Full text of DOI:10.1017/S0714980800004487

Il y a peu de renseignements sur les caractéristiques des personnes âgées qui utilisent les compléments. Nous avons
recruté un échantillon de commodité de personnes âgées (n = 128) vivant à l’intérieur de dix sites communautaires afin
d’identifier les facteurs associés à l’utilisation de compléments (vitamines-minéraux et / ou préparations à base d’her-
bes). Les données des variables nutritives, des antécédents de santé et de l’utilisation de compléments, ont été recueil-
lies par l’entremise d’un questionnaire-entrevue. Nous avons développé une échelle de propension à l’utilisation des
compléments qui a été associée à la majoration de l’utilisation de compléments à base d’herbes. Au niveau de l’analyse
bivariée, très peu de caractéristiques séparent les utilisateurs des non-utilisateurs.Toutefois, l’analyse multifactorielle
révèle que l’échelle de propension et un indice de masse corporelle inférieur sont des facteurs reliés au nombre total de
compléments utilisés. Lorsque l’on fait l’analyse des facteurs individuels de l’échelle au niveau bivarié, la disponibilité
alimentaire, les croyances au sujet de la santé et le rôle davantage proactif à sa propre santé correspondent à l’augmen-
tation de l’utilisation de compléments. Les premiers résultats suggèrent que les attitudes des personnes âgées envers
les compléments peuvent contribuer à leur décision d’ajouter ces derniers à leur alimentation. Des recherches
ultérieures sont nécessaires au sein d’une population canadienne de personnes âgées, choisi au hasard et diversifiée.
Information on the characteristics of seniors who use supplements is lacking. A convenience sample of community-liv-
ing seniors (n = 128) was recruited from 10 sites to identify factors associated with supplement use (vitamin–mineral
and/or herbal preparations). Data on nutritional variables, health history, and supplement use were collected with an
interview-administered questionnaire. A Propensity to Use Supplements scale was developed and was associated with
increased use of herbal supplements. At the bivariate level, few characteristics distinguished users and non-users.
Multivariate modelling revealed the Propensity scale and a low Body Mass Index to be factors related to the total num-
ber of supplements used. When individual scale items were examined at the bivariate level, food supply concerns,
health beliefs, and a proactive role in one’s own health were associated with increased supplement use. Initial results
suggest seniors’ attitudes towards supplements may be an important factor contributing to their decision to use sup-
plements. Further work is required in a more diverse, randomly selected, older Canadian population.
*
The authors wish to thank Richard Goy for his statistical advice, and Aimee Krupansky, Lynn Boudreau, Pat Vanderkooy, and
Desirée De Waal for their help as research assistants. Without the dedication of these individuals, this project would not have
succeeded. The assistance of the recruitment site management and the seniors themselves is also greatly appreciated.
Manuscript received: / manuscrit reçu : 20/03/01
Manuscript accepted: / manuscrit accepté : 20/01/03
Mots clés : vieillissement, vitamines-minéraux, préparations à base d’herbes, personnes âgées, compléments, phytothérapie
Key words: aging, vitamin–minerals, herbal preparations, older adults, supplements, alternative-drug use
Requests for offprints should be sent to: / Les demandes de tirés-a-part doivent être adressé e s à:
Heather H. Keller, R.D., Ph.D.
Associate Professor
Department of Family Relations and Applied Nutrition
University of Guelph
Guelph, ONN1G 2W1
(hkeller@uoguelph.ca)

use) and the number of vitamin–mineral and/or
herbal preparations used. It is hypothesized that those
seniors who use several supplements will be different
from those who use few or none.
