Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
It's time to have a nutritional policy for nursing homes
- Carlo Pedrolli (9 May 2006)
-
A step in the right direction
- John C Oldroyd (19 May 2006)
-
Family Style Mealtimes.
- Deborah Foong (23 May 2006)
-
Protecting mealtimes -does this improve nutrition of elderly hospital patients?
- Abhay K Das, Tina McDougall, Robert M West, Jacqueline A J Smithson (24 May 2006)
-
Unit of Analysis
- Grazyna T Adamiak (26 May 2006)
-
Council of Europe Recommendations on Food and Nutritional Care in Hospitals - The need for an appropriate eating environment to prevent hospital malnutrition
- Sumantra Ray, Poonam Rana (30 May 2006)
|
|
|||
|
Carlo Pedrolli, Nutrion Unit S. Chiara Hospital 38100 Trento - ITALY
Send response to journal:
|
It is time to have a clear, definite, nutritional policy in nursing homes. From a personal survey in Trento's district more than 20% of nursing homes residents are under a some form of nutritional support (3% on enteral nutrition, 6% on oral nutritional support, 15% on liquid or semiliquid diet). In few nursing homes the weight is recorded after referral and in even less it is monitored in a planned way. Nutritional status in nursing homes is a "shame" even in more progressed countries; it is time to wake up and to do something about it. If the way to begin is to introduce family style mealtimes, so be it! Competing interests: None declared |
|||
|
|
|||
|
John C Oldroyd, Lecturer in Public Health Deakin University, 221 Burwood Highway, Burwood, Victoria, Australia, 3125
Send response to journal:
|
To the editor, I would like to applaud the study undertaken by Kristel Nijs and colleagues[1]. Many elderly people in nursing homes, despite excellent clinical care, experience poor nutritional status associated with poor dentition, depression and disease related malnutrition. It is terribly distressing for family to see their elderly relatives declining in such as way. The solution has always been to provide high protein and energy dietary supplements from pharmaceutical companies, that while very carefully formulated, are of debatable palatability and uptake. Importantly, they are prescribed by physicians at enormous cost to the health service and there is limited evidence of their effectiveness[2]. Food based interventions are not ineffective: appropriate studies simply have not been done. This research is a step in the right direction. Jamie, can you make nursing homes your next project? 1. Nijs K A, de Graaf C, Kok F J, van Staveren W A. Effect of family style mealtimes on quality of life, physical performance, and body weight of nursing home residents: cluster randomised controlled trial. Bmj, 2006. 2. Milne A C, Avenell A, Potter J. Meta-analysis: protein and energy supplementation in older people. Ann Intern Med, 144(1): 37-48, 2006. Competing interests: None declared |
|||
|
|
|||
|
Deborah Foong, SHO Urology New Cross Hospital, Wolverhampton, WV10 0QP
Send response to journal:
|
It is clear from this study that patients reap various benefits from family style meatimes. Apart from the measurable benefits gained, another reason for providing this to patients is simply because it is a more pleasurable way of dining. We should all stop to consider what the previous generation have done for us. The least we can do is treat them with more care and humanity. Competing interests: None declared |
|||
|
|
|||
|
Abhay K Das, Consultant Physician Pontefract General Infirmary, Friarwood Lane, Pontefract WF8 1PL, Tina McDougall, Robert M West, Jacqueline A J Smithson
Send response to journal:
|
SIR- The study by Nijs et al on Dutch elderly nursing home residents showed that improving the ambience at mealtimes prevents decline in body weight(1). Nutritional status decline during hospital stay not only due to illness but also due to problems with feeding and catering policies, such as poor environment, lack of assistance with eating and interruption of meals by procedures and ward rounds(2). Another reason is limited time available to eat each meal in the hospital(3). In the UK, protecting mealtime is an imaginative solution to improve nutrition of hospital patients under ‘Better Hospital Food’ programme of Department of Health. We performed a pilot study in two elderly medical wards at Castle Hill Hospital, in West Yorkshire to check the effect of ‘protected mealtimes’ in elderly hospital patients. The intervention was mealtime protection for an hour during lunchtime and evening mealtime. It was an open study with four months interval between control (17 patients) and intervention (22 patients). We found protecting mealtimes helped in preventing weight loss (0.19Kg/week vs. 0.25Kg/week) and reduction in hand grip strength (0.53Kg vs. 0.60Kg). There was gain (0.03cm/week) in mid arm circumference with mealtime protection while there was reduction (0.02cm/week) in the control group (p=0.056). Interestingly, we did not find protecting mealtimes to improve the food intake (calories: 1121/day vs. 1275/day, protein: 44gm/day vs. 50gm/day). Similar studies showed equivocal results in the past among elderly hospital patients as mentioned by Milne et al(1). As the protection of mealtimes has consequences for medical care it will be worth to undertake an equivalence trial. References: 1. Nijs K A N D, Graaf C, Kok F J, Staveren W A. Effect of family style mealtimes on quality of life, physical performance, and body weight of nursing home residents: cluster randomised controlled trial. BMJ 2006;332:1180-84 2. Allison S. Institutional feeding of the elderly. Curr opin Clin Nutr Metab Care 2002;5:31-4 3. Sydner YM, Fjellstrom C. Food provision and the meal situation in elderly care-outcomes in different social contexts. J Hum Nutr Dietet 2005;18:45-52 Competing interests: None declared |
|||
|
|
|||
|
Grazyna T Adamiak, PhD, MPH&W, MA unemployed
Send response to journal:
|
