A report on the cost of disease-related malnutrition in England and a budget impact analysis of implementing the NICE clinical guidelines/quality standard on nutritional support in adults
The report was launched Monday 30th November 2015 at the King’s Fund, 11 Cavendish Square, London at a meeting chaired by Professor Alan Jackson, Professor of Human Nutrition, University of Southampton
A – The cost of malnutrition in England in 2011–12
Malnutrition, with and without associated disease, is a common clinical, public health and economic problem, with an estimated cost of £19.6 billion in England in 2011–12.
The public health and social care expenditure associated with malnutrition in adults and children in England in 2011–12, identified using the ‘Malnutrition Universal Screening Tool’ (‘MUST’), was estimated to be £19.6 billion, or about 15% of the total expenditure on health and social care. Most of this expenditure was due to healthcare rather than social care, and secondary rather than primary healthcare provision involving adults, predominantly older adults, rather than children. This pattern reflects the general distribution of the total expenditure on health and social care of all subjects in England.
The large contribution of institutionalised care to total costs was not only due to the high cost of institutionalisation, but also the high point prevalence of malnutrition in hospitals and care homes. However, since more than 90% of the malnutrition originates and exists outside these institutions, preventive measures should be undertaken in the community to reduce the clinical economic burden of malnutrition.
Given the large estimated annual cost of malnutrition (£19.6 billion), small fractional cost savings translate to large absolute savings (e.g. 1% cost saving corresponds to £196 million). Effective recognition and treatment of malnutrition and continuity of care within and between care settings are of key importance to achieving such goals.
B – Budget (cost) impact analysis involving implementation of the NICE clinical guidelines (CG32)/quality standard (QS24)
Interventions to combat malnutrition in the small proportion of malnourished patients targeted by the NICE clinical guidelines/quality standard on nutritional support in adults save rather than cost money. The estimated net cost saving of £172.2–229.2 million is due to reduced healthcare use.
Improvements in current nutritional care associated with fuller implementation of the NICE guideline/quality standard (identification of malnutrition and use of nutritional support in adults) not only result in better quality of care but also in a net cost saving. The investment necessary to implement better nutritional care is more than counteracted by the returns (the cost savings).
When the clinical guidelines/standard was applied to 85% of subjects with high risk of malnutrition in the population of malnourished adults targeted by the NICE guidelines/quality standard there was an overall net cost saving of £63.2–76.9 million (£119.20–145.09 thousand per 100,000 of the general population) depending on the type of nutritional support and the care setting(s). When they were applied to 85% of adults with medium and high risk of malnutrition according to ‘MUST’ the net cost saving was estimated to be £172.2–229.2 million in England or £324.8–432.3 thousand per 100,000 of the general population. These estimates exceeded those reported by NICE (£71,800 per 100,000 general population), which ranked the cost saving as the third highest relative to those associated with the implementation of other NICE clinical guidelines.
The above net cost savings, mostly due to appropriate use of oral nutritional supplements, represent only 0.4–3.3% of the total annual healthcare cost of disease-related malnutrition in adults, which amounted to £14.4 billion. However, the costing models involved only a proportion of patients with malnutrition presenting to healthcare workers (that targeted by the NICE guideline/quality standard), and only a fraction of this proportion received improved nutritional care. In addition, the large total cost of disease-related malnutrition included the cost of disease, much of which is not reversed by nutritional support alone.
The net cost saving was found to increase when the prevalence of malnutrition was high, when hospital admission rates were high, and when the gap between current care and desirable care was large. Rapid and reliable methods for nutritional screening were also found to produce a more favourable budget impact.
To improve the robustness of the costing model, future research should aim to establish an evidence base on healthcare use and cost of other forms of nutritional support for which little data exist (e.g. dietary advice, dietary modification and food fortification), and to further extend the evidence base on the effects of prescribable oral nutritional supplements on resource use in different care settings.
A – The cost of malnutrition in England in 2011–12
Cost of health and social care
1. The public expenditure on healthcare in England in 2011–12 was £101.6 billion, of which £90.6 billion was spent on behalf of the resident population. Most of the purchased healthcare was secondary healthcare (£68.8 billion) and the remainder primary healthcare (£21.6 billion). It was estimated that only about 11% of the expenditure involved children less than 18 years.
2. The public expenditure on social care in England was £26.1 billion (£26.5 billion with inclusion of the costs for service strategy, asylum seekers, and other adult services) with approximately equal distribution between children (36%), younger adults (<65 years; 30%) and older adults (≥ 65 years; 34%).
3. The total public expenditure on health and social care in England was estimated to be £127.5 billion, with children accounting for approximately a sixth and the remainder, approximately equally divided between younger and older adults.
