From menu to mouth: the decay pathway of nutrient intake from planned menu to consumed and characteristics of residents in an aged care facility with greater nutrient decay rates: a cross-sectional study

Abstract

Objectives
To observe the cascade of nutrient loss from meals planned to those provided and subsequently consumed by older people in residential care. A secondary aim was to determine the characteristics of residents with higher nutrient loss resulting in lower intake of key nutrients.

Design
A single-centre cross-sectional study.

Setting
An aged residential care facility in Christchurch, New Zealand.

Participants
All low and high level of care residents except those who are end of life, enterally fed or on short-term stay were invited to participate in the study. 54 of 60 selected residents who consumed all three main meals (breakfast, lunch and dinner) for three non-consecutive days were included in the analyses.

Main outcome measures
Nutrient contents of planned menu; nutrient contents of meals served and consumed using modified 3-day diet records; and percentage of planned nutrients served and consumed.

Results
Vitamins C, B12 and folate had the greatest total decay rates of 50% or more from that planned to be consumed to what was actually consumed, while unsaturated fats, beta carotene, iodine and zinc had the lowest decay rates of 25% or less. Male participants and lower care level residents consumed significantly more nutrients, compared with female participants and those receiving higher level care. Increased age, female gender, higher level of care, smaller meal size, pureed diet and lower body mass index were associated with larger decay rates and lower nutrient intakes.

Conclusions
Not all planned and served food and beverages are consumed, contributing to potential multiple nutrient deficiencies including energy and protein in the majority of aged-care residents. As a consequence, some nutrients may need to be oversupplied if consumption is to match planned intakes.

Full text

1
NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
From menu to mouth: the decay
pathway of nutrient intake from
planned menu to consumed and
characteristics of residents in an aged
care facility with greater nutrient decay
rates: a cross-sectional study
Wathsala Sripali Nanayakkara,1 Paula Skidmore,1 Leigh O'Brien,1 Tim Wilkinson,2
Chris Frampton,2 Richard Gearry2
To cite: NanayakkaraWS,
SkidmoreP, O'BrienL, etal.
From menu to mouth: the decay
pathway of nutrient intake from
planned menu to consumed and
characteristics of residents in an
aged care facility with greater
nutrient decay rates: a cross-
sectional study. BMJ Open
2019;9:e024044. doi:10.1136/
bmjopen-2018-024044
►Prepublication history and
additional material for this
paper are available online. To
view these les, please visit
the journal online (http:// dx. doi.
org/ 10. 1136/ bmjopen- 2018-
024044).
Received 10 May 2018
Revised 24 September 2019
Accepted 24 September 2019
1Department of Human Nutrition,
University of Otago, Dunedin,
New Zealand
2Department of Medicine,
University of Otago, Christchurch
School of Medicine,
Christchurch, New Zealand
Correspondence to
Ms Wathsala Sripali
Nanayakkara;
wathsala. k11@ gmail. com
Original research
© Author(s) (or their
employer(s)) 2019. Re-use
permitted under CC BY-NC. No
commercial re-use. See rights
and permissions. Published by
BMJ.
Strengths and limitations of this study
►One researcher had gathered, entered and analysed
all data, removing interobserver variability.
►All dietary data were collected using a 3-day diet
records (162 food records) where all foods served
and consumed were observed and recorded by one
researcher, increasing the accuracy of data being
collected.
►Included as many participants as possible to provide
a true picture of the data being collected.
►Only one aged residential care facility was included
in the study, and the results cannot be representa-
tive of all residential care facilities in New Zealand.
ABSTRACT
Objectives To observe the cascade of nutrient loss
from meals planned to those provided and subsequently
consumed by older people in residential care. A secondary
aim was to determine the characteristics of residents with
higher nutrient loss resulting in lower intake of key nutrients.
Design A single-centre cross-sectional study.
Setting An aged residential care facility in Christchurch,
New Zealand.
Participants All low and high level of care residents
except those who are end of life, enterally fed or on short-
term stay were invited to participate in the study. 54 of 60
selected residents who consumed all three main meals
(breakfast, lunch and dinner) for three non-consecutive
days were included in the analyses.
Main outcome measures Nutrient contents of planned
menu; nutrient contents of meals served and consumed
using modied 3-day diet records; and percentage of
planned nutrients served and consumed.
Results Vitamins C, B12 and folate had the greatest
total decay rates of 50% or more from that planned to
be consumed to what was actually consumed, while
unsaturated fats, beta carotene, iodine and zinc had the
lowest decay rates of 25% or less. Male participants and
lower care level residents consumed signicantly more
nutrients, compared with female participants and those
receiving higher level care. Increased age, female gender,
higher level of care, smaller meal size, pureed diet and
lower body mass index were associated with larger decay
rates and lower nutrient intakes.
Conclusions Not all planned and served food and beverages
are consumed, contributing to potential multiple nutrient
deciencies including energy and protein in the majority of
aged-care residents. As a consequence, some nutrients may
need to be oversupplied if consumption is to match planned
intakes.
INTRODUCTION
A worldwide high prevalence of malnutri-
tion, specifically protein–calorie undernu-
trition in older people living in aged-care or
nursing homes, has been well described.1–5 As
the population aged over 65 years increases,
prevention and early treatment of under-
nutrition could potentially avoid serious
impacts on morbidity and mortality. A recent
systematic review6 on malnutrition in nursing
homes found that approximately half the
residents (47%–62%) were at risk of malnu-
trition and 20%–39% were malnourished. In
New Zealand, it is reported that 47% of older
adults recently admitted to age-related resi-
dential care are malnourished and a further
43% are at risk of malnutrition.7
Malnutrition or poor nutritional status is
associated with multiple adverse outcomes
including risk of infections,8 9 pressure sores,10
falls,11 increased length of hospital stay12 and
mortality12 in older people. Hence, malnu-
trition affects quality of life and increases
healthcare costs. Causes of malnutrition in
older people are multifactorial. Some nutri-
tional factors associated with undernutrition
include poor appetite, low protein intake,
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from

2NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
slow eating and dysphagia4 as well as other known ‘geri-
atric syndromes’ such as depression, dementia, functional
dependency and multiple comorbidities.13 These studies
have all looked at the end result of malnutrition. The
common response is to ensure that meals in residential
care are planned to obviate this problem. However, what
is planned is not necessarily what is eaten.
To our knowledge, there are no studies that describe the
decay rates of nutrients from the menu to food consumed
by residents living in aged-care facilities. Therefore, the
purpose of this paper is to assess the ‘total decay rate’ of
selected nutrients between planned meals for an aged-
care facility and those actually consumed by the resi-
dent. We also aimed to determine the characteristics of
residents with higher decay rates and lower intake of key
nutrients. In our study, we considered a nutrient to have
the greatest decay rate if less than 50% of the planned
nutrient was consumed by the residents. Alternatively, we
defined nutrient with the lowest decay rate if more than
75% of the planned nutrient was consumed.
METHODS
The protocol for this single-centre cross-sectional study is
described elsewhere.14 Briefly, residents receiving lower
levels of care (rest home) and higher level of nursing and
clinical care (hospital level) at a single aged-care facility
in Christchurch, New Zealand, were invited to partic-
ipate in the study. Exclusion criteria were short-term
stay, end-of-life, and enterally fed residents. Informed
consents were obtained from the participants, next of kin
or their legal representative for the study and their clin-
ical records were accessed to obtain weight, height, body
mass index (BMI), functional levels, medical diagnoses
and medications.
The 4-week cycle menus including recipes were
obtained from the facility, and by analysing these menus,
the average daily nutrients content was determined for
‘meals planned’ for the standard serving size used at the
facility. The daily menu includes three main meals (break-
fast, lunch and dinner), three snacks (morning tea, after-
noon tea and supper) and beverages and is designed to
meet the recommended daily intakes (RDIs) of most resi-
dents. Residents can choose between a ‘standard serving’
size, and a ‘smaller serving’ size, which is approximately
50% the size of the standard serving for lunch and dinner
meals. The desserts were served only in standard serving
size. Pureed meals (standard size only) were also provided
to selected residents who required these. Information on
‘meals served’ and ‘meals consumed’ for individual partic-
ipants were obtained using a combination of weighed/
estimated 3-day diet records to minimise work load and
interruptions during meal services. To do this, food
items for small and standard serving sizes were weighed
using a Salter Model 1010 electronic kitchen scale with
accuracy to within ±1 g (Salter Housewares Ltd, Kent,
UK, range 1–2000 g), then each participant’s meal was
photographed before and after consumption to estimate
weights of food or beverage provided and consumed.
The weight of the food and beverages consumed was
calculated by subtracting the estimated wastage (eg,
0%, 25% and 50%) from each served item. A research
dietitian collected the diet records for three non-con-
secutive days (two weekdays and one weekend day) for
three main meals and three snacks as well as beverages
provided (including oral nutritional supplements) and
consumed by the participants. ‘Meals consumed’ also
included additional food consumed by participants such
as food or beverages brought in by their friends and
family or consumed outside the facility as reported by the
participant or the family. Where possible, packaging was
checked to determine the ingredients and the weight of
these extra food or beverages consumed. Similarly, any
extra food or beverages consumed overnight (between
supper and breakfast) were recorded the next morning
from participants who could recall. All nutritional anal-
yses were conducted using FoodWorks 8 Professional
Edition (version 8.0.3553, 2015, Xyris Software, Highgate
Hill, Australia). Only those participants who consumed
all three main meals at the facility (rather than eating
outside of the facility) were included for analyses.
Estimated energy requirement was calculated using the
Schofield Equation15 of ((0.049×W)+2.459)×AF for men
and ((0.038×W)+2.755)×AF for women aged >60 where
W is for weight and AF is activity factor of 1.2 for bed or
chair bound and 1.4 for sedentary but mobile residents.
The individual estimated protein requirement was calcu-
lated using the recommended 1.07 g/kg for males and
0.94 g/kg for females.16 Average nutrients served and
consumed were determined by calculating the average
from all individual diet records analysed. These were
then compared with either estimated average require-
ment or adequate intake (AI) for Australia and New
Zealand16 as applicable for mean nutrient intake by the
participants. The percentage of participants meeting the
RDI was calculated using the average of 3-day diet records
per participant. The average percentage of each nutrient
served and consumed was obtained by averaging the indi-
vidual nutrient results of food records compared with
those planned on the menu on the day.
Non-parametric Mann-Whitney U tests were conducted
to compare the percentage of planned nutrients consumed
between two groups (ie, gender, level of care and cognitive
function). Linear regression models were used to describe
the relationships between the decay rates of macronutrients
from planned to consumed with age and BMI. A p value of
<0.05 was considered statistically significant.
Patient and public involvement
Patients/participants were not involved in the development
of the research question and design and outcome measures.
RESULTS
Data from a total of 54 residents were included in the anal-
yses. Participant characteristics are shown in table 1. The
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from

