Introduction

A recent multicentre study found that up to 40% of patients admitted to the UK spinal cord injuries (SCI) centres were malnourished, and more than half of patients reported weight loss after SCI.1 Despite its importance, malnutrition is rarely identified by clinicians working in general hospitals2 or in SCI centres (SCIC).3 We reported previously that the provision of dietetic resources in SCIC is suboptimal,4 and therefore, it is imperative that hospital catering staff in conjunction with ward-based personnel are proficient in optimising menu design, and in adapting menus, that they have the ability to improve the mealtime atmosphere, and are able to deal with individual patient needs as well as offering a balanced menu that meets patients’ nutritional requirements.

Previous studies in general hospital patients have demonstrated that the dietary intake of hospitalised patients is often inadequate.5 Poor nutritional intake is associated with worse clinical outcomes,6 and appropriate nutritional care can improve these outcomes.7, 8, 9 The independent health watchdog in England, the Care Quality Commission, assesses all hospitals annually against Core Standard 5, meeting nutritional needs.10 The most recent Care Quality Commission national survey found that 20% of hospitals were not compliant in meeting this standard.10 Additionally, in a separate survey in 165 English hospital trusts, it was found that 20% of patients felt that they did not get the help they needed to eat their meals.11

There are currently limited data reporting how well the current catering system provides for the needs of patients with SCI. The aims of the present study were to (1) assess patients’ food intake and (2) measure satisfaction with current food provision as judged by patients and by the professional stakeholders: medical staff, nursing staff, managers and catering staff.

Methods

The study consists of an audit of patient food intake followed by an evaluation of the service by patients and professional stakeholders. Three questionnaires were employed (Figures 1, 2 and 3).

Figure 1
figure 1

Intake questionnaire.

Full size image
Figure 2
figure 2

Meal provision questionnaire—patients

Full size image
Figure 3
figure 3

Meal provision questionnaire—staff.

Full size image

Questionnaires

The first questionnaire addressed individual patients’ food intake; we invited patients to self-report (with help from nursing staff if the ability to write was impaired due to SCI) their actual food intake semi-quantitatively on a single day, by recording whether: all; greater than half; half; quarter; or none of the food served on the plate was consumed. The availability and consumption of snacks were also recorded. In addition, we recorded whether patients received oral nutritional supplements, vitamin and mineral supplements, and/or needed assistance to eat.

The second part of the audit addressed service users’ (Figure 2) and stakeholders’ (Figure 3) evaluation of the current food provision in the SCIC. We invited participants to rate, with a 5-point Likert scale (strongly agree to strongly disagree) their opinions on (1) food choice; (2) food ordering; (3) food delivery; (4) quality of food; and (5) overall satisfaction with meals. Staff members were asked to report on their perception of the service as provided to patients.

The third component was an open question that aimed to seek the subject’s views on how to improve food provision in the SCIC.

A pilot study (n=5) was performed to assess the content and time required to complete the questionnaire; feedback from this guided the development of the final version of the questionnaire.

Ethics

This study was conducted according to the guidelines outlined in the Declaration of Helsinki. Formal ethical permission to conduct the study was not required by the hospital review board, as this was considered to be a clinical audit not involving active patient participation.12 The questionnaires were approved by the local clinical audit department for phrasing and grammar of the questions.

Survey administration

The questionnaires were given to all in-patients (n=115) and to all the stakeholders working in the National Spinal Injuries Centre at Stoke Mandeville Hospital (n=167) during the period January 2011 to March 2011. Staff included doctors (n=15); nurses (n=102); allied health professionals (n=40); catering staff (n=7); and managers (n=3). One reminder was sent to staff by email 2 weeks after the initial distribution. Patients were reminded by a personal visit from the investigator.

The questionnaires were completed anonymously, and participants were asked to complete them without conferring with colleagues or fellow patients.

