Applied Physiology, Nutrition, and Metabolism

Adequate vitamin B12 and riboflavin status from menus alone in residential care facilities in the Lower Mainland, British Columbia

Journal: Applied Physiology, Nutrition, and Metabolism

Manuscript ID apnm-2018-0459.R1

Manuscript Type: Article

Date Submitted by the 07-Sep-2018 Author:

Complete List of Authors: Whitfield, Kyly; Mount Saint Vincent University, Department of Applied Human Nutrition da Silva, Liz;Draft Fraser Health Feldman, Fabio; Fraser Health; Simon Fraser University Department of Biomedical Physiology and Kinesiology Singh, Sonia; Fraser Health; University of British Columbia, Department of Family Practice McCann, Adrian; Bevital AS McAnena , Liadhan; Ulster University, Nutrition Innovation Centre for Food and Health Ward, Mary; Ulster University, Nutrition Innovation Centre for Food and Health McNulty, Helene; Ulster University, Nutrition Innovation Centre for Food and Health Barr, Susan; University of British Columbia, Green, Tim; South Australian Health and Medical Research Institute, Healthy Mothers, Babies and Children Theme

vitamin B12 (cobalamin), riboflavin (vitamin B2), older adults, residential Keyword: care, menu assessment, folate

Is the invited manuscript for consideration in a Special Not applicable (regular submission) Issue? :

https://mc06.manuscriptcentral.com/apnm-pubs Page 1 of 25 Applied Physiology, Nutrition, and Metabolism

1 Adequate vitamin B12 and riboflavin status from menus alone in residential care 2 facilities in the Lower Mainland, British Columbia 3 4 Kyly C Whitfield*1, Liz da Silva2, Fabio Feldman2,3, Sonia Singh2,4, Adrian McCann5, 5 Liadhan McAnena6, Mary Ward6, Helene McNulty6, Susan I Barr7, Tim J Green8 6 7 1. Department of Applied Human Nutrition, Mount Saint Vincent University, Halifax, 8 Canada 9 [email protected] 10 11 2. Fraser Health Authority, Surrey, Canada 12 [email protected] 13 14 3. Department of Biomedical Physiology and Kinesiology, Simon Fraser University, 15 Burnaby, Canada 16 [email protected] 17 18 4. Department of Family Practice, University of British Columbia, Vancouver, Canada 19 [email protected] 20 21 5. Bevital AS, Bergen, Norway Draft 22 [email protected] 23 24 6. Nutrition Innovation Centre for Food and Health, Ulster University, Coleraine, 25 Northern Ireland 26 [email protected] 27 [email protected] 28 [email protected] 29 30 7. Food, Nutrition and Health, University of British Columbia, Vancouver, Canada 31 [email protected] 32 33 8. Healthy Mothers, Babies and Children Theme, South Australia Health and Medical 34 Research Institute, Adelaide, South Australia, Australia 35 [email protected] 36 37 *Corresponding Author: Dr. Kyly C Whitfield, Assistant Professor, Department of 38 Applied Human Nutrition, Mount Saint Vincent University, 166 Bedford Hwy, Halifax, 39 Nova Scotia, B3M 2J6, Canada. Telephone: (902) 457-5978; Email: 40 [email protected] 41

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 2 of 25

42 ABSTRACT

43 Older adults have potential increased risk of nutrient deficiencies due to age-related

44 decreased dietary intake and malabsorption; it is important to ensure nutrient needs are

45 met to avoid adverse health outcomes. B vitamins are of particular interest: vitamin B12

46 deficiency can cause irreversible neurodegeneration; there is mandatory folic acid

47 fortification in Canada; and suboptimal riboflavin status has been reported among older

48 adults in the United Kingdom.

49 In this exploratory secondary analysis study we assessed vitamin B12 and riboflavin

50 biochemical status (via microparticle enzyme immunoassay and erythrocyte glutathione

51 reductase activity coefficient (EGRac), respectively), and the vitamin B12, riboflavin, and

52 folate content of menus served to a convenienceDraft sample of older adults (≥65 years) from

53 five residential care facilities within the Lower Mainland of British Columbia, Canada.

54 Diet was assessed from customized 28-day cycle meal plans.

55 Participants (n=207, 53 men and 154 women) were 86 ± 7 years, largely of European

56 descent (92%), and non-smokers (95%). The menus served had a low prevalence of

57 inadequacy for vitamin B12 and riboflavin (only 4% and 1% of menus contained less than

58 the estimated average requirement [EAR], respectively), but 93% contained less than the

59 EAR for folate. Mean ± SD serum total vitamin B12 concentration was 422 ± 209 pmol/L,

60 and EGRac was 1.30 ± 0.19. The majority of older adults in residential care were

61 provided with adequate vitamin B12 and riboflavin menu amounts, and only 5% were

62 vitamin B12 deficient (<148 pmol/L). However, 26% were riboflavin deficient (EGRac

63 ≥1.4), which may warrant further investigation.

64 65

https://mc06.manuscriptcentral.com/apnm-pubs Page 3 of 25 Applied Physiology, Nutrition, and Metabolism

66 67 KEYWORDS: vitamin B12 (cobalamin), riboflavin (vitamin B2), older adults, residential 68 care, menu assessment, folate

Draft

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 4 of 25

69 INTRODUCTION

70 Older adults are at potential risk of nutrient deficiencies due to decreased dietary intake,

71 dysphagia, and decreased absorption of some nutrients. Psychosocial factors such as

72 forgetfulness (missed meals), eating alone, and bereavement also contribute to this

73 increased risk (Keller 2007, Bales and Johnson 2014). The Canadian population is aging:

74 the proportion of Canadians over 65 years is expected to nearly double from 15% in 2013

75 to 24-28% by 2063 (National Population Projections Team et al. 2015), some of whom

76 will live their last years in residential care. There were nearly 250,000 Canadians living

77 in 4,600 residential care facilities in 2010 (Minister of Industry Statistics Canada 2011).

