Dietary supplements: prevalence of use in the New Zealand population.

Abstract

Objective: To determine prevalence of dietary supplement use in New Zealand and to describe types of supplements used.

Design: Secondary data analysis of two cross-sectional national surveys.

Subjects: In total, 3275 children aged 5-14 years, 4636 adults aged 15+ years.

Setting: Homes throughout New Zealand.

Main outcome measures: Prevalence data over 24 hours (children and adults) and over previous year (adults only), type of supplements consumed, pattern of dietary supplement use.

Results: Adults (24.2%) were more likely than children (5.4%) to consume any dietary supplement in the previous 24 hours. Over the previous year 59% of adults had consumed at least one supplement. Adult females were more frequent consumers than males both in the previous year (67.6%, 50.4%) and in the previous 24 hours (30.5%, 17.8%). Multivitamins and minerals were the type most frequently consumed by adults (19.8%) and children (2.1%). Older adults, compared with younger adults, are more likely to consume supplements daily and less likely to consume an individual supplement for a limited time.

Conclusions: The prevalence of use of dietary supplements of a population differs markedly depending on the duration of the period over which use of a supplement is recalled. In New Zealand adults, prevalence of use was 2.4 times higher recalling some use over the previous year, than when recalling use in the previous 24 hours. Use varied by age for adults and by ethnicity among adults and children.

Key words: dietary supplement, epidemiology, minerals, New Zealand, prevalence, questionnaire, vitamins.

INTRODUCTION

Foods may no longer be the sole source of nutrients, with nutrients sourced from both foods (frequently including those fortified with specific nutrients) and dietary supplements. (1) The term 'dietary supplements' once analogous with vitamins and minerals, is now used to include botanical supplements such as echinacea, various forms of garlic and a variety of physiologically active compounds derived from foods or other biological material. Although there has been a worldwide trend of an increase in the prevalence of dietary supplement use, (2-4) the necessity for and cost and safety of dietary supplements continues to be debated. (5)

The measurement and assessment of dietary supplement intake is an area with no international agreement, with little consistency in types of supplements specified, duration of time period of use, level of frequency of data collected and amounts consumed. The personal interview method in the home has been reviewed by Radimer who concluded that this was the best method of dietary-supplement data collection. (5) The time span pertaining to the period over which dietary supplement intake data are obtained will affect prevalence rates. Supplement use can be sporadic, irregular or regular and an individual may vary their supplement intake from month to month or week to week. (6) Information on supplement use 'over the previous year' will give higher prevalence rates than estimating prevalence based on consumption in 'the previous 24 hours' or 'in the previous week'. (4,7) This impact of the duration of the period over which prevalence rate is assessed can make comparisons between studies difficult.

Data for adults aged 15+years from the 1997 National Nutrition Survey (NNS97) (8) and for children aged 5-14 years from the 2002 Children's Nutrition Survey (CNS02) (9) were collected using the personal interview method, and provide the first opportunity to determine the prevalence of dietary supplement use in the New Zealand (NZ) population, and to describe the types of supplements used. Moreover, the questions used in the adult survey enable a comparison of prevalence rates, depending on whether data relate to a short period in the immediate past, or recall over a longer time period.

METHODS

The NNS97 was a voluntary cross-sectional survey of New Zealanders aged 15 years and above (n = 4636). The ethical approval, survey design and methodology are described elsewhere. (10) Information was collected for each participant regarding their use of dietary supplements in the previous year and in the previous 24 hours (as part of the 24-hour diet recall process, directly after foods and beverages were recalled). A dietary supplement was defined as 'any product intended for ingestion as a supplement to the diet'. Supplements were classified as 'vitamin and mineral supplements', including all single and multi-nutrient preparations and 'non-vitamin and mineral supplements', including herbal supplements, sports preparations and garlic products. The questions were interviewer-administered in the respondent's home and responses entered directly onto a laptop computer. Further to this, questions on supplement use in the previous year were put to participants who were asked how often they consumed each type of supplement that they reported taking. They were given seven options: less than one per month, one to three times per month, once per week, more than once per week, daily, episodic (regular supplement taking for a limited time period) and don't know. Participants were classified into a user category of daily, episodic or less than daily, based on the most frequently reported consumption of any supplement, over the previous year.

The CNS02 was a voluntary cross-sectional school-based survey of NZ children aged 5-14 years (n = 3275). The ethical approval, methodology and survey design are described in 'New Zealand Food: New Zealand Children'. (9) Interviews with children aged five to nine years were undertaken with a parent or caregiver present. Dietary supplements were defined and categorised as in NNS97. However, in CNS02, the use of dietary supplements was only asked within the 24-hour diet recall. No data were collected on supplement use in the previous year.