Prevalence of malnutrition
4. In healthcare, the prevalence of malnutrition varied with age and care setting. On admission to hospital it was estimated to be highest in those aged >65 years (33.6%), intermediate in adults <65 years (25.1%) (both measured using the ‘MUST’ for adults) and lowest in children (15%). At a given point in time the prevalence of malnutrition in hospitalised patients was considered to be higher than the admission prevalence, mainly because those with malnutrition have a longer length of hospital stay (30%) than those without.
5. In social care the prevalence of malnutrition also varied with age and care setting. In care home residents it was estimated to be 36% in older adults and only 24% in younger adults, who accounted for a minority of the care home population. The prevalence among older adults receiving day care and domiciliary (home) care was assumed to be 18%, and in younger adults 16%. Limited information in children receiving social care suggests that malnutrition affects only a small proportion (estimated to be 3%). For looked-after children, overweight and obesity are common and underweight distinctly uncommon.
6. At a given point in time in 2011–12 it was estimated that the number of people with malnutrition in hospital was only about 0.044 million (0.041 million publicly funded and 0.003 million privately funded), in care homes 0.142 million (0.078 million local authority funded and 0.064 million privately funded), and 0.066 million in sheltered housing. Of the total population of England in 2011–12 (53 million), 2.65million (5%) were estimated to be malnourished or at risk of malnutrition at a given point in time (2.12–3.18 million if 4–6% of the general population is malnourished or at risk of malnutrition (estimates based on an amalgamation of surveys in hospitals, care homes, sheltered housing, and national surveys of the general population). This means that at a given point in time only about 2% of the malnutrition was accounted for by hospital inpatients.
The cost of malnutrition in health and social care
7. The public expenditure on malnutrition in healthcare was estimated to be £15.2 billion, the majority of which was due to secondary care rather than primary care. Malnourished hospital inpatients were estimated to cost £7.7 billion and malnourished hospital outpatients £0.9 billion.
8. The public expenditure on malnutrition in social care was estimated to be £4.4 billion, more than 90% of which involved adults, predominantly older adults.
9. The total public expenditure on malnutrition in health and social care was estimated to be £19.6 billion, with older adults accounting for 52% of the total, younger adults for 42%, and children for 6%. Institutionalisation of malnourished people (hospital inpatients and care home residents) accounted for up to £9.3 billion.
10. A series of one-way sensitivity analyses examining healthcare costs involved changing assumptions about the prevalence and distribution of costs by age groups in hospital inpatients and outpatients, as well as other secondary care costs and primary care costs. The sensitivity analyses examining social care costs involved changing the assumptions about the prevalence of malnutrition in residential care, the costs for assessment and management of malnourished patients, and the costs of providing domiciliary and home care to those with malnutrition or at risk of malnutrition. These sensitivity analyses affected the overall expenditure on health and social care by less than 5% and generally by less than about 2%.
11. The estimated public expenditure on healthcare was £1917 per capita of population and on social care £500 (total £2417). In those with malnutrition or risk of malnutrition it was estimated to be £5763 per malnourished subject for healthcare (based on the point prevalence of malnutrition and annual expenditure on malnutrition) and £1645 for social care. The corresponding figures for non-malnourished subjects were £1715 and £440, respectively). This means that the expenditure in a hypothetical subject suffering from malnutrition during the entire year is 3.36 times greater compared to one without malnutrition during the same period. The incremental cost of malnutrition was £5239 per subject (cost of a subject with malnutrition minus the cost of a subject without malnutrition). Expressed per capita population the estimated annual cost of malnutrition is £370 and the incremental cost £263.
12. The cost of disease-related malnutrition (and malnutrition without disease) (£19.6 billion) is estimated to account for about 15% of the health and social care budget. Small percentage cost savings resulting from interventions translate to large absolute annual cost savings (e.g. a 1% reduction in the expenditure on malnutrition is £196 million per year).
13. Malnutrition exists in all care settings and all age groups. Strategies to combat its clinical and economic consequences should be joined up across care settings and age groups.
B – Budget (cost) impact analysis involving implementation of the NICE clinical guidelines/quality standard
14. A cost analysis was undertaken to examine the resource impact of implementing the NICE clinical guideline (CG32)/quality standard (QS24) on nutritional support in adults, involving only a small proportion of the total malnourished population found in England at a given point in time. The model comprised oral nutritional support with oral nutritional supplements (ONS) as well as non-ONS support, enteral tube feeding (ETF) and parenteral nutrition (PN) in hospital and community settings in England in 2011–12, and involved major modifications of a NICE costing template. It used data from the Health and Social Care Information Centre, national surveys on the prevalence of malnutrition in various care settings, and national surveys on the prevalence of home enteral and parenteral nutrition. Systematic reviews and meta-analyses involving interventions with ONS in hospital and community settings were also used. The analyses were undertaken in the light of expert opinion about clinical data, especially those relating to current practice.