3
NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
Table 1 Baseline demographic and features of 54 participants
Characteristics All (n=54) Men (n=12) Women (n=42)
Average age±SD (years) 86.7±6.4 86.6±5.3 86.8±6.8
Age range (years) 67–99 78–94 67–99
Ethnicity
NZ European 49 (91%) 12 (100%) 37 (88%)
Other 5 (9%) 0 (0%) 5 (12%)
Average BMI (kg/m2) 23.7 25.5 23.2
BMI <18.5 8 (15%) 0 (0%) 8 (19%)
BMI <20 12 (22%) 0 (0%) 12 (29%)
BMI 20–24.9 23 (43%) 7 (59%) 16 (38%)
BMI 25–29.9 14 (26%) 4 (33%) 10 (24%)
BMI≥30 5 (9%) 1 (8%) 4 (9%)
Level of care
Rest home 24 (44%) 6 (50%) 18 (43%)
Hospital 30 (56%) 6 (50%) 24 (57%)
Comorbidities
Cardiovascular 40 (74%) 10 (83%) 30 (71%)
Central nervous 28 (52%) 7 (58%) 21 (50%)
Respiratory 15 (28%) 2 (17%) 13 (31%)
Thyroid 6 (11%) 1 (8%) 5 (12%)
Musculoskeletal 30 (56%) 6 (50%) 24 (57%)
Diabetes 9 (17%) 2 (17%) 7 (17%)
Cognitive impairment 17 (31%) 4 (33%) 13 (31%)
Psychiatric 13 (24%) 1 (8%) 12 (29%)
Cancer 14 (26%) 6 (50%) 8 (19%)
Lower GI 11 (20%) 4 (33%) 7 (17%)
Functional levels
Able to mobilise 38 (70%) 8 (67%) 30 (71%)
Bed/chairbound 16 (30%) 4 (33%) 12 (29%)
Normal swallow 46 (85%) 10 (83%) 36 (86%)
Dysphagia requiring pureed diet 8 (15%) 2 (17%) 6 (14%)
Median number of regular medications (IQR) 8 (6.25–10) 9 (5–11) 8 (7–9.75)
BMI, body mass index; GI, gastrointestinal; NZ, New Zealand.
mean age was 86.7±6.4 years, and all except one participant
were aged over 70 years. Forty-two participants (78%) were
women and 56% of the participants were receiving hospital
(high) level of care with 30% bed or chair bound.
Half of the 54 participants received standard size meals,
and the other half received smaller meals for lunch.
All male participants received standard meals, while 15
(36%) female participants received standard meals, and
the remaining were on smaller meals. For macronutrients,
a large discrepancy in served protein was seen (residents
on smaller meals received 20% less protein on average
compared with those on standard size meals, online
supplementary file 1), while least difference was seen
in total fats served (2%). Both participants on standard
and small meals did not consume all nutrients served.
Participants on smaller meals were served and consumed
a significantly lower percentage of nutrients from the
planned menu compared with standard meals for most
nutrients except for total fat, saturated, monounsaturated
and polyunsaturated fats and cholesterol, niacin, vitamins
C and E, folic acid, folate, retinol, sodium and iron.
The average nutrients (of 162 diet records) served and
consumed by the 54 participants as well as number (%) of
participants meeting the RDIs are summarised in table 2.
The menu did not fulfil RDIs or AIs for retinol, water, fibre,
magnesium, calcium, phosphorous and iodine. Addition-
ally, the menu exceeded the recommended % fat contri-
bution to total energy, while %protein was suboptimal. The
majority of participants did not consume enough to meet
the RDIs for vitamins E, B6, B12, folate, potassium, iron,
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from

4NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
Table 2 Recommended intakes compared with nutrients planned on the menu, meals served and consumed and number of
participants (%) meeting the recommendation
RDI/AI
Men
>70 years
RDI/AI
Women
>70 years Meals planned
Average nutrients (% of planned nutrients) and overall
number of participants (%) who would meet the
recommendations according to
Served menu Food consumed
Average, n
(%)
Overall, n
(%)
Average, n
(%)
Overall, n
(%)
Energy (kJ) IR 8790 8060 (91) 39 (72) 6440 (73) 20 (37)
Total protein (g) 81* 57* 72.1 64.1 (89) 29 (54) 49.8 (69) 12 (22)
% protein 15–25 13.8 13.4 13 (24) 13.0 8 (15)
Total fat (g) NA 88.5 80.8 (92) 65.5 (74)
% fat 20–35 37.8 37.6 17 (31) 38.0 15 (28)
Saturated fat (g) NA 40.6 35.9 (88) 29.6 (72)
Polyunsaturated fat (g) NA 10.3 10.5 (111) 8.29 (88)
Monounsaturated fat (g) NA 29.0 26.6 (94) 21.3 (75)
Cholesterol (mg) NA 290 26.6 (94) 21.3 (75)
Total carbohydrates (g) NA 244 222 (90) 177 (72)
% carbohydrates 45–65 46.4 46.1 28 (52) 46.2 32 (59)
Sugars (g) NA 146 125 (85) 99.0 (67)
Water‡ (g) 3400† 2800† 2570 2230 (86) 0 (0) 1550 (60) 0 (0)
Fibre (g) 30† 25† 23.1 20.3 (86) 4 (7) 15.8 (67) 3 (6)
Total vitamin A equivalent (µg) NA 1260 1220 (96) 920 (73)
Retinol (µg) 900* 700* 566 496 (86) 3 (6) 404 (69) 0 (0)
Beta carotene equivalents (µg) NA 4200 4350 (121) 3080 (88)
Thiamin (mg) 1.20* 1.10* 2.42 1.80 (74) 37 (69) 1.43 (60) 29 (54)
Riboavin (mg) 1.60* 1.30* 2.69 2.39 (87) 49 (91) 1.94 (71) 48 (89)
Niacin (mg) NA 19.8 15.8 (79) 12.4 (62)
Niacin equivalents (mg) 16* 14* 34.0 28.5 (83) 54 (100) 22.2 (65) 44 (81)
Vitamin C (mg) 45* 232 84.5 (37) 49 (91) 64.2 (28) 34 (63)
Vitamin E (mg) 10† 7† 11.3 8.08 (72) 33 (61) 6.42 (28) 19 (35)
Vitamin B6 (mg) 1.70* 1.50* 1.57 1.08 (73) 2 (4) 0.84 (57) 2 (4)
Vitamin B12 (µg) 2.40* 4.03 2.40 (63) 26 (48) 1.89 (49) 15 (28)
Folic acid (µg) NA 259 205 (77) 158 (61)
Folate food (µg) NA 375 229 (62) 181 (49)
Total dietary folate equivalent
(µg)
400* 841 566 (66) 35 (65) 440 (52) 26 (48)
Sodium§ (mg) 460–920† 2400 1970 (82) 54 (100) 1550 (64) 53 (98)
Potassium (mg) 3800† 2800† 3320 2590 (76) 10 (19) 2050 (61) 3 (7)
Magnesium (mg) 420* 320* 268 257 (94) 7 (13) 201 (74) 1 (2)
Calcium (mg) 1300* 871 775 (89) 0 (0) 629 (72) 0 (0)
Phosphorus (mg) 1000* 1320 1140 (87) 39 (72) 917 (70) 22 (41)
Iron (mg) 8* 14.4 9.57 (66) 36 (67) 7.76 (54) 25 (46)
Zinc (mg) 14* 8* 8.63 8.32 (97) 21 (39) 6.43 (75) 8 (15)
Selenium (µg) 70* 60* 60.0 52.8 (91) 9 (17) 40.5 (70) 2 (4)
Iodine (µg) 150* 122 135 (108) 17 (31) 103 (82) 8 (15)
*Recommended dietary requirement.
†AI.
‡Total water includes water from food and uids.
§Discretionary salt added during cooking or at the table not included.
AI, adequate intake; IR, individual requirement; NA, not available; RDI, recommended daily intake.
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from