Use of self-prescribed vitamin–mineral and herbal
preparations among the older, community-dwelling
population is increasing; vitamin–mineral use ranges
from 26 to 80 per cent in this age group (Gray, Paganini-
Hill, & Ross, 1983; Betts & Rezek, 1989; Payette & Gray-
Donald, 1991; Bender, Levy, Schucker, & Yetley, 1992;
Gray, Hanlon, Fillenbaum, Wall, & Bales, 1996; McKen-
zie & Keller, 2001) and herbal use is exploding (Angus
Reid, 1997; Radimer, Subar, & Thompson, 2000; Eisen-
berg et al., 1998). Research on the characteristics of
older adults in the United States who use vitamin–min-
eral preparations is extensive, yet inconclusive. In com-
parison, research on the use of complementary health
care, such as herbal remedies, is in its infancy (Verhoef,
Russell, & Love, 1994; Millar, 1997; Montbriand, 2000;
Fries & Menzies, 2000).
This study was a subsection of a larger study, involving
the validation of a nutrition-risk screening tool
(SCREEN© – Seniors In the Community Risk Evalua-
tion for Eating and Nutrition) (Keller, McKenzie, &
Goy, 2001). Individuals participating in the SCREEN
study comprised the sample for this survey (Keller et
al.). Ten different sites were involved in the project,
includinga dayhospitaland rehabilitation centre, three
seniors’ apartment buildings, four supportive housing
units, and a seniors’ recreation centre. Participants
were recruited by supportive housing workers, mail, or
advertisements, or during nutrition information semi-
nars. Eligible individuals were those aged 55 years and
over, living in the community, who consented to a com-
prehensive nutritional assessment. A sample of 137
older adults was recruited and 128 completed the
study.
Those who have specifically examined the characteris-
tics of older (> 65 years of age) vitamin–mineral users
have not found significant associations between use
and age, gender, marital status (Ranno, Wardlaw, &
Gieger, 1988; Oakland & Thomsen, 1990), living
arrangements (Oakland & Thomsen), socio-economic
status (Oakland & Thomsen; Ranno et al.), or health
characteristics (Ranno et al.; Oakland & Thomsen;
Stoehr, Gaguli, Seaberg, Echement, & Belle, 1997).
Canadian research on the characteristics of supple-
ment users in the senior population is limited. While
factors such as health, exercise, education, and income
may be noted in these studies, they have not been
identified as factors associated with use (Donald,
Tapan, Hargreaves, Thompson, Overton, Peterson, &
Chao, 1992; Payette & Gray-Donald, 1991; Blais,
Maiga, & Aboubacar, 1997; Radimer et al., 2000). Ulti-
mately, it remains unclear which older adults are more
likely to use vitamin–mineral and/or herbal prepara-
tions. With a national health care system, it is antici-
pated that factors associated with use by Canadian
seniors may be different from those observed in the
American population.
As there has been little research on supplement use by
seniors, three focus groups were held, to determine
current supplement issues among community-living
older adults. These focus groups helped to identify
the important issues for inclusion in a structured,
interviewer-administered questionnaire. Focus group
questions were reviewed by nine dietitians for content
validity and included current and past supplement
use, purchase point, source of information, money
spent on supplements, and reasons for use and non-
use. Eight themes emerged from the focus group dis-
cussions, which led to the development of a 16-item
Propensity to Use Supplements scale (Figure 1). These
themes were food supply concerns, health beliefs, a
perceived need for supplements, cost, a preference for
“natural” treatments, confusion about what to use, a
distrust of Western medicine, and a proactive role in
one’s own health. All eight themes were addressed at
some point by the statements in the 16-item question-
naire, but the number of statements addressing a par-
ticular theme was based on the emphasis placed on
that theme during the focus group discussions. This
score was an attempt to try to reflect participant atti-
tudes toward supplement use; a higher score would
indicate a more positive attitude towards supplement
use. To ensure content validity and understandability
of items, the Propensity to Use Supplements questions
were reviewed by nine dietitians and piloted with 11
seniors before actual use. Items were then reworded
as necessary.