In both the NNS97 and CNS02 the brand name and amount consumed (including units) of supplements consumed per occasion were requested for each dietary supplement. If the supplement container was available, the brand name and dose were identified directly using a programmed barcode scanner. When the brand and product name were not known, a series of computer-generated questions were asked of the participant, to obtain the necessary information to enable classification of each dietary supplement.

The NNS97 and CNS02 were carried out using a multistage sampling design including stratification (NNS97 only), clustering and unequal probabilities of selection. Analyses were undertaken in Stata Statistical Software release 5.0 (College station TX: Stata Corporation, College Station, TX, USA) to allow for the complex sampling design. All proportions reported have been weighted to provide population estimates. Trends across age and sex groups were tested using the Wald test for linear trend, and were considered significant when P < 0.05.

RESULTS

Prevalence of use

Fifty-nine per cent of NZ adults consumed at least one dietary supplement 'in the previous year' with use higher for adult females than males (Table 1). This trend applied to the prevalence of consumption of both vitamin and mineral and non-vitamin and mineral supplements (females 59.5%, males 41.8% and females 34.5%, males 22.5%, respectively). The proportion of NZ adults consuming at least one dietary supplement in the previous 24 hours was lower (24.1%), and adult females were more likely than adult males to consume supplements across all age groups, irrespective of the period over which use was reported (Table 1). However, the age trend in prevalence of supplement use differed according to the duration of the time period reported (Table 1). Over the previous year, for both males and females, prevalence fell with age. In contrast, prevalence of use in the previous 24 hours increased with age for both sexes.

Younger males and females have a higher prevalence of episodic use, but a lower daily use when compared with older age groups (Table 2).

Supplement use by children aged 5-14 years in the previous 24 hours was comparatively low (Table 3). In the 7-10 years group, males had a significantly higher prevalence than females.

Supplement use in the previous 24 hours by both adults and children was highest for NZ European and others, followed by Maori and then Pacific (Table 4). Dietary supplement use among Maori and Pacific children was negligible or very small.

Identification of supplements

The brand and product name were not identified for 21% of the supplements, with 7.6% of these having insufficient information to match to a generic line. This failure does not markedly influence the prevalence data, but may underestimate the contributions of supplements to the total nutrient intake.

Types of supplements used

Children -- in previous 24 hours

The most frequently reported supplement types consumed by children in the previous 24-hour period were multivitamins and minerals (2.1%) and vitamin C (2.0%).

Adults -- in previous year

Multivitamins and minerals were the most frequently consumed type of supplement, followed by vitamin C and B-complex vitamins (Table 5). Females had a significantly higher prevalence of supplement use compared with males for multivitamins and minerals, garlic, iron, calcium, evening primrose oil and zinc. Among males, there was a decreasing prevalence of use with age for multivitamins and minerals, B-complex and vitamin C and an opposite trend for bee products.

Among females, prevalence of use fell with age for multivitamins and minerals, B-complex, iron and zinc. Evening primrose oil was used significantly more often by women aged 25-64 years compared with younger and older women.

Variety of supplements that adults used

A total of 5836 unique supplements were reported by adults as having been used in the previous year; about two-thirds of these were classified as vitamin and mineral supplements and one-third were non-vitamin and mineral supplements. The most frequently reported vitamin and mineral supplements were multivitamins and minerals (n = 898), vitamin B-complex supplements (n = 739) and vitamin C (n = 729). The most frequently reported non-vitamin and mineral supplements were garlic (all forms) (n = 523), evening primrose oil (n = 196) and bee products (n = 132).

DISCUSSION

The NNS97 is the most comprehensive survey to examine supplement use among NZ adults and the CNS02 the first national dietary survey of children to both include nutrients from supplements in the dietary assessment and also to document dietary supplement use. (9) The data provide the first opportunity in NZ to determine which groups in the population consume dietary supplements and which supplements they choose. Furthermore, these data provide an opportunity to explore two different estimates of prevalence of dietary supplement use: reported use 'over the previous year' and 'in the previous 24 hours'.

Supplement use 'in the previous 24 hours' provides data that can reasonably be compared with overseas studies, although such comparisons must be made with caution when definitions of a dietary supplement are not identical, when use is determined over time spans greater than 24 hours (e.g. three days, (11) two days, (12) past month (13)) and when the time period to which the data pertain are different.