Statistical analysis

Descriptive statistics were used to calculate the response frequency. Data are reported as medians (range). For numerical data on an ordinal level, the Mann–Whitney test was used, and for cross-tabulation on a nominal level, the χ2-test was performed. Comparisons were made between those who agreed (strongly agree and agree) and disagreed (strongly disagree and disagree). The data were analysed using Minitab 15 (Minitab Ltd, Coventry, UK) and significance was accepted if P<0.05.

The questionnaire items were analysed by principal components analyses (PCA) with direct oblimin rotation. A scree plot was used to determine the number of components to be extracted. Items with loadings above 0.5 in each component of the matrix were retained for scale construction, following appropriate reversal of items based on polarity of loadings. If scales consisted of a large number of items, these were explored with a further PCA to determine if they could be further broken down. This was analysed by using SPSS 17 (SPSS Inc., Brimingham, UK).

Results

A total of 105 questionnaires were distributed for the first part of the study and responses were received from 67 patients (63.8%; age range: 15–81 years (median: 49 year), 14.9% female, 55.7% tetraplegic; 30.7% complete SCI). Ten (14.9%) patients required assistance from hospital staff to complete their questionnaire. Of 61 patients with a nutrition-risk score, 27 (44.2%) were at malnutrition risk.

The second questionnaires were sent to 272 potential participants and responses were received from 56 patients (53.3%) and 117 stakeholders (70.5%) with an overall response rate of 61.3%.

Food intake

A total of 52% cent of SCI patients did not eat all the food served (Table 1). Approximately 1 in 10 patients ate less than half of the hospital meals. A total of 25% of patients were found to miss one or more meals. A significantly higher proportion of patients missed breakfast compared either with lunch or supper (12, 3 and 3%, respectively, χ2: 6.073, P=0.048).

Table 1 Distribution of food intake
Full size table

Snacks were not provided to the great majority of patients (90%), whereas oral nutrition supplement appeared to be used by a quarter (26%), and one third of the patients reported taking vitamin and mineral supplements.

Overall, 23 (35%) patients reported they required assistance to eat.

Meal provision

Table 2 and Figure 4 summarise patient and staff evaluation of SCIC meals.

Table 2 Evaluation summary from patients and staffs
Full size table
Figure 4
figure 4

Patients’ and Staffs’ evaluation of a SCIC meals.

Full size image

Overall, fewer patients (23.3%) than staff (50%) reported that the meal delivered failed to correspond to that ordered (P<0.01).

Fewer patients (55%) than staff (71.5%) felt that meals were served at an appropriate temperature (P<0.01).

Inadequate taste and lack of choice were the most commonly cited adverse comments (patients: 46.7%; staff: 49.3%), along with meals not served on time (patients:15%; staff: 24%) and interruptions during mealtimes (patients: 31%; staff: 31%).

The first PCA identified two components which accounted for 55.2% of the variance. The first factor (Q1, 6, 7, 8, 9, 10, 11, 12) accounted for 44.1% of the variance and the second (Q2, 3, 4, 5) accounted for 11.2%. As factor 1 was relatively large with eight items, these were subject to a further PCA. The resultant components of the second PCA accounted for 66.9% of the variance in these items. The first factor accounted for 56.7% of the variance, and the second a further 10.2%.

Therefore, principal component analyses of item scores identified three main factors: (i) food quality; (ii) food presentation; and (iii) food delivery (logistics). (Figure 5) represented the main sets of beliefs contained within the questionnaire. Items from each factor (Table 2) was used to make sub-scale sores, based on average item score. Differences between patients and staff beliefs on this sub-scale are shown in Figure 5.

Figure 5
figure 5

Mean scale scores plot.