It is suspected that seniors who use a few or no sup-
plements may be quite different in their attitudes
towards supplements from those who use many. As
supplements can react with each other and with tradi-
tional medications, it is important to be able to iden-
tify seniors who are potentially heavy users of
supplements. Additionally, research needs to explore
other covariates, outside of basic demographics, to
determine the profile of these seniors. This study is an
initial step towards understanding and characterizing
these users. The purpose of this study was to examine
the independent association among a wide variety of
specific participant factors (e.g., perceived health sta-
tus, nutritional status, attitudes towards supplement

*
p < 0.05 in bivariate comparison with total supplement use
Questions were answered agree, disagree, or no opinion
a
The interviewer-administered supplement-use ques-
tionnaire was included in a medical history interview.
Because of the larger study’s purpose, increasing use
of supplements could be compared to other variables
not typically collected in survey research. These varia-
bles included a detailed medical and nutritional his-
tory, anthropometric measurements, a diet history (7-
day), and a dietitian’s rating of nutritional status. The
medical history included such factors as number and
type of medications, surgeries, history of chronic dis-
ease, and other reported health problems. The nutri-
tional history allowed for the examination of many of
the risk factors associated with poor dietary intake,
including special diets, dentition, use of nutritional
supplements (e.g., Ensure), and grocery shopping/
cooking independence. Knee height, weight and
weight history, skin folds (subscapular, triceps), and
circumference of the wrist, calf, and arm were the
anthropometric measurements taken (Gibson, 1990).
A diet history was completed by a trained dietitian at
a separate appointment (Keller et al., 2001). The rating
of nutritional status was based on the clinical judge-
ment of the dietitian who completed the diet history
and reviewed the medical/nutritional information.
Further detail on the gathering of these data has been
reported elsewhere (Keller et al.).
internal reliability analysis of the Propensity to Use
Supplements scale. Results produced a low Cron-
bach’s alpha (α = 0.30), indicating poor internal relia-
bility. Removing items based on low, corrected, item–
total correlations (r < 0.20) did not produce a more
desirable α; for the remaining five items, α = 0.34 (a, e,
j, l, o; see Figure 1) (Streiner & Norman, 1995).
Multiple linear regression modelling was used to
determine if participant characteristics were associ-
ated with number of total supplements used (vita-
min–mineral and/or herbal preparations). The entire
Propensity to Use Supplements scale was used in this
analysis, despite the low Cronbach’s alpha. Addition-
ally, a separate regression analysis was completed to
determine those items on the Propensity to Use Sup-
plements scale that best predicted increasing use of
total supplements, as well as a bivariate analysis to
determine individual items associated with use.
A conceptual model guided the process of initial vari-
able selection for the full model to be assessed (Figure
2) (Kleinbaum, 1994). Sixteen variables were mod-
elled, based on information from the literature and
focus group discussions. Collinearity in the variables
was not evident (tolerance levels > 0.1). A generous p
value (p < 0.25) was used to ensure that all conceptual
model variables modestly associated with the out-
come were included in the initial full model (Klein-
baum). This strategy was used to reduce potential
covariates, as the sample size was modest (Klein-
baum). A total of eight variables from the conceptual
model fell below the designated significance level
(p < 0.25) using Pearson correlation and independent
SPSS (version 7.5) was used for all statistical analyses.
Bivariate associations were examined between varia-
bles of interest and use of vitamin–mineral and/or
herbal preparations, using t-test, χ2, and ANOVA.
Cronbach’s alpha (α) was also used, to perform an

t-tests (Figure 2). Included in these eight variables was
the Propensity to Use Supplements score. The initial
8-variable full model had an R2 of 0.34 (F = 7.8;
p < 0.01). Standard regression was then used to reduce
the initial full model in a sequential fashion, removing
the nuisance variables in a step-wise approach based
on their significance level, to achieve the final model
(Kleinbaum). A significant change in R2 was used to
identify when a sequential selection of these variables
should be stopped to maintain model stability.
A total of 128 community-living seniors participated
in the study (73.4% female). The average age of the
sample was 76.1 years (SD = 9.1, range: 50–94 yrs).
