SEARCH

SEARCH BY CITATION

Keywords:

  • hospital patients;
  • malnutrition;
  • nutritional care;
  • protected mealtimes

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References

Background:  Malnutrition is a common problem in hospitalised inpatients, resulting in a range of negative clinical, patient-centred and economic sequelae. Protected mealtimes (PM) aim to enhance the quality of the mealtime experience and maximise nutrient intake in hospitalised patients. The present study aimed to measure mealtime environment, patient experience and nutrient intake before and after the implementation of PM.

Methods:  PM were implemented in a large teaching hospital through a range of different approaches. Direct observations were used to assess ward-level mealtime environment (e.g. dining room use, removal of distractions) (40 versus 34 wards) and individual patient experience (e.g. assistance with eating, visitors present) (253 versus 237 patients), and nutrient intake was assessed with a weighed food intake at lunch (39 versus 60 patients) at baseline and after the implementation of PM, respectively.

Results:  Mealtime experience showed improvements in three objectives: more patients were monitored using food/fluid charts (32% versus 43%, P = 0.02), more were offered the opportunity to wash hands (30% versus 40%, P = 0.03) and more were served meals at uncluttered tables (54% versus 64%, P = 0.04). There was no difference in the number of patients experiencing mealtime interruptions (32% versus 25%, P = 0.14). There was no difference in energy intake (1088 versus 837 kJ, P = 0.25) and a decrease in protein intake (14.0 versus 7.5 g, P = 0.04) after PM.

Conclusions:  Only minor improvements in mealtime experience were made after the implementation of PM and so it is not unexpected that macronutrient intake did not improve. The implementation of PM needs to be evaluated to ensure improvements in mealtime experience are made such that measurable improvements in nutritional and clinical outcomes ensue.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References

The prevalence of malnutrition amongst adults admitted to hospitals in the UK (BAPEN, 2010) and Europe (Schindler et al., 2010) is estimated to be between 10% and 50%. Once in hospital, patients’ nutritional status often continues to deteriorate (Stratton et al., 2003), most commonly in older patients (Larsson et al., 1990). Hospital costs are increased by up to 60% in malnourished patients as a result of longer hospital stays and increased resource utilisation (Braunschweig et al., 2000) and the estimated expenditure for disease-related malnutrition in the UK exceeds £7.3 billion (Elia et al., 2005).

One cause of malnutrition is inadequate nutrient intake, and the reasons for this are varied. Barriers to food access, such as interruptions at mealtimes and unpleasant eating environments (Burke, 1997; Naithani et al., 2009), are especially common in older patients and those with co-morbidities (Naithani et al., 2010). ‘Protected mealtimes’ (PM) is a UK wide initiative that aimed to address some of these barriers.

PM can be defined as ‘periods on a hospital ward when all non-urgent clinical activity stops. During these times, patients are able to eat without being interrupted and staff can offer assistance’ (Hospital Caterers Association., 2004). The implementation of PM is one of the key action points in the Council of Europe resolution: Food and Nutritional Care in Hospitals (Council of Europe Alliance 2003), and is included in the most recent UK Government strategy ‘Improving Nutritional Care’ (Department of Health, 2007). Although PM are seen as an important approach to tackling hospital malnutrition, there is a lack of consistent evidence to demonstrate that they improve mealtime experiences and, most importantly, increase food and nutrient intake. Despite methodological limitations, some previous studies suggest that PM decrease mealtime interruptions, although their impact on food intake is unclear (Das et al., 2006; Weekes, 2008; Chambers et al., 2009; Stuckey et al., 2009).

The present study aimed to compare mealtime environment, patient mealtime experience and nutritional intake before and after the implementation of PM in two large teaching hospitals.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References

The present study measured both mealtime experience and food intake using identical methods at baseline, before PM (June/July 2008) and after the introduction of PM (October/November 2009) at Charing Cross Hospital and Hammersmith Hospital, London, UK. The PM initiative was introduced through a guideline document, dissemination through the Directorate of Nursing, notices on the intranet and provision of large signs indicating that PM were in progress and the times of meals for each ward.