78 Although only 7% of adults over 65 years live in residential care facilities, this proportion

79 climbs to 30% among those over 85Draft years (Garner et al. 2018). Since diet is one of the

80 few modifiable risk factors for healthy aging at this life stage (Keller 2007), it is

81 important to identify high risk micronutrient deficiencies to prevent disease, decrease

82 health costs, and improve quality of life.

83

84 Vitamin B12 deficiency is purported to be more common among older adults due to

85 decreased absorption of food-bound vitamin B12 because of a decrease in gastric acid and

86 intrinsic factor production (McNulty and Scott 2008). In addition, mandatory folic acid

87 fortification in Canada and elsewhere increases the concern of vitamin B12 deficiency in

88 this group as high dietary folate intake can mask the hematological markers of B12

89 deficiency, often prolonging time to diagnosis while irreversible neurodegeneration

90 persists (McNulty and Scott 2008, Macfarlane et al. 2011). For these reasons, adults over

91 the age of 51 years are advised to consume most of their vitamin B12 from fortified foods

https://mc06.manuscriptcentral.com/apnm-pubs Page 5 of 25 Applied Physiology, Nutrition, and Metabolism

92 or supplements (Institute of Medicine 1998a). The Canadian Health Measures Survey

93 (CHMS; 2007-09) found that only one fifth of free-living older adults aged 60-79 years

94 had marginal or deficient vitamin B12 status (total serum vitamin B12 ≤ 220 pmol/L)

95 (Macfarlane et al. 2011). However, little is known about the dietary or biochemical

96 vitamin B12 status among older Canadian adults living in residential care facilities.

97

98 Riboflavin (vitamin B2) status of older adults is also of interest due to recent findings

99 from the United Kingdom, where nearly half of free-living adults >65 years had

100 suboptimal or deficient biochemical riboflavin status (as indicated by erythrocyte

101 glutathione reductase activity coefficient (EGRac) > 1.3), despite the mean daily dietary

102 intake of riboflavin from food and supplementsDraft among this group being considerably

103 above the Reference Nutrient Intake (Bates et al. 2014). Poor riboflavin status can

104 manifest with neurodegeneration, anemia, endocrine dysfunction, and skin and eye

105 inflammation (Said and Ross 2014).

106

107 Older adults living in residential care facilities could be at increased risk of B-vitamin

108 deficiencies due to anorexia and cachexia associated with conditions like dementia

109 (Donini et al. 2013, Sekerak and Stewart 2014), as well as exposure to folic acid

110 fortification (Macfarlane et al. 2011). Therefore we investigated vitamin B12 and

111 riboflavin status, and calculated the amounts of vitamin B12, riboflavin, and folate in

112 menus provided to older adults living in residential care facilities throughout the Lower

113 Mainland of British Columbia, Canada.

114

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 6 of 25

115 MATERIALS AND METHODS

116 Study design

117 This cross-sectional analysis used baseline data from a larger study investigating high-

118 dose vitamin D supplementation among older adults living in residential care facilities

119 (Feldman et al. 2014). Briefly, a convenience sample of older adults was recruited from

120 five residential care facilities in the Lower Mainland of British Columbia, Canada

121 between November 2012 and February 2013. Eligibility criteria required that participants

122 be aged ≥65 years, have resided in the facility for >3 months, and not be receiving enteral

123 nutrition support. Written, informed consent was obtained from participants or their

124 designated representative. Ethics approval was obtained from the University of British

125 Columbia Clinical Research Ethics BoardDraft (H12-02503) and Fraser Health Research

126 Ethics Board (FHREB 2012-058).

127

128 Blood and data collection

129 Demographics, health information, anthropometric measures, and information on whether

130 or not micronutrient supplements were used (but not supplement composition) were

131 collected from participants’ medical charts. Non-fasting blood samples were collected

132 from each participant into two evacuated tubes, one trace element-free (for serum) and

133 one containing EDTA (for washed erythrocytes; Vaccutainer, Becton Dickinson).

134 Samples were transported on ice to the BC Children’s Hospital Research Institute

135 (Vancouver, Canada) for processing and storage. Blood samples were spun in a 4°C

136 refrigerated centrifuge, serum was collected, and then erythrocytes were washed three

137 times with PBS (Amresco); all samples were stored at -80°C until analysis.

https://mc06.manuscriptcentral.com/apnm-pubs Page 7 of 25 Applied Physiology, Nutrition, and Metabolism

138

139 Biochemical vitamin analysis

140 Serum total vitamin B12 was measured by microparticle enzyme immunoassay using an

141 AxSYM immunochemical automated analyzer (Abbott Laboratories) at the BC

142 Children’s Hospital Research Institute (Vancouver, Canada). Riboflavin status was

143 assessed using EGRac, a functional indicator that provides a measure of tissue riboflavin

144 saturation that has long been viewed as the gold standard for biochemical riboflavin

145 assessment (McAuley et al. 2016). EGRac is a ratio of the in vitro activity of glutathione

146 reductase in washed red cells before and after the addition of excess riboflavin-dependent

147 cofactor, flavin adenine dinucleotide (McAuley et al. 2016). Lower EGRac ratios (closer

148 to a ratio of 1.0) are indicative of riboflavinDraft sufficiency, while higher ratios indicate poor

149 status (Wilson et al. 2012). Washed erythrocytes were batch analyzed at Ulster University

150 (Coleraine, Northern Ireland) using an iLab 650 Clinical Chemistry Analyzer

151 (Instrumentation Laboratory).