Twenty-five per cent of Finnish adults reported using a dietary supplement (over three days) in 1992, (11) which is very similar to the 24% of NZ adults reported in the present paper (1997 data) and the 24% of Australian adults (7) (1996 data), both reporting previous 24-hour use. The lower prevalence rates reported in overseas data from a decade earlier (UK 11% reported use over seven days, (14) Netherlands 10-16% reported using a supplement over two days (12)) may simply reflect the trend to increased supplement use. US data on prevalence of dietary supplement use in the past month for adults aged 20 years and older, between 1971 and 2000, showed increases from 27.5% to 47.1% for male and from 37.9% to 56.7% for female adults aged 20 years and older. (13)

It appears from the different patterns of dietary supplement use, obtained by reporting intake over the previous year and recalling actual intake in the previous 24 hours, that older and younger New Zealanders have very different practices with respect to dietary supplements. Younger adults are more likely than older adults to use supplements on an episodic or less than daily basis. In contrast, older adults are more likely to be regular users, as evidenced by both their higher reported pattern of daily use (in the previous year) and higher actual daily use (in the previous 24 hours).

A partial picture of dietary supplement use is obtained from recalling use in the previous 24 hours. It appears to be an underestimate of overall use, as a considerable number use supplements on a non-regular basis: in short bursts or episodes or irregularly. This use is captured by the documentation of use over the previous year and highlights the more occasional or short-term use by younger adults compared with more dedicated, regular use by older adults.

There are also limited data for international comparisons with NZ children, particularly as data were only collected on supplement use in the previous 24 hours. British data for children aged 4-18 years reported that 19% were 'currently taking a vitamin or mineral supplement'. (15) US data for children from NHANES 1971-2000 found that frequency of dietary supplement use over the previous month has remained fairly constant at approximately 34% for those aged 6-11 years but lower for the 12-15 years group, at approximately 21%. (4) Data from 15 275 English and Scottish children showed a prevalence of 16% use of dietary supplements over the past month. (16) It is likely that for children, as for adults, reporting frequency of use over a specified time, compared with recalling actual use in the immediate past 24-hour period, provides a different 'indicator' of dietary supplement use. Therefore, although it appears that NZ children are lower consumers of supplements compared with children in other countries, we cannot conclude this with any certainty until data are collected over a similar time period. However, we can conclude that the consumption of dietary supplements by children is considerably lower than for NZ adults based on 'actual use' in the previous 24 hours.

CONCLUSIONS

On the basis of use in the previous 24 hours supplement use is very low in NZ children (5.4%), but higher in adult females (30.5%) and males (17.8%). The most frequently consumed type of supplement was multivitamins and minerals by both adults and children. Prevalence of use estimates for adults differ markedly depending on the duration of the time period over which use is reported. Older adults are more likely to be regular users as evidenced by their higher use in the previous 24 hours', but younger adults report a higher episodic use in the previous year. Comparatively fewer younger adults have consistent supplement taking habits.

Surveys of the prevalence of use of dietary supplements must determine whether the aim is to assess regular or episodic and irregular use as well as focusing on determining supplement types consumed.

ACKNOWLEDGEMENTS

The 1997 National Nutrition Survey and the 2002 National Children's Nutrition Survey were funded by the Ministry of Health (NZ). Principal Investigators of the NNS97 were: University of Otago: Dr Jim Faed, Dr Elaine Ferguson, Mr Peter Herbison, Dr Caroline Horwath, Dr Ted Nye, Mrs Winsome Parnell, Dr Papaarangi Reid (Wellington School of Medicine), Professor David Russell, Associate Professor Rob Walker, Dr Barry Wilson, Dr Noela Wilson; University of Auckland: Dr Colin Tukuitonga. Principal Investigators of the CNS02 were: University of Auckland: Dr Robert Scragg, Mr David Schaaf; Massey University: Mr Eljon Fitzgerald; University of Otago: Mrs Winsome Parnell, Dr Noela Wilson.

Statistical advice was obtained from M. Wohlers (LINZ[R] and A. Gray, and programming services from C. Blakey (LINZ[R]).

REFERENCES

1 Milner JA. Functional foods: the US perspective. Am J Clin Nutr 2000; 71: 1654S-9S.

2 MacLennan AH, Wilson DH, Taylor AW. The escalating cost and prevalence of alternative medicine. Prev Med 2002; 35: 166-73.

3 Tindle HA, Davis RB, Phillips RS, Eisenberg DM. Trends in use of complementary and alternative medicine by US adults: 1997-2002. Altern Ther Health Med 2005; 11: 42-9.

4 Briefel RR, Johnson CL. Secular trends in dietary intake in the United States. Annu Rev Nutr 2004; 24: 401-31.

5 Radimer KL. Methodological issues in assessing dietary supplement use in children (Comment). J Am Diet Assoc 2005; 105: 703-8.

6 Block G, Sinha R, Gridley G. Collection of dietary-supplement data and implications for analysis. Am J Clin Nutr 1994; 59 (1 Suppl.): 232S-9S.

7 Lawrence MA, Rutishauser IHE, Lewis JL. A practical approach to monitoring nutrient supplement intake of Australian adults. Aust J Nutr Diet 2001; 58: 98-103.