“Widowed” was the most common marital status
(43.7%) and 64.1 per cent of the participants lived
alone. Two-thirds (66.7%) had obtained a high school
diploma, and of these high school graduates, 46 per
cent had completed some type of post-secondary pro-
gram. Smokers were not common among the partici-
pants, with only 7 per cent currently smoking at the
time of interview (Table 1). The majority reported a lit-
tle (36%) or a great deal (29%) of difficulty in complet-
ing every day tasks, due to their health, although 83
per cent believed their health was good to excellent
compared to that of others their own age. Subjects
reported an average of 5.2 medical problems
(SD = 2.8) and took an average of 4.3 prescription
medications (SD = 3.5) and 1 over-the-counter medi-
cation (SD = 1.1) (Table 2).
→
→
The majority of subjects (71%) self-rated their appetite
as normal to good, and 29 per cent were following a
special diet, with diabetic, low salt, low fat, vegetar-
ian, and renal diets reported most frequently. Partici-
pants typically spent less than CAN$50 per week on
groceries, with the average reported expenditure
between $20 and $30 per person per week. The major-
ity of participants (60.9%) had maintained their
weight in the past year. Using Body Mass Index (BMI)
(Health and Welfare Canada, 1988) categories, the
majority of participants were within the ideal body
weight for height category (BMI = 20–25; 37.5%) or
were above this ideal range (BMI _ 27; 36.7%). The
dietitians’ rating of nutrition risk averaged 4.8
(SD = 2.4) on a 10-point risk scale (1 = low risk,
10 = high risk), yet almost half of the participants
(47.7%) were classified in the moderate (rating 5–7) or
high (rating 8–10) nutrition risk categories.
More than three quarters of the participants (80%)
used one or more vitamin–mineral and/or herbal
preparation(s). Total supplement use averaged 2.4
(SD = 2.4, range: 1–13).
Individually, 72% were users of vitamin–minerals and
46% used herbal preparations. With a total of 17 dif-
ferent vitamin–mineral preparations reported, the
most common were vitamin E, multivitamin–miner-
als, vitamin C, and calcium. The average number of
individual vitamin–mineral preparations taken was
2.3 (SD = 1.6), ranging from 1 through 8.
*
p < 0.25 was used to define the initial full model; origi-
nal list of potential covariates taken from conceptual
model

**(p < 0.01) between users and non-users
a
Vitamin–mineral users
b
Education information is based on n = 93
*
(p < 0.05 )between users and non-users; **(p < 0.01) between users and non-users
Vitamin–mineral users
a
b
c
OTC = Over the Counter
Range 2–14, maximum score possible: 16

Of the 26 herbs reported, herbal teas, garlic pills, Echi-
nacea, and cod liver oil were the most prevalent
(McKenzie & Keller, 2001). Herbal teas were classified
under herbal preparations, based on their ingredient
list (e.g., chamomile, dandelion, rosehip), as was done
previously in the NHANES III investigation (Radimer
et al., 2000). On average, participants used 1.7 herbal
preparations (SD = 1.13, range: 1–6).
cance values: age (-0.16, p = 0.08); self-rated health
(5.1, p = 0.03); number of medications (-0.19, p = 0.04);
weight change (2.1, p = 0.15); dietitian rating of nutri-
tion risk (-0.12, p = 0.19); BMI (-0.16, p = 0.08); money
spent on groceries (5.4, p = 0.02); and Propensity to
Use Supplements Scale (0.54 p = 0.00).