15. The costing model involved three steps: calculation of the extra cost (investment) needed to change the current pathway of nutritional care to a proposed pathway incorporating the NICE guideline/quality standard; the cost saving arising from reduced healthcare use associated with the proposed pathway; and the overall net balance (budget impact) calculated as the difference between the first two steps.
16. In the base case analysis, which assumed that 90% of malnourished subjects were screened and about 85% of those at high risk of malnutrition were provided with nutritional support, either directly by a dietitian or indirectly without a dietitian, according to local policy.
17. Five models, complementary to each other, were used to evaluate the budget impact: ONS in hospital (inpatients and outpatients) and community (new general practice registrations and care home admissions); oral (ONS and non-ONS) nutritional support in hospital inpatients and outpatients; oral nutritional support in hospital and community (new general practice registrations and care home admissions); oral nutritional support in hospital and community settings (as above) plus enteral tube feeding in hospital; and oral nutritional support and enteral and parenteral nutritional support in hospital and community settings.
18. Using all five models, the base case analyses indicated a cost saving (£101.8 – £126.6 million depending on the model) that exceeded the extra cost of implementing the proposed pathway of high quality care (£19.2 – £61.2 million). The result was an overall net cost saving of £63.2 – £76.9 million (£119.20 – £145.09 thousand per 100,000 population).
19. The largest single largest extra cost due to the implementation of the proposed pathway of care was nutritional screening (all models). In the models involving all care settings the extra costs amounted to £19.7 million or about one-third of the total extra costs, even when the costs of providing extra nutritional support with ONS, ETF and PN were taken into account.
20. Depending on the model used, older adults were estimated to account for 47–51% of the costs, 50–64% of the cost saving, and 52–76% of the net cost saving.
21. When the models were modified to ensure that the proposed pathways involved an intervention in 85% of those at medium + high risk of malnutrition according to ‘MUST’, the overall net cost saving increased to £172.2–229.2 million in England (£324.8–432.3 thousand per 100,000 people in the general population).
22. A series of one-way sensitivity analyses were undertaken which involved varying the assumptions about variables affecting the costs. These included rates of hospital admissions, prevalence of malnutrition, time taken to screen (and salary scale of person undertaking the screening), and cost savings (including reduction in length of hospital stay, hospital admissions and GP and outpatient visits.
23. The models were sensitive to variations in admission rates (±20% variation in admission rate affected the net monetary balance by ±20–25%), prevalence of malnutrition (± 3% variation affected the net balance by ±11.1–13.8%) and time taken to screen a subject (variation of ±4 minutes (base case value 5 minutes) affected the net balance by 11.1–13.8%). The models were much less sensitive to variations in the assumptions about the prevalence of malnutrition in subjects newly registering with their GP and those in care homes, and the pay scales of those undertaking nutritional assessment.
24. The single most important variable affecting the net balance was the cost saving due to the effect of ONS in reducing length of hospital stay (13.9 ± 6.6% according to a random effects meta-analysis of 12 studies involving patients with malnutrition. Variation in the cost saving by ±25% of the actual value (i.e. 13.9 ± 3.475%) affected the net balance by up to ±34%. There was uncertainty about the effectiveness of oral non-ONS nutritional support (e.g. dietary modification and/or dietary advice provided by dietitians), which could potentially have a substantial effect on the net cost saving. Other sensitivity analyses involving greatly restricted populations of malnourished subjects in the community setting (only new registrations at GP surgeries and new admissions to care homes) had little impact on the final net cost saving.
25. The models could be made more robust by evidence-based information on the effects of nutritional support in routine clinical practice. One of the important areas that needs to be addressed is the effect of oral non-ONS nutritional support on clinical and economic outcomes.
26. The estimated net cost saving was found to be greater than that reported by NICE in 2012, in which nutritional support in adults was ranked third after hypertension (clinical guideline (CG) 34) and long-term contraception (CG 30) among other cost saving schemes involving implementation of NICE clinical guidelines. While there are many differences between the current economic model and that used by NICE, a key difference is that in the current model more screening and more nutritional support were undertaken.
27. All the costing models suggest that nutritional support in adults produces a net cost saving, with important clinical implications. Local economies are likely to experience larger net cost savings from implementation of the NICE clinical guidelines/standard when the prevalence of malnutrition is high, when the rate of hospital admissions is high, and when there is a large gap between current nutritional care and desirable, high quality nutritional care.
28. Since interventions in one care setting can influence the clinical and economic burden in another, separate funding streams may create problems if one setting bears the costs the other the economic benefits. An integrated system of care with a single funding stream that follows the patient may avoid such problems.