5
NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
Table 3 Comparison of average nutrients served and consumed among men and women
Meals planned
Meals served (% of planned) Meals consumed (% of planned)
P value*Male (n=12) Female (n=42) Male (n=12) Female (n=42)
Energy (kJ) 8790 9010 (103) 7780 (88) 8000 (89) 6000 (68) <0.001
Energy (kcal) 2100 2150 (103) 1860 (88) 1910 (89) 1430 (68) <0.001
Total protein (g) 72.1 71.3 (96) 62.0 (86) 63.3 (86) 45.9 (65) <0.001
% protein 13.8 13.5 13.4 13.5 12.8 0.06
Total fat (g) 88.5 91.3 (102) 77.8 (89) 81.7 (92) 60.9 (70) <0.001
% fat 37.8 37.5 37.5 37.6 37.9 0.74
Total carbohydrates (g) 244 245 (99) 216 (88) 216 (87) 166 (68) <0.001
% carbohydrates 46.4 44.6 46.3 44.2 46.4 0.29
Water (g) 2570 2590 (100) 2120 (82) 1990 (77) 1430 (55) <0.001
Fibre (g) 23.1 22.2 (93) 19.8 (85) 19.4 (81) 14.7 (65) <0.001
*Mann-Whitney U test for % nutrients consumed between male and female participants.
Table 4 Comparison of average nutrients served and consumed among rest home and hospital level of care participants
Meals planned
Meals served (% of planned) Meals consumed (% of planned)
P value*Rest home (n=24) Hospital (n=30) Rest home (n=24) Hospital (n=30)
Energy (kJ) 8790 8300 (94) 7860
(88)
6950 (78) 6040 (68) <0.001
Energy (kcal) 2100 1980 (94) 1880 (88) 1660 (78) 1440 (68) <0.001
Total protein (g) 72.1 61.6 (85) 66.0 (91) 51.1 (72) 48.7 (68) 0.170
% protein 13.8 12.5 14.2 12.2 13.6 0.009
Total fat (g) 88.5 87.1 (101) 75.8 (85) 72.6 (84) 59.9 (67) <0.001
% fat 37.8 39.5 36.2 39.8 36.5 0.003
Total carbohydrates (g) 244 225 (91) 220 (90) 189 (76) 167 (68) 0.006
% carbohydrates 46.4 45.2 46.9 45.0 47.1 0.06
Water (g) 2570 2260 (87) 2200 (85) 1680 (65) 1450 (57) 0.007
Fibre (g) 23.1 21.0 (87) 19.8 (86) 17.5 (73) 14.4 (66) 0.006
*Mann-Whitney U test for % nutrients consumed between rest home and hospital level of care participants.
zinc and selenium in addition to those nutrients noted
above.
An average of 8790 kJ and 72.1 g of protein per day were
planned on the menu; however, on average 91% of energy
and 89% of protein were served and just 73% and 69% of
planned energy and protein, respectively, were consumed.
The proportions of total fat (74%) and carbohydrate (72%)
intake were similar to total energy consumed at 73%. More
than half (61%) of the participants did not meet their esti-
mated energy requirement, and 39 participants (72%) did
not meet their estimated protein needs.
Nutrients with the greatest decay rates (<50% planned
nutrients consumed) were: vitamin C (28%), followed by
vitamin B12 (49%) and folate (49%). The greatest decay
occurred between the amount of nutrients planned and
served. The nutrients that were consumed the most with
lowest decay rates (>75% planned nutrients consumed)
were: polyunsaturated fats (88%), beta carotenes (88%)
and iodine (82%).
Sixteen (30%) of the participants were receiving oral
nutritional supplements that provided on average 1330 kJ
(16% towards total intake per day) and 12.3 g protein
(17%) served and 1140 kJ (21%) and 9.77 g protein (21%)
consumed among these participants on supplements.
Table 3 summarises the differences between men and
women for nutrients served and consumed. As mentioned
previously, all male participants were served the stan-
dard meal serving size, and therefore, the served macro-
nutrients were close to 100% of that planned, and these
participants consumed approximately 90% of the planned
macronutrients. Women where the majority (64%) had
selected smaller meals at mid-day were served just below
90% of planned macronutrients on average and consumed
approximately 70% of planned. Hence, there were signif-
icant differences (p<0.001) between men and women for
the decay rates for consumed macronutrients except for
%protein, %fat and %carbohydrates contribution to total
energy.
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from

6NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
Figure 1 Decay rates of average energy, macronutrients, calcium and iron from planned menu by meal sizes, gender and
levels of care.
Table 4 summarises the nutrients served and consumed
by rest home and hospital level participants. Hospital
level participants had significantly greater decay rates
for all nutrients listed (p<0.05) except for protein and
sugar compared with rest home participants. However,
hospital participants were served more protein and sugar
compared with rest home participants.
Figure 1 depicts the decay rates of average energy, macro-
nutrients, calcium and iron planned, served and consumed
by meal size, gender and levels of care. In general, male
participants had the smallest decay rates, while small
meal size, female and hospital-level participants had the
greatest decay rates resulting in lower intakes of energy and
nutrients. Female and hospital participants had the same
decay patterns for energy and iron.
Seven (41%) of the participants with cognitive impair-
ment were either assisted or fully fed by the staff. The decay
rates were similar, and there were no statistically signifi-
cant differences between those with or without cognitive
impairment (online supplementary file 2). Similarly, there
were no significant differences between the decay rates of
macronutrients between mobile and bed or chair-bound
participants (online supplementary file 3). In contrast, six
participants who received pureed meals consumed signifi-
cantly lower total energy, fat, carbohydrates and water with
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from

7
NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
greater decay rates compared with participants on regular
textured diet (online supplementary file 4).
Linear regression models showed significant nega-
tive associations between decay rates (from planned to
consumed nutrients) and age for macronutrients. Specif-
ically, the reductions were 11.5% (p=0.004) for energy,
12.5% (p=0.005) for protein, 9.7% (p=0.055) for fat and
11.7% (p=0.003) for carbohydrates for every 10-year
increase in age. Similarly, there were significant negative
associations between decay rates and BMI. These reduc-
tions were 16.5% (p=0.002) for energy, 14.3% (p=0.020)
for protein, 17.5% (p=0.010) for fat and 14.7% (p=0.007)
for carbohydrates for every 10 kg/m2 decline in BMI.
DISCUSSION
To our knowledge, this is the first study to investigate the
cascade effect of nutrient intake decay from a planned
menu to what is actually consumed by residents living in an
aged-care facility. Our primary aim was to observe the total
decay rate (percentage of planned nutrients consumed by
the residents) to help identify those who are most at risk of
inadequate nutrient intake. The results showed that not all
food and beverages served were consumed by the residents.
Those consuming smaller size meals, female and hospital
(high) level of care participants had the greatest decay
rates for total energy, macronutrients, iron and calcium. We
acknowledge that in our study, only female residents (64%
of female residents) had selected small meal size, and the
decay rates for meals served and consumed are affected for
women for this reason. Among macronutrients, least decay
rates were seen in fats followed by carbohydrates and protein
had the greatest decay rates. While protein intake is known
to be important in older adults, particularly to prevent or
reduce the effects of sarcopenia, our study indicated that
the majority (72%) of aged-care residents did not consume
adequate protein. Similarly 61% of these residents did not
meet their estimated energy requirements. Furthermore,
increasing age and reduction in BMI were also associated
with significantly lower energy and macronutrient intake.
We found that vitamin C had the greatest decay rates;
however, 91% of the participants were served and 63%
consumed the recommended 45 mg/day on average.
The discrepancy was due to the fact that vitamin C-rich
items such as orange juice, cranberry juice and kiwifruit
on the menu were not necessarily selected by the partic-
ipants. Additionally, not all participants consumed all of
the canned or fresh fruit being served, hence the drop
in percentage of residents meeting the recommended
intake. Other common food wastes that contributed to
decay rates of macronutrients included bread or toast,
particularly the crusts, preordered sandwiches at the
evening meal, coleslaw, green salads, milk served with
breakfast cereals, porridge and in tea, meats, pasta
dishes, instant puddings offered as pureed desserts,
some vegetables such as courgettes, leeks and cabbage.
The greater decay rates of protein may be due to items
such as meats and milk-containing food and beverages
being left unconsumed. However, it was evident that the
%protein contribution to total energy was below recom-
mendation and % fat exceeded the recommended level
on the menu, and those participants on smaller meals
were served significantly less protein (meats) but similar
amounts of fats to those on standard serving size.
We did not observe any significant differences in nutri-
tional intake or decay rates between those with cognitive
impairment and those without. All food intakes from
breakfast to supper were observed; however, the details
of food consumed overnight were collected from those
who could recall. We believe that this would not have
made a significant impact on the results as the menu
(which covers all food offered from breakfast to supper)
is intended to meet most residents’ nutritional require-
ments. The reason for not observing significant differ-
ences may be due to feeding assistance provided by staff
to 41% of the participants with cognitive impairment.
Indeed, the provision of regular feeding or eating assis-
tance has been shown to improve food intake in aged-
care residents.17 However, feeding assistance is often
not provided in care facilities due to staff and time
constraints. Provision of such assistance is also important
since malnourished patients are more likely to consume
≤50% meals compared with well-nourished patients.18
Our study found that, on average, residents consumed
6440 kJ (1540 kcal/day), which is similar to that described
in previously published studies17 19–22 (average 1507–1522
kcal/day). However, the average protein intake of 49.8 g/
day was much lower than that described previously17 20–22
(54.8–58 g/day). Similar to our study, Buckinx et al21 also
found that not all meals served in nursing homes were
consumed by the residents and that there were signif-
icant differences between energy and macronutrient
contents of foods served and consumed. Equally, a Cana-
dian study19 found that despite an average normal BMI,
these elderly residents did not meet the requirements for
macronutrient and micronutrient intake. These studies
support our findings that the majority of aged-care resi-
dents are at risk of nutritional inadequacy.
It is important to note that although we found that male
participants had the lowest decay rates for nutrients, the
majority of study participants including men were found
to have inadequate consumption of multiple nutrients
namely protein, water, fibre, vitamins A, E, B6, B12, folate,
potassium, magnesium, calcium, phosphorus, iron, zinc,
selenium and iodine. Unfortunately, the RDI of 11 (69%)
of these 16 nutrients was not met by the planned menu,
particularly for men who have higher RDIs.
We should emphasise that a residential care institu-
tion should deliberately plan to oversupply meals to its
residents, rather than not serving enough. This means
that what is served should, on average, always be greater
than what is consumed. However, this is the first study to
quantify these gaps and to show which nutrients may be at
greater risk of being underconsumed. However, serving
meals that are too large can discourage people from
eating. A solution may be to offer second servings at meal
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from