The study was registered as an audit because it compared clinical practice with published standards and did not require ethical approval. However, it was conducted within an ethical framework, ensuring patient confidentiality, and only data relevant to the study objectives were collected.

Mealtime experience

Direct observations were undertaken using a two part questionnaire designed to assess aspects of the PM objectives for: (i) ward-level mealtime environment (e.g. ratio of patients to staff, dining room use, removal of distractions, etc) and (ii) individual patient mealtime experience (e.g. receiving assistance, visitors present, etc). The aspects of environment and experience that were assessed are detailed in Table 1. All adult wards were eligible, except for private wards and the intensive care units. A mealtime environment was observed on all eligible wards and the patient mealtime experience was measured on any eligible patient on these wards, except for those receiving enteral or parenteral nutrition. Ward managers were aware the evaluation was ongoing, although they were unaware of the day that their ward would be observed to avoid influencing ward activity. Each ward was randomly assigned to undergo either a lunch or supper observation using random number tables. Two researchers (one dietitian/student dietitian and one dietetic assistant/nurse/healthcare assistant) observed each mealtime. Both performed the ward-level direct observations, and then each researcher selected one bay to observe approximately five patients (a total of approximately ten patients per ward), aiming to maximise patient observations.

Table 1.   Direct observations of environment and patient experience during mealtimes at baseline and after the introduction of protected mealtimes (PM)
ObjectiveBaselineProtected mealtimesP
  1. *40 wards at baseline and 34 after PM underwent direct observation of ward-level mealtime experience. Differences in the denominator reflect either missing data or that the question was not applicable.

  2. 253 patients at baseline and 237 after PM underwent direct observation of patient mealtime experience. Differences in the denominator reflect either missing data or that the question was not applicable.

Mealtime environment (ward level)*
 Mobile patients offered opportunity to eat in a day room, n/N (%) 23/40(58) 22/34(65)0.69
 Day room in use, n/N (%) 14/23(61) 10/22(45)0.46
 Patients toileted before mealtimes, n/N (%) 13/37(35) 11/25(44)0.66
 Food service was announced, n/N (%) 12/38(32) 11/34(32)1.00
 Possible to close bay doors, n/N (%) 27/36(75) 18/33(55)0.13
 Bay doors closed and a nurse stayed to assist patients, n/N (%)  2/18(11)  2/16(13)1.00
 Distractions (e.g. television) were turned off, n/N (%) 15/27(56) 10/19(53)1.00
 Number of patients on ward, mean (SD) 18.0(8.1) 20.0(5.7)0.19
 Number of nursing/assistant staff on ward, mean (SD)  5.5(1.8)  5.8(2.0)0.40
 Number of staff assisting with meal service, mean (SD)  1.9(1.4)  2.4(1.4)0.14
 Number of staff assisting patients with meals, mean (SD)  1.7(1.7)  2.1(1.5)0.40
 % of staff assisting with meal service, mean (SD) 39.1(31.7) 41.4(24.5)0.74
 % of staff assisting patients with meals, mean (SD) 34.0(33.2) 36.0(26.6)0.76
Patient mealtime experience (individual patients)
 Patients on food or fluid charts, n/N (%) 78/243(32)100/234(43)0.02
 Patients offered opportunity to wash hands, n/N (%) 73/243(30)91/227(40)0.03
 Table clean and clutter free, n/N (%)134/247(54)149/233(64)0.04
 Patient able to sit out in chair, n/N (%)189/250(76)155/229(68)0.07
 If able to sit in chair, patients assisted to do so, n/N (%) 74/135(55) 50/150(33)0.10
 If unable to sit in chair, patients repositioned in bed, n/N (%) 49/60(82) 77/85(91)0.19
 Patients receiving the correct food, n/N (%)193/208(93)193/201(96)0.23
 Patients requiring assistance with eating, n/N (%) 48/240(20) 48/222(22)0.75
 Patients receiving assistance where required, n/N (%) 31/36(86) 38/43(88)0.94
 Patients undisturbed during mealtimes, n/N (%)166/243(68)175/234(75)0.14
 Patients with visitors at mealtime, n/N (%) 60/236(25) 44/226(20)0.36
 Patients with visitors who provided feeding assistance, n/N (%) 25/60(42) 20/44(46)0.85
 Food or fluid charts completed before plates removed, n/N (%) 29/67(43) 50/88(57)0.12