152

153 Menu assessment

154 All participants consumed diets based on a standardized 28-day cycle meal plan designed

155 to align with recommendations from Canada’s Food Guide (Health Canada 2011), and

156 customized to meet individual dietary needs (i.e. therapeutic restrictions, allergies,

157 texture) and preferences. The 28-day customized menus served to each participant were

158 assessed, and individual means for total energy (daily kcal), macronutrient profile,

159 riboflavin, vitamin B12, and folate (as dietary folate equivalents, accounting for

160 differences in bioavailability of folate and folic acid) content were computed for each

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 8 of 25

161 participant using CBORD® clinical nutrition assessment software (CBORD Group Inc,

162 Ithaca, NY, USA). Nutrient composition data were either from the Canadian Nutrient File

163 (Health Canada, Ottawa, Ontario, Canada) or the manufacturer for pre-prepared food

164 items. Dietary Reference Intakes (DRI) were used to evaluate dietary adequacy of the

165 menus; specifically we defined inadequacy as the proportion of the group with menus

166 below the Estimated Average Requirement (EAR) for the nutrient (Otten et al. 2006).

167 Note that although participants may have been assigned similar diets, their nutrient values

168 differed based on personal preferences (e.g. beverage and dessert choices) or other

169 considerations. Here we present menu analysis, which reflects the nutrient content within

170 menus rather than actual dietary intake. Moreover, although use of micronutrient

171 supplements was recorded from participants’Draft medical charts, data on the specific

172 nutrients consumed in supplemental form were not available; thus nutrient amounts

173 reflect intake only from foods provided by menus.

174

175 Data analysis

176 Descriptive statistics were computed and expressed as mean ± SD or n (%). The

177 following cut-offs for total serum vitamin B12 concentration, which were also used for the

178 Canadian Health Measures Survey, were used for vitamin B12 status: deficient, < 148

179 pmol/L; marginal, 148-220 pmol/L; sufficient, ≥221 pmol/L (Macfarlane et al. 2011).

180 Riboflavin status cut-offs were: deficient, EGRac ≥ 1.4; suboptimal, EGRac 1.3 - 1.4;

181 sufficient, EGRac < 1.3 (Horigan et al. 2010).

182

https://mc06.manuscriptcentral.com/apnm-pubs Page 9 of 25 Applied Physiology, Nutrition, and Metabolism

183 Vitamin B12 concentration, EGRac, and menu total energy, carbohydrate, protein, fat,

184 vitamin B12, folate (DFE), and riboflavin were not normally distributed (Shapiro Wilks

185 test p<0.05), and could not be transformed into a normal distribution, so a Mann Whitney

186 U test was employed to test differences by sex and supplement consumption. Age was

187 normally distributed; an independent t-test was used to assess differences in age by sex.

188 BMI and health diagnoses were ln-transformed to achieve normality before using an

189 independent t-test to assess differences by sex. Chi square tests were used to assess

190 whether there were differences in the proportions for vitamin B12 and riboflavin status

191 categories by sex and supplement use; and ethnicity, smoking status, BMI category, food

192 allergy, %

193 assess associations between vitaminDraft B12 concentration or EGRac and time spent in

194 residential care (months), age (years), and nutrient content of menus. All analyses were

195 performed with SPSS for Macintosh version 23.0 (IBM Corp., Armonk, NY), with a

196 significance level of p<0.05.

197

198 RESULTS

199 Of the 739 residents approached in the five residential care facilities, data were collected

200 from 236 (n=329 unable to consent and/or designate could not be contacted; n=100

201 refused; n=58 did not meet eligibility criteria; n=16 in palliative care). Of those, complete

202 biomarker and dietary data were available for 207 participants. Demographic

203 characteristics of participants can be found in Table 1. Most participants were of

204 European descent (92%) and non-smokers (95%), with a BMI (mean ± SD) of 25.0 ± 5.4

205 kg/m2. Women were significantly older than men: 86 versus 83 years (p=0.003). Most

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 10 of 25

206 participants had between 4-5 medically diagnosed conditions, with a range from 0 (2

207 women) to 15 (1 woman). There were few differences between the study participants and

208 other residents from the same facilities (Feldman et al. 2014). One notable exception was

209 that the study included fewer participants with dementia than non-participating residents

210 in the same facilities (p=0.002).

211

212 Biochemical status markers for vitamin B12 and riboflavin did not differ between men

213 and women, as shown in Table 2. Most participants were vitamin B12 sufficient (81%;

214 ≥221 pmol/L) and only 5% were deficient (<148 pmol/L), and the mean ± SD serum

215 vitamin B12 concentration was 422 ± 209 pmol/L. The mean ± SD EGRac ratio was 1.30

216 ± 0.19; just over one quarter of participantsDraft were classified as riboflavin deficient

217 (EGRac ≥ 1.4; 26%) while 14% were considered suboptimal (EGRac ≤1.3 to < 1.4).

218

219 The nutrient content of menus served is shown in Table 3. Approximately half of

220 participants consumed a general diet, and a similar proportion had a diet with regular

221 texture. Only 3 (1%) of menus served contained less than the EAR for riboflavin of 1.1

222 mg/day for men and 0.9 mg/day for women (Institute of Medicine 1998b). Similarly, only

223 2% of men’s and 5% of women’s menus served contained vitamin B12 below the EAR of

224 2.0 μg/d (Institute of Medicine 1998a). The majority of vitamin B12 would be food-

225 bound, given the low prevalence of supplement use and consumption of vitamin B12-

226 fortified foods (fortified soy beverage for the vegetarian diet only, n=3). Conversely,

227 93% of menus served contained less than the folate EAR of 320 μg/d DFE (Institute of

228 Medicine 1998c). Micronutrient supplementation was not common, with less than one

https://mc06.manuscriptcentral.com/apnm-pubs Page 11 of 25 Applied Physiology, Nutrition, and Metabolism

229 quarter of participants consuming an oral micronutrient supplement, and only 4%

230 receiving a monthly intramuscular vitamin B12 injection.