8NanayakkaraWS, etal. BMJ Open 2019;9:e024044. doi:10.1136/bmjopen-2018-024044
Open access
times and to offer snacks between meals. In both cases,
ideally, these would be oversupplied with nutrients known
to be at risk, such as protein, fibre, vitamins A, B6, magne-
sium, calcium and zinc. Furthermore, given protein had
one of the greater decay rates, the menu should focus on
offering more protein at each meal rather than fats as fats
are generally well consumed by the residents.
As with any study, there are potential limitations to this
research. As all recruitment occurred at one facility, and
all participants were Caucasian, the study participants may
not be representative of all New Zealand aged-care facili-
ties. One research dietitian collected all dietary data elim-
inating interobserver error and used weighed food record
in combination with a validated visual assessment method
of plate waste.23 Our study did not screen for malnutri-
tion; however, 22% of the participants were found to have
low BMI (<20 kg/m2 including 15% with BMI <18.5). We
suggest future studies compare differences in nutrient
intake between well-nourished and malnourished resi-
dents identified using a nutritional screening tool such
as Mini Nutritional Assessment.24 Some important nutri-
ents such as vitamin D were not analysed in this study;
however, they could be assessed in future studies.
CONCLUSION
Despite menu planning, approximately 90% of nutrients
planned on the menu are served to the residents and about
70% are consumed. Subsequently, multiple nutrient defi-
ciencies exist including suboptimal energy and protein
intake in aged-care residents in New Zealand. Female sex,
increasing age, smaller meal size, hospital level of care,
lower BMI and consuming a pureed diet were associated
with lower nutrient intake with greater nutrient decay rates.
However, male residents were more likely to consume
most of nutrients planned and served at the facility. The
study highlighted the need to increase protein and energy
content on the menu, and we suggest routinely offering
protein-rich snacks as in-between meal snacks and offering
larger meals or alternatively second helping to residents.
Contributors WSN, RG, LO, TW and PS were responsible for the planning of this
study. WSN was responsible for participant recruitment, data collection and entry
and drafting the manuscript under the supervision of RG, LO, TW and PS; CF was
responsible for support with statistical analysis and reporting. All authors provided
critical review and revision of the manuscript and have read and approved the nal
version of the manuscript.
Funding The authors have not declared a specic grant for this research from any
funding agency in the public, commercial or not-for-prot sectors.
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval The study was approved by The University of Otago Human Ethics
(Health) Committee (H15/053).
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement No data are available.
Open access This is an open access article distributed in accordance with the
Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which
permits others to distribute, remix, adapt, build upon this work non-commercially,
and license their derivative works on different terms, provided the original work is
properly cited, appropriate credit is given, any changes made indicated, and the use
is non-commercial. See:http:// creativecommons. org/ licenses/ by- nc/ 4. 0/.
REFERENCES
1 Borgstrom Bolmsjo B, Jakobsson U, Molstad S, etal. The nutritional
situation in Swedish nursing homes - a longitudinal study. Archives of
gerontology and geriatrics 2015;60:128–33.
2 Gaskill D, Black LJ, Isenring EA, etal. Malnutrition prevalence and
nutrition issues in residential aged care facilities. Australas J Ageing
2008;27:189–94.