Multivariate statistics produced a final significant
model, which revealed two factors to be independ-
ently and significantly associated with the increasing
use of supplements (F = 22.6, p < 0.01, R2 = 0.34); a
higher Propensity to Use Supplements score (β = 0.54)
and a lower BMI (β = -0.15). With removal of each
non-significant covariate individually, the final
reduced model contained only these two variables
and retained an R2 = 0.34. Further examination of the
16 items within the Propensity to Use Supplements
scale with regression analysis identified only 1 item,
“I have no one to talk to about whether I should use
supplements or herbal remedies”, to be predictive of
increasing supplement use. However, bivariate analy-
ses revealed four questions to be significantly associ-
ated with increasing supplement use (Figure 1). These
items reflected three of the eight themes found within
the scale and included food supply safety concerns,
health beliefs, and a proactive role in one’s own
health.
Characteristics of the users varied, depending on the
supplement category (vitamin–mineral or herbal
preparation), but few were found to be significant in
the bivariate analyses (Tables 1 and 2). Vitamin–min-
eral users had a significantly lower BMI (χ2 = 11.0,
p = 0.01) than non-users. Herbal remedy users were
significantly younger (t = 2.3; p < 0.05) were more
likely to be married (χ2 = 13.4, p < 0.01), and were
likely to weigh a lower percentage of ideal body
weight (current weight/ideal weight x 100) (F = 7.7,
p < 0.01). Participants who used herbal remedies had
a higher Propensity to Use Supplements score than
non-users of these remedies (F = 4.0, p < 0.05), unlike
vitamin–mineral users, whose score was higher, yet
not significantly different.
During model development, a total of eight variables
from the conceptual model (Figure 2) fell below the
designated significance level (p < 0.25). The following
are the variables, their coefficients, and their signifi-
β
β
*
p < 0.05; ** p < 0.01
a
collinearity was not evident

may suggest that seniors who are more “health con-
scious” in terms of maintaining an ideal body weight
also are more likely to use herbal preparations. While,
in the present study, users of herbal preparations were
at a lower percentage IBW than non-users, generaliza-
tions should be made with caution. In the present
study, further examination of other weight variables,
such as weight change and BMI, revealed no differ-
ences between users and non-users. Contrary to pop-
ular beliefs, the majority of herbal remedy users are
not “naturalists” who are against synthetic chemicals
or are dissatisfied with traditional medicine but are
people who incorporate herbal therapies into their
total health care. This seems consistent with the find-
ings of Foster, Philips, Hamel, & Eisenberg (2000),
who found that of the users of alternative therapies
aged 65 years and over, 95% had seen a physician in
the past year.
Previous work has produced conflicting results when
describing seniors who use vitamin–mineral prepara-
tions. Bivariate analysis in the present study found
that most demographic, health, and socio-economic
characteristics were not useful in differentiating vita-
min–mineral supplement use from non-use (Tables 1
and 2). BMI was the exception: vitamin–mineral users
had the highest proportion of those within the BMI of
20 to 25, whereas non-users were more likely to be
overweight or underweight.
Previous work has also found a relationship between
BMI and vitamin–mineral use. Data from a popula-
tion-based cohort study of 2,152 middle- to older-age
American adults found that supplement use was
more prevalent among individuals with a lower BMI
(Lyle, Mares-Perlman, Klein, & Greger, 1998). In fact,
the use of vitamin and minerals by those with a lower
body weight has been interpreted as beneficial, espe-
cially if those individuals are not getting adequate
nutrition from dietary sources (Gray et al., 1996).
However, the majority of users in this study had a
BMI ranging from 20 to 25, which is considered the
“ideal” and is associated, for most people, with the
lowest risk of illness (Health and Welfare Canada,
1988). Also, participants from the present study were
unlikely to have low nutrient intakes (McKenzie,
1999).
Scaling analyses suggest that this tool requires further
development, although elimination of items due to
low item–total correlation did not improve results, a
fact which suggests that multiple attributes of “use”
are being measured, rather than a uni-dimensional
scale. Despite the lack of internal consistency in this
16-item scale, it demonstrated validity, in that users of
herbal remedies had significantly higher Propensity to
Use Supplements scores than non-users. Additionally,
users of vitamin–mineral preparations also had higher
scores (8.4 vs. 6.6) than non-users, although this differ-
ence was not significant. Key questions may be those
found to be significantly associated with total supple-
ment use in the bivariate analyses. The literature sug-
gests that there is some basis for the Propensity to Use
Supplements scale: concern with the safety of the food
supply, being proactive about one’s health, and seeing
health from a holistic perspective have been reported
in the literature describing older adult users of sup-
plements (Ipsos-Reid, 2001; Powell, 2002; Astin, 1998;
Kaptchuk & Eisenberg, 1998; Rajendran, Thompson,
& Reich, 2001; Cassileth, 1999).