Nutrient intake

A weighed food intake at a lunch was undertaken on a subsample of inpatients. Inclusion criteria were patients who were at ‘high risk’ of malnutrition because they have the most benefit to derive from PM, and were screened using the INSYST nutritional screening tool (Tammam et al., 2009). Exclusion criteria were patients who were nil-by-mouth or receiving enteral or parenteral nutrition. The patients recruited after the introduction of PM were selected so as to broadly match the characteristics of the baseline patient population, aiming to minimise confounding variables, including gender, age group, diet type (e.g. modified consistency), diagnosis and level of feeding assistance (independent, minimal, moderate, total assistance) (Simmons et al., 2008).

The hospitals use a ready plated and sealed catering system (Hickson et al., 2007). To measure the weight of food served, three samples of every meal (e.g. chicken meal) were cooked and the average weight of each component (e.g. roast chicken, roasted vegetables and dauphinoise potatoes) was measured. To measure the weight of food consumed, following a patient’s lunch, the waste of each component was weighed separately and the amount consumed was calculated. Energy and protein intake and the percentage consumed of energy and protein served were calculated, with standard recipes provided by the food manufacturer, using a computerised food composition programme (dietplan6; Forestfield Software, Horsham, UK).

Statistical analysis

The sample size calculation was based upon detecting a 420 kJ (100 kcal) difference in energy intakes between the baseline and PM groups assuming a power of either 90% or 80%, α = 0.05 and SD of 630 kJ (150 kcal) (Hickson et al., 2007). This would require a sample size of 50 (90% power) or 37 (80% power) patients in each group. Because only 39 patients were recruited at baseline, a total of 60 patients was the target for the PM group to increase the power.

Categorical data are presented as frequencies (n/N) and percentages (%) and are compared between baseline and PM using the chi-squared test. Continuous data are presented as the mean (SD) or medians and the interquartile range (IQR), depending on distribution, and compared using either a t-test or the Mann–whitney U-test using spss, version 15 (SPSS Inc., Chicago, IL, USA). P < 0.05 was considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References

For the evaluation of mealtime environment and experience, 40 wards and 253 patients were observed at baseline, and 34 wards and 237 patients were observed after the PM. Table 1 indicates that there were few differences in the mealtime environment and patient mealtime experience after the introduction of PM. There was no significant difference in the proportion of patients experiencing mealtime interruptions (P = 0.14). The only aspects that changed significantly were that more patients were monitored using food and/or fluid charts (32% versus 43%, P = 0.02), more were offered the opportunity to wash hands (30% versus 40%, P = 0.03) and more were served meals at clean and uncluttered tables (54% versus 64%, P = 0.04). No ward had all the PM objectives met for all patients.

The weighed food intake indicated no impact of PM on energy intake [1088 kJ (260 kcal) versus 837 kJ (200 kcal), P = 0.25] and a decrease in protein intake (14.0 versus 7.5 g, P = 0.04) (Table 2).

Table 2.   Energy and protein served and consumed at the lunchtime meal, as measured by weighed food intake
 Baseline (n = 39)Protected mealtimes (n = 60)P
  1. IQR, interquartile range.