231

232 Participants consuming oral micronutrient supplements had significantly better nutrient

233 status: higher vitamin B12 concentrations (474 ± 186 vs. 407 ± 213; p=0.024) and lower

234 EGRac (1.20 ± 0.15 vs. 1.33 ± 0.20; p<0.001) compared to participants not consuming an

235 oral micronutrient supplement (Table 2). Similarly, the proportion of participants

236 classified as riboflavin deficient/suboptimal/sufficient differed significantly by

237 supplement consumption (p=0.007). However, vitamin B12 status classifications by

238 supplement consumption did not differ (p=0.136). The few participants receiving a

239 monthly vitamin B12 injection did notDraft have higher vitamin B12 concentrations (543 ± 188

240 pmol/L; n=8) compared to those not receiving vitamin B12 injections (417 ± 209 pmol/L;

241 n=199; p=0.074).

242

243 There were no significant correlations between length of time spent in residential care

244 and EGRac (ρ=0.027, p=0.697) or vitamin B12 concentration (ρ=0.084, p=0.231). Of

245 participants not taking oral supplements (n=161), the riboflavin content of menus served

246 was significantly inversely associated with ERGac (ρ=-0.235, p=0.003). Of those

247 participants not receiving oral supplements or monthly vitamin B12 injections (n=155),

248 there was a weak but significant inverse correlation between dietary vitamin B12 from

249 menus served and total vitamin B12 concentrations (ρ=-0.170, p=0.034).

250

251 DISCUSSION

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 12 of 25

252 This exploratory secondary analysis study highlights that standardized residential care

253 menus contained adequate amounts of vitamin B12 and riboflavin (as reflected by a low

254 prevalence of menus providing less than the EAR), and that the older adults living in

255 residential care facilities in British Columbia generally had sufficient vitamin B12 status

256 despite low consumption of oral supplements and fortified foods. Interestingly, despite

257 adequacy in the menus served, 26% of participants were classified with riboflavin

258 deficiency.

259

260 The most recent Canadian Health Measures Survey (CHMS) data indicate that 79% of

261 adults aged 60-79 years (n=1,035) had sufficient serum total vitamin B12 concentrations

262 (>220 pmol/L) (Macfarlane et al. 2011),Draft which mirrors the results of this study, where

263 81% of participants were vitamin B12 sufficient.

264

265 It is well established that high folate intake can mask the hematologic markers of vitamin

266 B12 deficiency, allowing the progression of vitamin B12-dependent neuropathies (Crider

267 et al. 2011). This is a particular risk among older adults who may already be at a higher

268 risk of vitamin B12 deficiency due to atrophic gastritis (Institute of Medicine 1998a). We

269 assessed the folate content of menus (as dietary folate equivalents), and found that despite

270 mandatory folic acid fortification in Canada, 93% of menus did not provide the EAR for

271 folate of 320 μg/d (Institute of Medicine 1998c). Similarly, a recent cross-sectional

272 assessment of micronutrient adequacy in diets consumed by residents of 32 long-term

273 care facilities across Canada found median folate intakes below the EAR: 266.6 and

274 219.2 μg/day among men with a regular (n=127) and modified (n=70) diets, and 239.8

https://mc06.manuscriptcentral.com/apnm-pubs Page 13 of 25 Applied Physiology, Nutrition, and Metabolism

275 and 175.3 μg/day among women consuming a regular (n=297) and modified (n=138)

276 diets (Keller et al. 2018). In addition, folate intake was reported to be low among older

277 adults (>70 years) in the Canadian Community Health Survey Cycle 2.2 (CCHS), with

278 23.1% of men (n=1,520) and 47.0% of women (n=2,610) having intakes below the EAR

279 (Health Canada 2009). This is surprising given that of all age groups assessed in the

280 CHMS, adults aged 60-79 years had the highest median red blood cell folate

281 concentrations of 1409 nmol/L (Colapinto et al. 2011).

282

283 We found that 40% of participants had suboptimal or deficient biochemical riboflavin

284 status (EGRac ≥ 1.3). Little is known about the biochemical riboflavin status of older

285 Canadians, however our team recentlyDraft found a high prevalence (70%; n=34/49) of

286 EGRac ≥ 1.3 among a convenience sample of women of childbearing age (20-40 years)

287 in Vancouver, Canada (Whitfield et al. 2015). A similar trend in riboflavin status by age

288 was seen in the United Kingdom National Diet & Nutrition Survey, where EGRac

289 improved with age (Ruston et al. 2004). Among older adults (50-64 years) in the United

290 Kingdom, the prevalence of EGRac > 1.3 was 54% (men) and 50% (women), while

291 younger adults (19-24 years) had higher prevalences of 82% (men) and 77% (women)

292 (Ruston et al. 2004). In the United Kingdom this may be explained by lower dietary

293 riboflavin intake among younger adults (Ruston et al. 2004). Conversely, dietary

294 riboflavin intake in Canada differs from consumption patterns in the United Kingdom,

295 with data from the CCHS indicating that absolute riboflavin intake is higher among

296 younger people compared to older adults (Health Canada 2009). The mean riboflavin

297 intake for young adults (aged 19 – 30 years) was 2.41 mg/day (young men, n=1,804) and

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 14 of 25

298 1.72 mg/day (young women, n=1,854), compared to older adults, 1.86 mg/day (older

299 men, n=1,520) and 1.53 mg/day (older women, n=2,610). Of note, though, the prevalence

300 of inadequacy (%

301 young men, and 5.9% (men) and 4.4% (women) among adults >70 years (prevalence of

302 inadequacy for women 19-30 years was not reported due to high sample variability)

303 (Health Canada 2009).