Population surveys investigating the prevalence of
complementary and alternative medicine (CAM)
reveal CAM users to be typically white, female, afflu-
ent, middle-aged, and well-educated (Verhoef et al.,
1994; Blais et al., 1997; Millar 1997). Yet specific exami-
nation of herbal remedy users is required and the
characteristics associated with being an herbal rem-
edy user are unknown. Research among low-income
American populations has shown that this low-
income group uses herbs, yet there were no significant
relationships relating herb use to gender, income, edu-
cation level, or race (Planta, Gundersen, & Petitt, 2000;
Wolsko et al., 2000). To date, limited research has
focused specifically on older Canadian adults and
their use of herbal preparations. In the present study,
in the bivariate analyses, there were few characteris-
tics where herb users varied significantly from non-
users; being younger, being married, and weighing a
lower percentage of ideal body weight characterized
herb users. NHANES III, in the United States, also
identified herb users as being younger (55–74 years)
(Radimer et al., 2000).
At present there is no clear profile of those seniors
who use several vitamin–mineral or herbal prepara-
tions. Due to the small sample size in this study, only
eight variables identified in the conceptual frame-
work were multivariately modelled in an attempt to
identify this profile. This is the first known report to
model a variety of nutrition risk factors, in addition to
demographic factor, on vitamin–mineral and/or
herbal remedy use in older Canadian adults. The two
retained variables, lower BMI and higher Propensity
to Use Supplements score, suggest that high users of
supplements may be health conscious and proactive
about their health.
Ideal body weight (IBW) has not been examined in
previous supplement literature, but this association

Multivariate analyses have been completed by both
Gray et al. (1996) and Astin (1998) yet were different,
in that their sole focus was on either vitamin–mineral
preparations or alternative health care. Examining
predictors of vitamin–mineral use among study par-
ticipants aged 65 years and older in the United States,
Gray et al. (1996) identified users as white women,
with a higher education, a lower BMI, greater use of
prescription drugs, frequent use of a health care pro-
vider, and supplemental health insurance. However,
BMI was the only variable consistent with the present
study findings. Greater use of prescription drugs and
health care providers in Gray et al.’s work may indi-
cate access issues that are different between Canada
and the United States.
requires further development and validation, it is
apparent that attitudes towards health, our food sup-
ply, and traditional and complementary therapies are
important factors in determining use of supplements,
and further work on this initial version of the Propen-
sity to Use Supplements scale is warranted. Addition-
ally, examining health and nutrition covariates, as was
begun in this work, should be continued for more
diverse, randomly selected, older Canadian popula-
tions.
Previous work led to conflicting results when supple-
ment use was examined at the bivariate level. The use
of a multivariate approach sheds light on some new
and interesting associations, which help to character-
ize the user better, and when specifically examined,
reveal that a Propensity to Use Supplements scale
offers some interesting concepts to be considered in
future supplement research among older Canadian
adults. Given that supplement use will likely continue
to increase, further Canadian research is needed to
understand which seniors are self-prescribing supple-
ments and why they are doing so.