Full size image

Discussion

To our knowledge, this is the first study that reported the nutrient intake and evaluated the current meal provision among SCI patients and stakeholders. Their inevitable disabilities and the need for extended hospitalisation for acute care and rehabilitation in this group of patients renders extrapolation from unselected hospital populations unwise. We believe, however, that our sample size and response rate (which are comparable to other published studies) lend sufficient weight for our conclusions to be robust.13, 14

Food is the treatment for most malnourished patients in hospital. Although oral nutrition supplement are used to supplement food intake in a proportion of patients, artificial nutrition support, such as enteral and parenteral nutrition, is only used in a minority of SCI patients.1

It is important to assess patients’ food intake and to review their experience so that improvements can be made where necessary to optimise food intake during the hospital stay. SCI patients are vulnerable to malnutrition,1 and the present audit provides evidence that improvements are indeed still needed. Recent national11, 15 and international16 reports have highlighted the importance of nutritional screening, which is a multidisciplinary responsibility, as an important step in fighting hospital malnutrition. Through an effective nutrition screening followed by a successful implementation of appropriate nutrition care plan, which will include monitoring patients’ intake if necessary, will help address the nutritional needs of this group.

We found that 52% of hospitalised patients did not eat all the food served; of these patients, 1 in 3 (34%) were found to miss one or more meals. Based on the UK’s estimated average requirements and reference nutrient intake,17 the deficit from a single missed meal would be equivalent to 300–600 kcal and 1520, g protein per day (16–33% of daily requirement).

The study also demonstrated that snacks, which can contribute substantial energy (e.g., 400 kcal, 6 g protein, or approximately 15% of daily requirement), were not provided to/consumed by the majority of patients (90%), indicating an important missed opportunity in SCI patients.

To improve the current food provision and mealtime experience, patients and staff were invited to provide feedback for the current meal service in the SCIC. Inadequate taste, lack of choice and inappropriate texture were among the most commonly cited issues. This is of obvious importance when patients rely on hospital food to meet their nutritional needs. Furthermore, if the meals do not meet patients’ expectations, increasing portion sizes would be unlikely to result in higher intake.

To tackle hospital malnutrition, the implementation of protected mealtimes is one of the key areas in the Council of Europe resolution: Food and Nutritional Care in Hospitals,16 and this is included in the recent UK Government policy ‘Improving Nutritional Care’.15 Some of our patients and staff reported, however, that meals were not served on time and that there are frequent interruptions during mealtimes, suggesting this could relate to a lack of implementing of intended best practice. Indeed, recent evidence suggests that only minor improvements in mealtime experience can be expected after the implementation of ‘protected mealtime’,18 questioning its effectiveness, and perhaps further effort should be focused on dining companions or fortified hospital food.18

The present audit also support the need for more training of staff (catering staff and nursing staff) to educate regarding the importance of an effective meal service for hospital patients, and improve understanding of the benefits and implications of their role in the hospital.15 Ward staff may also require more assistance and consistent supervision to ensure the whole food service is adequate, and maximises food intake.15

The main limitation of this audit was that only one day’s food intake was assessed, and therefore, assumptions have necessarily been made about food provision and net intake across a very prolonged hospital stay sometimes. More data from each patient would provide stronger evidence to permit a more conclusive comparison with the national standard.19 However, it is highly probable that by just asking for one day’s intake, a higher response rate will be achieved.

After the survey, the SCI management team (physician, senior nurses and managers), catering management team, facility management team, occupational therapist, speech and language therapist, patient liaison officer and dietitian met to review the food-service needs. A nutrition steering group was set up to oversee food provision, and to evaluate the implementation of nutrition policy on a continuing basis.

In conclusion, this is the first study to report the nutrient intake and mealtime experience as perceived by both patients and carers in a UK SCIC. It identified some areas where there had probably already been improvement, but left little doubt that current SCI hospital catering and its implementation is under-performing to the probable disadvantage of patients. A greater need for tailoring of meals to meet the demands of the different patient groups was identified if we are to optimise patients’ nutritional intake. Effort should be made to ensure greater choice and better presentation for all meals. Periodic quality control helps to see that recommendations and patients’ expectations are being met.

Data archiving

There were no data to deposit.