304

305 The relatively high prevalence of suboptimal and deficient biochemical riboflavin status

306 despite the low prevalence of menu inadequacy may be attributed to the biomarker,

307 EGRac. Although EGRac is currently most commonly employed biomarker (McAuley et

308 al. 2016), there is some uncertainty aboutDraft the interpretation and clinical significance of

309 EGRac cut-offs. Ruston et al noted that EGRac is highly sensitive to even small levels of

310 glutathione reductase desaturation (Ruston et al. 2004), potentially because compared to

311 other enzymes involved with energy metabolism, glutathione reductase loses FAD at an

312 early stage of riboflavin deficiency (Ross and Hansen 1992). Others have noted that

313 EGRac values above the cut-off of 1.3 are not well aligned with physiological or

314 functional changes (Powers et al. 2011). However, there is potential for adverse

315 functional outcomes even with sub-clinical riboflavin deficiency (McAuley et al. 2016),

316 so the 40% prevalence of suboptimal or deficient status (EGRac ≥ 1.3) we report here

317 may warrant continued investigation of riboflavin status of Canadians. Of note, an expert

318 panel formed by the European Food Safety Authority recently stated that urinary

319 riboflavin excretion should be used to assess status, with EGRac as a supportive

https://mc06.manuscriptcentral.com/apnm-pubs Page 15 of 25 Applied Physiology, Nutrition, and Metabolism

320 secondary biomarker (EFSA NDA Panel on Dietetic Products et al. 2017), so any future

321 research should also assess urinary riboflavin.

322

323 The vitamin B12 available in menus (mean ± SD) was 3.9 ± 1.0 μg/day among men and

324 3.6 ± 1.1 μg/day among women in this study, which are similar to intakes reported among

325 adults >70 years in the CCHS (mean (SE) intake was 4.4 (0.3) μg/day among men and

326 3.5 (0.2) μg/day among women (Health Canada 2009)), and other Canadians living in

327 long-term care facilities (median intakes between 3.1 and 4.2 μg/day (Keller et al. 2018)).

328 These values are higher than free-living Italian women >60 years (n=286) who consumed

329 2.76 ± 2.13 μg vitamin B12 daily (Bolzetta et al. 2015) and free-living older adults in

330 Turkey who consumed between 2.6 Draft± 3.4 (women; n=185) and 3.5 ± 3.3 (men; n=184)

331 μg vitamin B12 daily (Keser et al. 2015). However, the importance of crystalline vitamin

332 B12 consumption, which is recommended over food-bound vitamin B12, cannot be

333 overlooked. Very few participants consumed a diet that included vitamin B12-fortified

334 foods (fortified soy beverage for the vegetarian diet only, n=3/207) despite the Institute of

335 Medicine recommendation that most vitamin B12 be sourced from fortified foods or

336 supplements for those over the age of 51 years (Institute of Medicine 1998a). Of the 9

337 participants whose menus provided vitamin B12 at levels below the EAR, 4 were

338 consuming oral micronutrient supplements, so their total intakes were likely adequate.

339 None were receiving monthly intramuscular vitamin B12 injections. However, it is

340 difficult to attribute vitamin B12 intake below the EAR to diet type or texture as there was

341 no pattern among low consumers, who had a mix of general, dysphagia, nutrient-dense,

342 lactose-restricted, and diabetic diets, and regular, easy to chew, and pureed diets.

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 16 of 25

343

344 The riboflavin content of menus served to men in this study was 2.0 ± 0.7 mg/day, which

345 is similar to the intake of men >70 years in the CCHS, 1.86 (0.05) mg/day (Health

346 Canada 2009) and long-term care facilities in Canada, (median 2.0 mg/day) (Keller et al.

347 2018). However, riboflavin in the menus served to women in this study (1.9 ± 0.8

348 mg/day) was higher than the riboflavin consumed by women >70 years in the CCHS,

349 1.53 (0.04) mg/day (Health Canada 2009), and consumed among free-living Italian

350 women, 1.19 ± 0.35 mg/day (Bolzetta et al. 2015) and women in Canadian long-term care

351 (1.6 to 1.7 mg/day) (Keller et al. 2018). While it is difficult to make direct comparisons

352 between the CCHS and our study, since we assessed the nutrient content of menus served

353 rather than actual dietary intake, thisDraft trend contrasts a recent assessment of nutrient

354 intakes among four groups of Finish older adults - healthy, diseased, home-dwelling, and

355 in assisted care facilities. The Finish study reported that malnutrition was highest, and

356 micronutrient intake lowest, among those living in residential care, however, this study

357 assessed only select nutrients and neither vitamin B12 nor riboflavin were included in

358 these analyses (Jyvakorpi et al. 2015).

359

360 The major strength of this study was the ability to compare biomarker data to dietary

361 nutrient availability and supplement use for vitamin B12 and riboflavin. Also, despite the

362 growing number of Canadians expected to move into residential care facilities, this is an

363 understudied population. Of note, both the CCHS and CHMS include only free-living

364 Canadians (not in institutions). However as an exploratory secondary analysis, the study

365 does have a few limitations. Dietary data were based on nutrient content of menus served

https://mc06.manuscriptcentral.com/apnm-pubs Page 17 of 25 Applied Physiology, Nutrition, and Metabolism

366 not food consumed, and therefore may overestimate intake. Also, guests of individuals

367 residing in these facilities are free to bring food and drink, so there is also potential for

368 underestimating intake. For those using supplements, nutrient intake provided by

369 supplements was not quantified. Another limitation of our study was that participants had

370 an average of 4-5 co-diagnoses (with an upper range of 15), and although medications

371 were not part of this chart review, we expect polypharmacy was also common (Andrew et

372 al. 2017), which could influence nutritional status (Jyrkkä et al. 2012). Although we were

373 able to assess the folate content of menus, the lack of a matching folate biochemical

374 marker is a limitation. Participation among those with dementia was about 10% lower

375 than those who did not consent to participate (Feldman et al. 2014). As, such, there may

376 be a small systematic bias in our sampleDraft favoring those without dementia. Finally, the

377 recommendation to assess riboflavin status using both EGRac and urinary riboflavin

378 excretion (EFSA NDA Panel on Dietetic Products et al. 2017) came after this study was

379 completed, so urinary riboflavin is not reported here.