3 Kaiser MJ, Bauer JM, Rämsch C, etal. Frequency of malnutrition in
older adults: a multinational perspective using the mini nutritional
assessment. J Am Geriatr Soc 2010;58:1734–8.
4 Keller HH. Malnutrition in institutionalized elderly: how and why? J
Am Geriatr Soc 1993;41:1212–8.
5 Saletti A, Lindgren EY, Johansson L, etal. Nutritional status
according to mini nutritional assessment in an institutionalized elderly
population in Sweden. Gerontology 2000;46:139–45.
6 Bell CL, Tamura BK, Masaki KH, etal. Prevalence and measures
of nutritional compromise among nursing home patients: weight
loss, low body mass index, malnutrition, and feeding dependency,
a systematic review of the literature. J Am Med Dir Assoc
2013;14:94–100.
7 Watkin RS, Gammon CS, Allen J, etal. Screening for nutrition
risk and dysphagia among older adults recently admitted to age
related residential care within the waitemata dhb region. Nutrients.
Conference: Nutrition Society of New Zealand Annual Conference
2016;9.
8 Gavazzi G, Krause K-H. Ageing and infection. Lancet Infect Dis
2002;2:659–66.
9 Schneider SM, Veyres P, Pivot X, etal. Malnutrition is an
independent factor associated with nosocomial infections. Br J Nutr
2004;92:105–11.
10 Shahin ESM, Meijers JMM, Schols JMGA, etal. The relationship
between malnutrition parameters and pressure ulcers in hospitals
and nursing homes. Nutrition 2010;26:886–9.
11 Johnson CS. The association between nutritional risk and falls
among frail elderly. J Nutr Health Aging 2003;7:247–50.
12 Stratton RJ, King CL, Stroud MA, etal. ‘Malnutrition Universal
Screening Tool’ predicts mortality and length of hospital stay in
acutely ill elderly. British Journal of Nutrition 2006;95:325–30.
13 Saka B, Kaya O, Ozturk GB, etal. Malnutrition in the elderly and
its relationship with other geriatric syndromes. Clinical Nutrition
2010;29:745–8.
14 Nanayakkara W, Gearry R, Muir J, etal. Gastrointestinal symptoms
and FODMAP intake of aged-care residents from Christchurch, New
Zealand. Nutrients 2017;9:1083.
15 Schoeld WN. Predicting basal metabolic rate, new standards
and review of previous work. Human nutrition. Clinical nutrition
1985;39:5–41.
16 National Health and Medical Research Council AGDoHaA, New
Zealand Minisry of Health. Nutrient reference values for Australia
and New Zealand including recommended dietary intakes. Canberra
2006.
17 Keller HH, Carrier N, Slaughter SE, etal. Prevalence and
determinants of poor food intake of residents living in long-term care.
J Am Med Dir Assoc 2017;18:941–7.
18 Agarwal E, Ferguson M, Banks M, etal. Nutritional status and dietary
intake of acute care patients: results from the nutrition care day
survey 2010. Clinical Nutrition 2012;31:41–7.
19 Aghdassi E, McArthur M, Liu B, etal. Dietary intake of elderly living in
Toronto long-term care facilities: comparison to the dietary reference
intake. Rejuvenation Res 2007;10:301–10.
20 Iuliano S, Poon S, Wang X, etal. Dairy food supplementation may
reduce malnutrition risk in institutionalised elderly. British Journal of
Nutrition 2017;117:142–7.
21 Buckinx F, Allepaerts S, Paquot N, etal. Energy and nutrient content
of food served and consumed by nursing home residents. J Nutr
Health Aging 2017;21:727–32.
22 Iuliano S, Olden A, Woods J. Meeting the nutritional needs of elderly
residents in aged-care: are we doing enough? J Nutr Health Aging
2013;17:503–8.
23 Sherwin AJ, Nowson CA, McPhee J, etal. Nutrient intake at meals in
residential care facilites for the aged: validated visual estimation of
plate waste 1998;55:188–93.
24 Guigoz Y, Vellas B, Garry PJ. Assessing the nutritional status of
the elderly: the mini nutritional assessment as part of the geriatric
evaluation. Nutr Rev 1996;54:S59–65.
on October 16, 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-024044 on 16 October 2019. Downloaded from