Astin (1998) used multiple regression analyses to
develop a theoretical model to account for the increas-
ing use of alternative forms of health care. Satisfac-
tion, need, and philosophical values were predictors
of alternative health care. This study identified
“health philosophy” as an important concept in the
use of alternative care and paralleled the Propensity
to Use Supplements concept of the present study. Fur-
ther examination of the 16 items within the actual Pro-
pensity scale revealed that only 1 question was
significant in the multiple regression analysis. This 1
question suggests that seniors who have no one they
trust to talk to about supplements are more likely to
use these products. This suggests a potential gap in
traditional health care: seniors are interested in and
concerned about the use of supplements and are likely
to try them. Further work, perhaps qualitative, should
explore the decision-making process of seniors when
choosing to use supplements.
Angus Reid. (1997, August 22–24). Use and dangers of
alternative medicines and practices [Poll]. Retrieved
January 23, 1998, from http://www.angusreid.com/
pressrel/alternat.htm
Astin, J.A. (1998). Why patients use alternative medicine:
Results of a national study. Journal of the American
Medical Association, 279(19), 1548–1553.
Bender, M., Levy, A., Schucker, R., & YeE. (1992). Trends in
prevalence and magnitude of vitamin and mineral
supplement usage and correlation with health status.
Journal of the American Dietetic Association, 92(9), 1096–
1101.
Although this study provides some interesting
insights into supplement use by older adults, it has
limitations. As a small convenience sample was used,
generalizations to other seniors are not possible. As
well, despite seniors having been recruited from 10
diverse settings, there was some homogeneity in sup-
plement use, as almost three quarters were taking
some kind of supplement. This homogeneity led to
decisions about defining the dependent variable (total
supplement use as compared to the use/non-use
described in previous work) and may have influenced
the multivariate model results. Caution is warranted
when interpreting this finding, as statistical models
are limited by the sample on which they are based and
the variables selected for modelling. However, this
study provided the opportunity to explore, in a more
thorough manner, factors associated with supplement
use that have not previously been investigated. Spe-
cifically, a scale that examines the attitudes towards
supplement use was developed. Although this scale
Betts, N., & Rezek, J. (1989). Attitudes of rural and urban
elderly concerning supplement use. Journal of Nutrition
for the Elderly, 8(3/4), 67–77.
Blais, R., Maiga, A., & Aboubacar, A. (1997). How different
are users and non-users of alternative medicine.
Canadian Journal of Public Health, 88(3), 159–162.
Cassileth, B. (1999). Complementary therapies: Overview
and state of the art. Cancer Nursing, 22(1), 85–90.
Donald, E., Tapan, K., Hargreaves, J.A., Thompson, G.,
Overton, T., Peterson, R.D., & Chao, E. (1992). Dietary
intake and biochemical status of a selected group of
older Albertans taking or not taking micronutrient
supplements. Journal of the Canadian Dietetic Association,
53(1), 39–43.
Eisenberg, D.M., Davis, R.B., Ettner, S.L., Appel, S., Wilkey,
S., Van Rompay, M., & Kessler, R.C. (1998). Trends in
alternative medicine use in the United States, 1990–
1997. Journal of the American Medical Association, 280(18),
1569–1575.

Foster, D., Philips, R., Hamel, M., & Eisenberg, D. (2000).
Alternative medicine use in older adults. Journal of the
American Geriatrics Society, 48(12).
Montbriand, M.J. (2000). Senior and health-professionals’
perceptions and communication about prescription and
alternative therapies. Canadian Journal of Aging, 19(1),
35–56.
Oakland, M.J., & Thomsen, P.A. (1990). Beliefs about and
usage of vitamin/mineral supplements by the elderly
participants of rural congregate meal programs in
central Iowa. Journal of the American Dietetic Association,
90(5), 715–716.
Fries, C.J., & Menzies, K.S. (2000). Gullible fools or desperate
pragmatists? A profile of people who use rejected
alternative health care providers. The Canadian Journal of
Public Health, 91(3), 217–219.
Gibson, R (1990). Principles of nutritional assessment. New
York: Oxford University Press.
Payette, H., & Gray-Donald, K. (1991). Do vitamin and
mineral supplements improve the dietary intake of
elderly Canadians? Canadian Journal of Public Health, 82,
58–59.
