan39). See Stratton et al.(Reference Stratton, Green and Elia12) and Stratton(Reference Stratton44) for a more detailed review.
Health care use and costs with oral nutritional supplements
There has been an increasing interest in the effects of ONS on health care use and costs. In the acute setting, reductions in the length of hospital stay and complications and a reduction in associated costs with ONS have been well documented(Reference Elia, Stratton and Russell3). The British Association for Parenteral and Enteral Nutrition's health economic report found that the average net cost saving associated with ONS use in specific groups of hospitalised patients was about £850 per patient (2003 prices) and concluded that ‘ONS can produce a net cost saving and be cost-effective in selected patient groups (such as patients undergoing gastrointestinal or orthopaedic surgery)’. The evidence-base for the use of nutritional support (ONS and tube feeding) to prevent pressure ulcers has been systematically examined(Reference Stratton, Ek and Engfer29). This review found that nutritional support had important clinical and statistically significant effects on reducing the development of pressure ulcers in at-risk populations. Based on this systematic review, a simple cost analysis indicated a reduction in costs with the use of ONS to prevent pressure ulcers(Reference Elia and Stratton45). For all stages of pressure ulcers, there was a net cost saving in favour of ONS, which was significant for stage III (effect size 0·12 (95% CI 0·00, 0·11; P=0·04)) and stage IV ulcers (0·12 (95% CI 0·01, 0·11; P=0·04)). This corresponded to a net cost saving of £5 (stage I) to £460 per patient (stage IV) (2004 prices)(Reference Elia and Stratton45).
Hospital admissions/readmissions
Studies are now starting to assess the effects of ONS use, either entirely in the community or partly in the community following initiation in hospitals, on hospital admissions and readmissions. Recently, two large RCT have highlighted significant reductions in hospital readmissions following the use of ONS in community settings, in older people with a wide range of conditions and in individuals with benign gastrointestinal disease(Reference Gariballa, Forster and Walters46, Reference Norman, Kirchner and Freudenreich47). Other trials have also shown lower (not significantly so) hospital admissions/readmissions with the use of ONS v. routine care(Reference Edington, Barnes and Bryan48–Reference Chapman, Visvanathan and Hammond51) (Fig. 3). When combined in a meta-analysis, the results highlight a highly significant reduction in hospital admissions with ONS (OR 0·56 (95% CI 0·41, 0·77), six RCT). The cost savings associated with such reduction in health care use are important, and the associated improvements to the patients’ quality of life are also likely to be hugely beneficial, but require verification.
Reduction in hospital admissions/readmissions with the use of oral nutritional supplements (ONS). *Significant reduction, P<0·05 (individual study references:(Reference Gariballa, Forster and Walters46–Reference Chapman, Visvanathan and Hammond51)).
Encouraging appropriate use and maximising compliance
ONS, like any type of treatment, need to be used appropriately, and as effectively as possible. In addition to using ONS for their prescribed indications (Table 1) and in line with the evidence-base for their use, there are other matters to consider. Nutrition support should be tailored to the individual as much as is feasible and an individual's preference for the type of treatment they require should be taken into account, including the type of supplement they prefer, the time and way in which they receive and consume it (or other interventions that they may prefer). Of course, ensuring compliance to supplements is vital if improvements in nutritional intake and outcomes are required. Maximising the acceptability of supplements to patients may include using a variety of flavours, textures, temperatures, serving formats and consistencies. However, other patients are happy consuming the same supplement (type and flavour) for long periods of time. The energy density, volume and macronutrient profile of the supplement, the time and way in which it is administered and the duration of supplementation are factors that may affect compliance(Reference Stratton, Green and Elia12, Reference Nieuwenhuizen, Weenen and Rigby52). Recently, studies have suggested that increasing the energy density of liquid feeds improves both nutritional intake and compliance with supplementation, over the use of standard energy dense feeds in hospital and community patients(Reference Hubbard, Holdoway and Kerr53–Reference Hubbard, Buchan and Sanders55). Increasing the energy density of ONS minimises the volume needed to be consumed by a malnourished patient, and so encourages compliance and nutritional intake(Reference Hubbard, Holdoway and Kerr53–Reference Hubbard, Buchan and Sanders55). It is also likely that encouragement and education about the reasons for taking supplements would be beneficial. Overall, a greater understanding is still needed to maximise compliance with supplementation, especially over long periods of time, so that benefits can be achieved and sustained.
Like other treatments in clinical practice, it is recommended that the use of ONS (and other forms of nutritional support, such as dietary counselling) is appropriate and monitored. This typically involves reviewing the indications for the use of ONS, setting appropriate goals at the onset and reviewing these goals at intervals (the goals may include improvements in nutritional intake, weight and clinical/functional outcomes). The optimal dose and timing of supplementation for a patient is also important and should be reviewed regularly. It is necessary to review whether the termination of ONS is needed and whether the addition of other forms of nutritional support is needed. A recent publication supported by a number of national organisations (British Association for Parenteral and Enteral Nutrition, Parenteral and Enteral Nutrition Group, British Dietetic Association and National Nurses Nutrition Group) has highlighted some examples of good practice for the appropriate use of ONS in older people, while also summarising the evidence to support their use(Reference Beck, Church and Duncan56). The NICE guidelines also highlight the importance of monitoring the use of all forms of nutritional support(20).
Conclusion
Appropriate, evidence-based use of ONS should be an essential part of the management of disease-related malnutrition, particularly in the current economic environment we are facing. Malnutrition needs to be identified and treated effectively, since failure to do so can lead to disease complications and delay recovery from illness, at enormous cost to the healthcare service. Prompt identification and treatment is required to attenuate the detrimental impact of malnutrition on health and the quality of life. The appropriate use of ONS should be an integral part of the management of disease-related malnutrition. Perhaps surprisingly, the use of ONS (mainly ready-made, multi-nutrient ONS) is the only form of oral nutritional support with a substantial evidence-base supporting its use. The evidence from systematic reviews and meta-analyses supports the use of ONS in a wide variety of conditions, but the evidence is strongest in the acutely ill, in older patients and in malnourished patients, in whom a range of clinical benefits, including a significant reduction in mortality and complications, such as infections and pressure ulcers, have been demonstrated. Appropriate use of ONS is likely to lead to economic benefits associated with reduced health care use, including a reduction in the number of hospital admissions and duration of hospital stay.
Acknowledgements
Both R. S. and M. E. contributed to the writing of the manuscript and both authors read and approved the final manuscript. R. S. is also an employee of Nutricia Ltd. M. E. has no conflict of interest.