The present study reports on the effects of a preventive intervention designed as family style mealtimes on quality of life, physical performance and body weight of nursing home residents in Netherlands. However, there remain some tricky issues not really explored in the paper of Nijs et al. The main problem is the appropriateness of individual randomisation used by researchers. The definition of ”a family” employed in the study is problematic. The patients in the control group are served meals in a "balanced setting of residents (typically six per table) that appear to be stable and predetermined. In opposite, the patients in the control group are served meals at random seats. These are "assigned on basis of availability (typically six residents per table)". It seems that the intervention as well the control groups consist of in size almost equal groups of residents, six per table, and that these groups are the targeted clusters, not the individuals them selves. In behavioural, such as life style intervention or environmental interventions there is always risk of contamination, leading to choice of cluster randomisation. In this study individuals were allocated to the intervention and control groups of intervention despite that the intervention considered an environmental change, not easily isolated between the participating individuals. Thus, the use of the concept of "family style meals" is crucial, as it suggests that the ad hoc "families" should be the unit of allocation as well analysis. The contamination by ”the family style” is the crucial content of the intervention directed at creating a familiar environment for the intervention group (Slymen and Hovell, 1997). In the Introduction the Authors point out that a familiar environment is important in the context of deterioration etc. As the subjects of intervention are older people this fact may be important. A more stable and familiar setting may act as an important condition of the behavioural change. The Authors report that during the intervention of six months there are some dropouts’ as well new residents enter the two wards respectively when beds become vacant. This occurs not by choice. The new residents are not accounted for as environmental agents and not treated as subjects of the intervention despite that their continuous participation in the meals. A simple calculation shows that the changes in the composition of the wards differ, and thus the social environments are not comparable. The change in the composition of the experimental group is 29 persons as compared to 38 in the control group, which makes a difference of almost ten people. In addition, not only new participants in the common meals, the deaths may affect the well-being of the residents as well, in particular in the intervention group practicising a more familiar style of meals, which means establishment of more close relations. The behavioural change could not a priori be practised individually. This is the main argument for cluster allocation and cluster analysis on table-level. In addition, the Authors do not appear to adjust their analysis by inflation factor, and a comparison of contamination versus inflation is the base for design decision as well study size considerations (Slymen and Hovell, 1997). The Authors did not use intention-to-treat analysis despite that by dropouts we are assuming individuals who are no longer exposed to the intervention but who, nevertheless, have their outcomes measured (Slymen and Hovell, 1997). In the case of the present study the personnel' as well patients' behaviour is the target of the intervention. However, only patients’ behavioural change is measured and estimated. The intervention is based on interactions between the personnel and the meal participants, and thus clearly a fact that should be taken into account when deciding the principal unit of the experiment. According to Freemantle and Wood the use of hierarchical models is to prefer as "they allow all the information and variation at each level of the data to be explored while retaining the validity of the analysis." They also point out that considerable debate surrounds the choice of the unit of analysis in cluster analysis. The model selected should be well matched to the underlying structure of the data. Slymen and Hovell (1997) conclude that for many studies there is no choice but to use cluster randomisation. This is often the case in certain institutional settings, such as schools or families, where the institution may preclude random assignment of individuals. References Slymen, DJ, Hovell MF. Cluster versus Individual randomisation in Adolescent Tobacco and Alcohol Studies: Illustrations for design decision. International Journal of Epidemiology 1997; 26:765-771. Nick Freemantle, John Wood. Authors reply: Letters. Cluster randomised trials. BMJ 1999; 318:1286 (8 May) Competing interests: None declared |
|||
|
|
|||
|
Sumantra Ray, Clinical Research Fellow & Hon Clinical Teacher Division of Medicine & Therapeutics, Ninewells Hospital & Medical School, University of Dundee DD1 9, Poonam Rana
Send response to journal:
|
The Council of Europe (CoE) recommendations on Food and Nutritional
care in Hospitals (1) incorporate 114 recommendations under the following
5 themes:
These recommendations link very closely with national guidelines produced in the UK by SIGN (Scottish Intercollegiate Guidelines Network), NICE (national Institute of Clinical Excellence) and QIS (Quality Improvement Scotland - Standards for Food, Fluid and Nutritional Care). The overarching role of the Council of Europe Alliance (UK) is to put these guidelines into practice through practical initiatives at both local and national levels. Kristel et al (2) have effectively highlighted the importance of the eating environment and its impact on patient care, in their paper looking at the effects of family style mealtimes in nursing homes. Linking this to theme 4, recommendation 6 from the CoE document
(Bedside or dining room eating?) it becomes evident that the following
areas require attention in a hospital setting:
The above are based on evidence from Holmes (3) and McGlone et al (4), which have outlined the difficulties of eating in bed as well as the psychosocial factors which may limit a patient's food intake if it were available yet beyond reach. In the light of the above and emerging evidence from Kristel et al it is necessary to envisage randomised controlled trials looking at the effects of a family style eating environment on different groups of patients in a hospital setting. Dr Poonam Rana, Clinical Research Fellow Dr Sumantra Ray, Clinical Research Fellow & Hon Clinical Teacher References: 1. Food and Nutritional Care in Hospitals: How to prevent Undernutrition Report and recommendations of the Committee of Experts on Nutrition, Food Safety and Consumer Protection. Council of Europe Publishing, 2003. 2.Kristel A N D Nijs, Cees de Graaf, Frans J Kok, and Wija A van Staveren :'Effect of family style mealtimes on quality of life, physical performance, and body weight of nursing home residents: cluster randomised controlled trial',BMJ 2006; 332: 1180-1184 3. Holmes S. Nutrition: a necessary adjunct to hospital care? J R Soc Health 1999;119:175-9 4. McGlone PC, Dickerson JWT, Davies GJ. The feeding of patients in hospital: a review . J Royal Soc Health 1995 (oct): 282-8. Competing interests: None declared |
|||