Energy (kJ)
 Served (kJ) mean (SD)2268 (720)2251 (1013)0.94
 Consumed (kJ) median (IQR)1088 (301–2144)837 (837–3502)0.25
 Consumed (% of served) mean (SD)52 (21–88)43.5 (12–80)0.31
Energy (kcal)
 Served (kcal) mean (SD)542 (172)538 (242)0.94
 Consumed (kcal) median (IQR)260 (72–513)200 (36–434)0.25
Protein
 Served (g), mean (SD)25.3 (8.5)23.5 (11.7)0.40
 Consumed (g), median (IQR)14.0 (4.0–26.0)7.5 (1.0–18.0)0.04
 Consumed (% of served), mean (SD)63 (25–90)41 (6–76)0.06

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References

Only small improvements have been made in the patient mealtime experience after the introduction of PM in these hospitals. Importantly, the implementation of PM did not result in a reduction in mealtime interruptions, with a large minority (25%) still experiencing disruptions. Although some minor improvements in patient mealtime experience occurred, in general, there was little improvement, and this is potentially attributable to inadequate implementation. Factors critical for successful implementation of PM include: incorporating PM into hospital policy, promotion, communication and leadership at all levels, and structured education for all staff groups (NPSA, 2006). At these hospitals, PM was only a guideline and there was no structured education to support PM, simply intranet notifications and new ward signage. These data indicate that the comprehensive planning and support of PM implementation is vital.

Because there were no significant improvements in ward mealtime environment, and few improvements in patient mealtime experience, it is unsurprising the nutrient intake did not improve. The reason for a lower protein intake is unclear. The present study provides evidence that PM implementation was not wholly successful. Although it may be conventional wisdom to assume that improved nutritional care in hospitals will improve nutrient intake (Wilson, 2009), there is a surprising lack of quality evidence to support this (Weekes et al., 2009). Driving improvements in nutritional care is an important goal for the UK Government (Brotherton, 2010); however, further evidence is required to demonstrate that PM comprises a route for achieving improved nutritional care and nutrient intake.

Future research is required to evaluate the implementation of PM. It is vital to ensure that fully operational PM, with resulting improvements in ward mealtime environment and patient mealtime experience; otherwise, intensive weighed food intake studies are unlikely to demonstrate improvements in intake. Only then can we evaluate whether an improved eating environment in line with PM objectives results in measurable improvements in nutritional and clinical outcomes. If the implementation of PM is shown not to improve such outcomes, we urgently need to explore other possible strategies to prevent hospital related malnutrition. Similarly, robust evidence to support the use of PM will encourage more hospitals to work harder to implement PM successfully.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References

The authors thank Stella Kourtellidou and all of the dietitians at Charing Cross Hospital and Hammersmith Hospital for assisting in the data collection and, in particular, Gopika Chandratheva for assistance with the weighed food intake. The authors thank the nursing staff on each ward who assisted with the direct observations.

Conflict of interests, sources of funding and authorship

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References

None of the authors have any conflicts of interest to declare.

No funding was obtained to carry out this work.