380

381 To conclude, the present study showed that there was a low prevalence of menu

382 inadequacy (%

383 content of menus served. Despite mandatory folic acid fortification in Canada, 93% of

384 menus served contained less than the folate EAR. There was a very low level of

385 biochemical vitamin B12 deficiency of only 5%. However, 26% of participants were

386 riboflavin deficient, which warrants continued investigation. Based on these results, older

387 adults living in Canadian residential care facilities are unlikely to require

388 supplementation of these B vitamins.

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 18 of 25

389

390 ACKNOWLEDGEMENTS

391 The UBC Vitamin Research Fund funded this study. K.C.W received support from the

392 Canadian Institutes of Health Research Frederick Banting and Charles Best Canada

393 Graduate Scholarships Doctoral Award. We would like to acknowledge the work of the

394 following registered dietitians at the Fraser Health Authority: Janine Seto, Fiona Huynh,

395 D’Arcy McDay, and Tavia Moffitt.

396 397 398 CONFLICT OF INTEREST 399 400 The authors have no conflicts of interest to report. Draft

https://mc06.manuscriptcentral.com/apnm-pubs Page 19 of 25 Applied Physiology, Nutrition, and Metabolism

401 REFERENCES 402 Andrew, M.K., Purcell, C.A., Marshall, E.G., Varatharasan, N., Clarke, B., and Bowles, 403 S.K. 2017. Polypharmacy and use of potentially inappropriate medications in long- 404 term care facilities: does coordinated primary care make a difference? Int. J. Pharm. 405 Pract. ePub. doi: 10.1111/ijpp.12397. 406 Bales, C.W., and Johnson, M.A. 2014. Nutrition in older adults. In Modern Nutrition in 407 Health and Disease, 11th edition. Edited by A.C. Ross, B. Caballero, R.J. Cousins, 408 K.L. Tuckets, and T.R. Ziegler. Lippincott Williams & Wilkins, Baltimore, MD. pp. 409 744–56. 410 Bates, B., Lennox, A., Prentice, A., Bates, C., Page, P., Nicholson, S., et al. 2014. United 411 Kingdom National Diet and Nutrition Survey Results from Years 1, 2, 3 and 4 412 (combined) of the Rolling Programme (2008/2009 - 2011/2012). London, UK. 413 Bolzetta, F., Veronese, N., De Rui, M., Berton, L., Toffanello, E.D., Carraro, S., et al. 414 2015. Are the Recommended Dietary Allowances for Vitamins Appropriate for 415 Elderly People? J. Acad. Nutr. Diet. 115(11): 1789–1797. Elsevier Inc. doi: 416 10.1016/j.jand.2015.04.013. 417 Colapinto, C.K., O’Connor, D.L., and Tremblay, M.S. 2011. Folate status of the 418 population in the Canadian Health Measures Survey. Can. Med. Assoc. J. 183(2): 419 e100-06. doi: 10.1503/cmaj.100568.Draft 420 Crider, K.S., Bailey, L.B., and Berry, R.J. 2011. Folic acid food fortification-its history, 421 effect, concerns, and future directions. Nutrients 3: 370–384. doi: 422 10.3390/nu3030370. 423 Donini, L.M., Poggiogalle, E., Piredda, M., Pinto, A., Barbagallo, M., Cucinotta, D., et 424 al. 2013. Anorexia and eating patterns in the elderly. PLoS One 8(5): e63539. doi: 425 10.1371/journal.pone.0063539. 426 EFSA NDA Panel on Dietetic Products, Turck, D., Bresson, J., Burlingame, B., Dean, T., 427 Fairweather‐Tait, S., et al. 2017. Dietary Reference Values for riboflavin. EFSA J. 428 15(8): 1–65. doi: 10.2903/j.efsa.2017.4919. 429 Feldman, F., Moore, C., da Silva, L., Gaspard, G., Gustafson, L., Singh, S., et al. 2014. 430 Effectiveness and safety of a high-dose weekly vitamin D (20,000 IU) protocol in 431 older adults living in residential care. J. Am. Geriatr. Soc. 82(8): 1546–1550. doi: 432 10.1111/jgs.12927. 433 Garner, R., Tanuseputro, P., Manuel, D.G., and Sanmartin, C. 2018. Transitions to long- 434 term and residential care among older Canadians. Heal. Reports 29(5): 13–23 435 Statistics Canada Catalogue no. 82–003–X. 436 Health Canada. 2009. Canadian Community Health Survey Cycle 2.2, Nutrition (2004) - 437 Nutrient Intakes from Food, Provincial, Regional and National Summary Data 438 Tables, Volume 2. Ottawa, Canada. 439 Health Canada. 2011. Eating well with Canada’s food guide. Publication 4651. 440 Horigan, G., McNulty, H., Ward, M., Strain, J.J., Purvis, J., and Scott, J.M. 2010. 441 Riboflavin lowers blood pressure in cardiovascular disease patients homozygous for