Gray, G., Paganini-Hill, A., & Ross, R. (1983). Dietary intake
and nutrient supplement use in a southern California
retirement community. The American Journal of Clinical
Nutrition, 38, 122–128.
Planta, M., Gundersen, B., & Petitt, J. (2000). Prevalence of
the use of herbal products in a low-income population.
Family Medicine, 32(4), 252–257.
Gray, S.L., Hanlon, J.T., Fillenbaum, G.G., Wall, W.E., &
Bales, C. (1996). Predictors of nutritional supplement
use by the elderly. Pharmacotherapy, 16(4), 715–720.
Health and Welfare Canada. (1988). Promoting healthy
weights: A discussion paper. Ottawa, ON: Minister of
Supply and Services.
Powell, D. (2002). Food safety: A scientific perspective.
Canadian Journal of Dietetic Practice and Research 63(1).
Radimer, K.L., Subar, A.F., & Thompson, F.E. (2000).
Nonvitamin, nonmineral dietary supplements: Issues
and findings from NHANES III. Journal of the American
Dietetic Association, 100, 447–454.
Ipsos-Reid. (2001, October 9). Canadians and food safety.
Retrieved November 25, 2002, from http://www.ipsos-
reid.com/media/
dsp_displaypr_cdn.cfm?id_to_view = 1324
Rajendran, P., Thompson, R., & Reich, S. (2001). The use of
alternative therapies by patients with Parkinson’s
disease. Neurology, 57(5), 790–794.
Ranno, B., Wardlaw, G., & Gieger, C. (1988). What
characterizes elderly women who overuse vitamin and
mineral supplements? Journal of the American Dietetic
Association, 88(3), 347–348.
Kaptchuk, T.J., & Eisenberg, D.M. (1998). The persuasive
appeal of alternative medicine. Annals of Internal
Medicine, 129, 1061–1065.
Keller, H.H., McKenzie, J.D., & Goy, R.E. (2001). Construct
validation and test-retest reliability of SCREEN Seniors
in the Community: Risk Evaluation for Eating and
Nutrition. Journal of Gerontology, Medical Sciences,
56A(9),:M552-M558. .
Stoehr, G.P., Gaguli, M., Seaberg, E.C., Echement, D.A., &
Belle, S. (1997). Over-the-counter medication use in an
older rural community: The movies project. Journal of
the American Geriatrics Society, 45(2), 158–165.
Kleinbaum, D. (1994). Logistic regression: A self-learning text.
New York: Springer.
Lyle, B.J., Mares-Perlman, J.A., Klein, B.E., Klein, R., &
Greger, J.L. (1998). Supplement users differ from
nonusers in demographic, lifestyle, dietary and health
characteristics. Journal of Nutrition, 128, 2355–2362.
McKenzie, J.D. (1999). Vitamin–mineral supplement and herbal
remedy use in community-living older adults. Unpublished
master’s thesis, Department of Family Relations and
Applied Nutrition, University of Guelph, Guelph, ON.
McKenzie, J.D., & Keller, H.H. (2001). Use of vitamin–
mineral supplements and herbal remedies among
community-living older adults. Canadian Journal of
Public Health, 92(4), 286–289.
Streiner, D., Norman, F. (1995). Health measurement scales–A
practical guide to their development and use (2nd ed.). New
York: Oxford University Press.
Verhoef, M., Russell, M., & Love, E. (1994). Alternative
medicine use in rural Alberta. Canadian Journal of Public
Health, 85(5), 308–309.
Wolsko, P., Ware, L., Kutner, J., Lin,C., Albertson, G., Cyran,
L., Schilling, L., & Anderson, R. (2000). Alternative/
complementary medicine: Wider usage than generally
appreciated. The Journal of Alternative and
Complementary Medicine, 6(4), 321–326.
Millar, J. (1997). Use of alternative health care practitioners
by Canadians. Canadian Journal of Public Health, 88(3),
154–158.