MH, AC and KW designed the study. AC undertook the data collection. MH, AC and KW were all involved in the analysis and interpretation of the data. MH wrote the initial draft of this report and all authors agreed and contributed to the final version submitted for publication.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of interests, sources of funding and authorship
  9. References
  • BAPEN (2010) Nutrition Screening Survey in the UK in 2008: Hospitals, Care Homes and Mental Health Units. Redditch, UK: British Association of Parenteral and Enteral Nutrition.
  • Braunschweig, C., Gomez, S. & Sheean, P.M. (2000) Impact of declines in nutritional status on outcomes in adult patients hospitalized for more than 7 days. J. Am. Diet. Assoc. 100, 13161322.
  • Brotherton, A. (2010) Driving improvements in nutritional care. J. Hum. Nutr. Diet. 23, 557558.
  • Burke, A. (1997) Hungry in Hospital? London: Association of Community Health Councils for England and Wales.
  • Chambers, C., Maxwell, R., Cahalane, S., Cannon, E., Kenny, E. & Hickey, R. (2009) Do protected mealtimes result in fewer interruptions to the patients’ meals? 26th International Conference of the International Society for Quality in Healthcare.
  • Council of Europe Alliance (2003) Council of Europe resolution food and nutritional care in hospital: 10 key characteristics of nutritional care in hospitals. [online]. Available at http://www.bapen.org.uk/pdfs/coe_leaflet.pdf (accessed 12 January 2011).
  • Das, A.K., McDougall, T., Smithson, J.A. & West, R.M. (2006) Benefits of family mealtimes for nursing home residents: protecting mealtimes may similarly benefit elderly inpatients. BMJ 332, 13341335.
  • Department of Health (2007) Improving Nutrition Care: A Joint Action Plan from the Department of Health and the Nutrition Summit Stakeholders. London: Department of Health.
  • Elia, M., Stratton, R.J., Russell, C., Green, C.J. & Pan, F. (2005) The cost of disease-related malnutrition in the UK and economic considerations for the use of oral nutritional supplements (ONS) in adults: BAPEN.
  • Hickson, M., Fearnley, L., Thomas, J. & Evans, S. (2007) Does a new steam meal catering system meet patient requirements in hospital? J. Hum. Nutr. Diet. 20, 476485.
  • Hospital Caterers Association (2004) Protected mealtimes policy. [online]. Available at http://www.hospitalcaterers.org/documents/pmd.pdf (accessed 12 January 2011).
  • Larsson, J., Unosson, M., Ek, A.C., Nilsson, L., Thorslund, S. & Jurulf, P. (1990) Effect of dietary supplement on nutritional status and clinical outcomes in 501 geriatric patients – a randomised study. Clin. Nutr. 9, 179184.
  • Naithani, S., Thomas, J.E., Whelan, K., Morgan, M. & Gulliford, M.C. (2009) Experiences of food access in hospital. A new questionnaire measure. Clin. Nutr. 28, 625630.
  • Naithani, S., Whelan, K., Thomas, J.E. & Gulliford, M.C. (2010) Multiple morbidity is associated with increased problems of food access in hospital: a cross-sectional survey utilising the cumulative illness rating scale. J. Hum. Nutr. Diet. 23, 575582.
  • National Patient Safety Agency (NPSA) (2006) Protected Mealtimes Review: Findings and Recommendations Report [online]. Available at http://www.nrls.npsa.nhs.uk/resources/?entryid45=59806 (accessed 10 May 2011).
  • Schindler, K., Pernicka, E., Laviano, A., Howard, P., Schütz, T., Bauer, P., Grecu, I., Jonkers, C., Kondrup, J., Ljungqvist, O., Mouhieddine, M., Pichard, C., Singer, P., Schneider, S., Schuh, C., Hiesmayr, M. & NutritionDay Audit Team (2010) How nutritional risk is assessed and managed in European hospitals: a survey of 21,007 patients findings from the 2007–2008 cross-sectional Nutrition Day survey. Clin. Nutr. 29, 552559.
  • Simmons, S.F., Keeler, E., Zhuo, X., Hickey, K.A., Sato, H.W. & Schnelle, J.F. (2008) Prevention of unintentional weight loss in nursing home residents: a controlled trial of feeding assistance. JAGS 56, 14661473.
  • Stratton, R.J., Green, C.J. & Elia, M. (2003) Disease-Related Malnutrition: An Evidence-Based Approach to Treatment. Wallingford: CABI Publishing.
  • Stuckey, C.M., O’Malley, G., Matthias, S., Hawkins, A., Geldart, S., Waever, S.A. & Smith, T.R. (2009) The introduction of ‘protected mealtimes’ increases nutritional intake in acute medical patients. Gut 58, A79A80.
  • Tammam, J.D., Gardner, L. & Hickson, M. (2009) Validity, reliability and acceptability of the imperial nutritional screening system (INSYST): a tool that does not require the body mass index. J. Hum. Nutr. Diet. 22, 536544.
  • Weekes, C.E. (2008) The effect of protected mealtimes on meal interruptions, feeding assistance, energy and protein intake and plate waste. Proc. Nutr. Soc. 67(OCE), E119.
  • Weekes, C.E., Spiro, A., Baldwin, C., Whelan, K., Thomas, J.E., Parkin, D. & Emery, P.W. (2009) A review of the evidence for the impact of improving nutritional care on nutritional and clinical outcomes and cost. J. Hum. Nutr. Diet. 22, 324335.
  • Wilson, R.C. (2009) The importance of nutritional care. J. Hum. Nutr. Diet. 22, 185186.