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 20 of 25

442 the 677C ->T polymorphism in MTHFR. J. Hypertens. 28: 478–486. doi: 443 10.1097/hjh.0b013e328334c126. 444 Institute of Medicine. 1998a. Vitamin B12. In Dietary Reference Intakes for Thiamin, 445 Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and 446 Choline. National Academy Press, Washington, DC. pp. 306–356. 447 Institute of Medicine. 1998b. Riboflavin. In Dietary Reference Intakes for Thiamin, 448 Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and 449 Choline. National Academies Press, Washington, DC. pp. 87–122. 450 Institute of Medicine. 1998c. Folate. In Dietary Reference Intakes for Thiamin, 451 Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and 452 Choline. National Academy Press, Washington, DC. pp. 196–305. 453 Jyrkkä, J., Mursu, J., Enlund, H., and Lönnroos, E. 2012. Polypharmacy and nutritional 454 status in elderly people. Curr. Opin. Clin. Nutr. Metab. Care 15: 1–6. doi: 455 10.1097/mco.0b013e32834d155a. 456 Jyvakorpi, S.K., Pitkala, K.H., Puranen, T.M., Bjorkman, M.P., Kautiainen, H., 457 Strandberg, T.E., et al. 2015. Low protein and micronutrient intakes in 458 heterogeneous older population samples. Arch. Gerontol. Geriatr. 61(3): 464–471. 459 Elsevier Ireland Ltd. doi: 10.1016/j.archger.2015.06.022. 460 Keller, H.H. 2007. Promoting food intakeDraft in older adults living in the community: a 461 review. Appl. Physiol. Nutr. Metab. 32(6): 991–1000. doi: 10.1139/h07-067. 462 Keller, H.H., Lengyel, C., Carrier, N., Slaughter, S.E., Morrison, J., Duncan, A.M., et al. 463 2018. Prevalence of inadequate micronutrient intakes of Canadian long-term care 464 residents. Br. J. Nutr. 119(9): 1047–56. doi: 10.1017/S0007114518000107. 465 Keser, A., Yabanci Ayhan, N., Bilgic, P., Tayfur, M., and Slmsek, I. 2015. Determination 466 of dietary status as a risk factor of cardiovascular heart disease in Turkish elderly 467 people. Ecol. Food Nutr. 54(4): 328–41. doi: 10.1080/03670244.2014.992520. 468 Macfarlane, A.J., Greene-Finestone, L.S., and Shi, Y. 2011. Vitamin B-12 and 469 homocysteine status in a folate-replete population: results from the Canadian Health 470 Measures Survey. Am. J. Clin. Nutr. 94: 1079–87. doi: 10.3945/ajcn.111.020230.1. 471 McAuley, E., McNulty, H., Hughes, C., Strain, J.J., and Ward, M. 2016. Riboflavin 472 status, MTHFR genotype and blood pressure: current evidence and implications for 473 personalised nutrition. Proc. Nutr. Soc. 75(3): 405–414. doi: 474 10.1017/S0029665116000197. 475 McNulty, H., and Scott, J.M. 2008. Intake and status of folate and related B-vitamins: 476 considerations and challenges in achieving optimal status. Br. J. Nutr. 99(Suppl 3): 477 S48–S54. doi: 10.1017/S0007114508006855. 478 Minister of Industry Statistics Canada. 2011. Residential Care Facilities 2009/2010. 479 Ottawa, Canada. 480 National Population Projections Team, Bohnert, N., Chagnon, J., and Dion, P. 2015. 481 Population Projections for Canada (2013 to 2063), Provinces and Territories (2013 482 to 2038). Ottawa, Canada.

https://mc06.manuscriptcentral.com/apnm-pubs Page 21 of 25 Applied Physiology, Nutrition, and Metabolism

483 Otten, J.J., Pitzi Hellwig, J., and Meyers, L.D. 2006. Dietary reference intakes: the 484 essential guide to nutrient requirements. In Institute of Medicine. Washington, DC, 485 USA. 486 Powers, H.J., Hill, M.H., Mushtaq, S., Dainty, J.R., Majsak-Newman, G., and Williams, 487 E.A. 2011. Correcting a marginal riboflavin deficiency improves hematologic status 488 in young women in the United Kingdom (RIBOFEM). Am. J. Clin. Nutr. 93: 1274– 489 84. doi: 10.3945/ajcn.110.008409.1. 490 Ross, N.S., and Hansen, T.P. 1992. Riboflavin deficiency is associated with selective 491 preservation of critical flavoenzyme-dependent metabolic pathways. Biofactors 3(3): 492 195–190. 493 Ruston, D., Hoare, J., Henderson, L., Gregory, J., Bates, C.J., Prentice, A., et al. 2004. 494 The National Diet & Nutrition Survey: adults aged 19 to 64 years Nutritional status 495 (anthropometry and blood analytes), blood pressure, and physical activity. London, 496 UK. 497 Said, H.M., and Ross, A.C. 2014. Riboflavin. In Modern Nutrition in Health and Disease, 498 11th edition. Edited by A.C. Ross, B. Calallero, R.J. Cousins, K.L. Tucker, and T.R. 499 Ziegler. Lippincott Williams & Wilkins, Baltimore, MD. pp. 325–330. 500 Sekerak, R.J., and Stewart, J.T. 2014. Caring for the patient with end-stage dementia. 501 Ann. Long-Term Care Clin. CareDraft Aging 22(12): 36–43. 502 Whitfield, K.C., Karakochuk, C.D., Liu, Y., McCann, A., Talukder, A., Kroeun, H., et al. 503 2015. Poor thiamin and riboflavin status is common among women of childbearing 504 age in rural and urban Cambodia. J. Nutr. 145: 628–33. doi: 505 10.3945/jn.114.203604.study. 506 Wilson, C.P., Ward, M., McNulty, H., Strain, J.J., Trouton, T.G., Horigan, G., et al. 2012. 507 Riboflavin offers a targeted strategy for managing hypertension in patients with the 508 MTHFR 677TT genotype: a 4-y follow-up. Am. J. Clin. Nutr. 95: 766–772. doi: 509 10.3945/ajcn.111.026245.1.

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 22 of 25

511 TABLES 512 513 Table 1: Demographic characteristicsa All Men Women n=207 n=53 n=154 p value Age, yearsb 85.7 ± 6.7 83.3 ± 7.2 86.5 ± 6.3 0.003 Ethnicityc European descent 191 (92%) 49 (93%) 142 (92%) South Asian 7 (3%) 2 (4%) 5 (3%) 0.391 Chinese or Southeast Asian 5 (2%) - 5 (3%) Other 4 (2%) 2 (4%) 2 (1%) Smokerc 10 (5%) 4 (8%) 6 (4%) 0.285 BMI, kg/m2b,d 25.0 ± 5.4 25.6 ± 5.0 24.8 ± 5.6 0.239 Underweight (≤ 18.5)c 23 (12%) 3 (6%) 20 (14%) Normal (18.51-24.99)c 85 (43%) 24 (45%) 61 (42%) 0.446 Overweight (25-29.99) c 56 (28%) 17 (32%) 39 (27%) Obese (≥ 30) c 33 (17%) 9 (17%) 24 (17%) Number of Health Diagnosesb 4.6 ± 2.5 4.5 ± 2.7 4.6 ± 2.5 0.078 Range 0 – 15 0 – 12 0 – 15 514 a data expressed as mean ± SD or n (%) 515 b differences by sex were assessed usingDraft an independent t-test 516 c differences by sex were assessed using x2 test 517 d BMI data available for only n=197 (n=10 women missing data)

https://mc06.manuscriptcentral.com/apnm-pubs Page 23 of 25 Applied Physiology, Nutrition, and Metabolism

a 518 Table 2: Biochemical vitamin B12 and riboflavin status All Men Women Oral No oral supplement supplement n=207 n=53 n=154 p value n=46 n=161 p value b Serum vitamin B12, pmol/L 422 ± 209 390 ± 204 433 ± 210 0.190 474 ± 186 407 ± 213 0.024 Deficient (<148)c 11 (5%) 5 (9%) 6 (4%) 1 (2%) 10 (6%) Marginal (≥148 to 220)c 28 (14%) 9 (17%) 19 (12%) 0.945 3 (7%) 25 (16%) 0.136 Sufficient (≥221)c 168 (81%) 39 (74%) 129 (84%) 42 (91%) 126 (78%) EGRac, ratiob 1.30 ± 0.19 1.27 ± 0.14 1.31 ± 0.21 0.763 1.20 ± 0.15 1.33 ± 0.20 <0.001 Deficient (≥ 1.4)c 54 (26%) 13 (25%) 41 (~27%) 6 (13%) 48 (30%) Suboptimal (≤1.3 to < 1.4)c 28 (14%) 7 (13%) 21 (~14%) 0.182 3 (7%) 25 (15%) 0.007 Sufficient (< 1.3)c 125 (60%) 33 (62%) 92 (60%) 37 (80%) 88 (55%) 519 a data expressed as mean ± SD or n (%) 520 b differences by sex and supplement use were assessed usingDraft a Mann Whitney U test 521 c differences by sex and supplement use were assessed using a x2 test

https://mc06.manuscriptcentral.com/apnm-pubs Applied Physiology, Nutrition, and Metabolism Page 24 of 25

522 Table 3: Menu classifications, nutrient composition of menus, and micronutrient 523 supplement use of participantsa All Men Women n=207 n=53 n=154 p value Diet typeb General 112 (54%) 29 (55%) 83 (54%) Diabetic 36 (18%) 12 (23%) 24 (16%) Dysphagia (thin or thick) 21 (10%) 6 (11%) 15 (10%) 0.170 Nutrient dense 17 (8%) 1 (2%) 16 (10%) Other c,d 21 (10%) 5 (9%) 16 (10%) Diet Textureb Regular 112 (54%) 28 (53%) 84 (55%) Easy to chew 42 (20%) 12 (23%) 30 (20%) 0.905 Cut-up 16 (8%) 3 (6%) 13 (8%) Minced 19 (9%) 5 (9%) 14 (9%) Pureed 16 (8%) 5 (9%) 11 (7%) Full fluids 2 (1%) - 2 (1%) Allergyb 18 (9%) 4 (8%) 14 (9%) 0.731 Menu total energy, kcal/de 1792 ± 409 1892 ± 394 1757 ± 409 0.013 Menu carbohydrate, g/de 238 ± 57 251 ± 56 223 ± 57 0.030 Menu protein, g/de Draft74 ± 17 80 ± 20 72 ± 16 0.006 Menu fat, g/de 61 ± 16 64 ± 16 60 ± 16 0.101 c,e Menu vitamin B12, μg/d 3.7 ± 1.1 3.9 ± 1.0 3.6 ± 1.1 0.026 % < EARb,f 9 (4%) 1 (2%) 8 (5%) 0.308 Menu riboflavin, mg/de 1.9 ± 0.7 2.0 ± 0.7 1.9 ± 0.8 0.135 % < EARb,g 3 (1%) 1 (2%) 2 (1%) 0.757 Menu folate (DFE), μg/de 220 ± 61 238 ± 64 214 ± 59 0.009 % < EARb,h 193 (93%) 48 (91%) 145 (94%) 0.369 Supplement Useb Oral micronutrient supplement 46 (22%) 14 (26%) 32 (21%) 0.395 i Monthly vitamin B12 injection 8 (4%) 2 (4%) 6 (4%) 0.968 524 DFE, dietary folate equivalents; EAR, estimated average requirement 525 a data expressed as mean ± SD or n (%) 526 b differences by sex were assessed using a x2 test 527 c other diets include: for men, n=1 each of high fibre, healthy heart, potassium restricted, 528 renal failure (no dialysis), and sodium restricted; for women, n=6 sodium reduced, n=4 529 high fibre, n=2 lactose restricted, n=2 lacto-ovo vegetarian, n=1 lacto-vegetarian, and 530 n=1 gluten-free. 531 d only 3 individuals (n=2 women lacto-ovo vegetarian, n=1 woman lacto-vegetarian) 532 consumed vitamin B12-fortified foods in the form of a fortified soy beverage. 533 e differences by sex were assessed using a Mann Whitney U test f 534 vitamin B12 EAR is 2.0 μg/d for both men and women (Institute of Medicine 1998a). 535 g riboflavin EAR is 1.1 mg/day for men and 0.9 mg/day for women (Institute of Medicine 536 1998b). 537 h folate EAR is 320 μg/d for both men and women (Institute of Medicine 1998c).

https://mc06.manuscriptcentral.com/apnm-pubs Page 25 of 25 Applied Physiology, Nutrition, and Metabolism

538 i n=2 (1 man, 1 woman) took both an oral micronutrient supplement and received a 539 monthly vitamin B12 injection

Draft

https://mc06.manuscriptcentral.com/apnm-pubs