8 Russell DG, Parnell WR, Wilson NC. NZ Food: NZ People. Key Results of the National Nutrition Survey. Wellington, New Zealand: Ministry of Health, 1999.

9 Parnell WR, Scragg RKR, Wilson NC, Schaaf D, Fitzgerald EDH. NZ Food: NZ Children: Key Results of the 2002 National Children's Nutrition Survey. Wellington, New Zealand: Ministry of Health, 2003.

10 Parnell WR, Wilson NC, Russell DG. Methodological advances in the 1997 NZ National Nutrition Survey. N Z Med J 2001; 114: 123-6.

11 Kaartinen P, Ovaskainen ML, Pietinen P. The use of dietary supplements among Finnish adults. Scand J Nutr 1997; 41: 13-17.

12 Dorant E, van den Brandt PA, Hamstra AM et al. The use of vitamins, minerals and other dietary supplements in The Netherlands. Int J Vitam Nutr Res 1993; 63: 4-10.

13 Ervin RB, Wright JD, Kennedy-Stephenson J. Use of dietary supplements in the United States, 1988-94. Vital Health Stat 1999; 11: 1-14.

14 Gregory J, Foster K, Tyler H, Wiseman M. The Dietary and Nutrition Survey of British Adults. London: HMSO, 1990.

15 Gregory J, Low S, Bates CJ et al. National Diet and Nutrition Survey: Young People 4-18 Years. London: TSO, 2000.

16 Bristow A, Qureshi S, Rona RJ, Chinn S. The use of nutritional supplements by 4-12 year olds in England and Scotland. Eur J Clin Nutr 1997; 51: 366-9.

Winsome R. PARNELL, (1) Noela C. WILSON (2) and Claire SMITH (2)

(1) Department of Human Nutrition and (2) Life in New Zealand Activity and Health Research Unit, Research Unit, University of Otago, Dunedin, New Zealand

W.R. Parnell, PhD, Senior Lecturer 

N.C. Wilson, PhD, Formerly Director

C. Smith, MSc, Formerly Nutritionist 

Correspondence: W.R. Parnell, Department of Human Nutrition, University of Otago, PO Box 56, Dunedin, New Zealand. Email: [e-mail]

Table 1 Prevalence of adult supplement use (two time periods)

                            Previous year
Age       Males             Females           P for difference
(years)   % (CI (a))        % (CI)            between sexes

15-24     58.5 (50.4-66.5)  71.1 (65.1-77.1)   0.0137
          (n = 254)         (n = 346)
25-44     52.3 (47.6-57.0)  69.9 (66.5-73.3)  <0.0001
          (n = 759)         (n = 1205)
45-64     46.3 (41.0-52.0)  67.2 (62.7-72.0)  <0.0001
          (n = 588)         (n = 667)
65+       40.5 (34.0-47.1)  58.4 (52.5-64.2)  <0.0001
          (n = 326)         (n = 491)
All ages  50.4 (47.3-53.4)  67.6 (65.3-70.0)  <0.0001
          (n = 1917)        (n = 2709)

                            Previous 24 hours
Age       Males             Females           P for difference
(years)   % (CI)            % (CI)            between sexes

15-24     12.5 (7.6-17.3)   20.5 (14.1-26.9)   0.0505
          (n = 254)         (n = 346)
25-44     16.1 (12.5-19.8)  28.6 (25.1-32.0)  <0.0001
          (n = 759)         (n = 1205)
45-64     21.6 (17.8-26.2)  36.8 (32.2-41.5)  <0.0001
          (n = 588)         (n = 667)
65+       22.8 (17.5-28.1)  37.0 (31.3-42.7)   0.0004
          (n = 326)         (n = 491)
All ages  17.8 (15.5-20.0)  30.5 (28.2-32.9)  <0.0001
          (n = 1927)        (n = 2709)

(a) 95% confidence intervals.

Table 2 Prevalence pattern of adult dietary supplement use in the
previous year

                             Males % (CI (a))
Age (years)      Daily        Episodic (b)  <Daily (c)

15-24            24.2         52.8          22.9
                 (16.1-32.4)  (42.9-62.7)   (14.5-31.3)
25-44            31.1         48.6          20.3
                 (25.6-36.5)  (41.9-55.4)   (14.8-25.7)
45-64            54.7         34.1          11.2
                 (45.8-63.5)  (25.8-42.8)   (6.3-16.1)
65+              73.4         15.4          10.9
                 (64.6-82.8)  (8.6-22.3)    (4.0-17.8)
P-value for      <0.0001      <0.0001       <0.0001
  trend for age
All years        39.8         42.5          17.6
                 (35.6-44.0)  (38.1-46.9)   (14.3-21.0)

                            Females % (CI (a))
Age (years)      Daily        Episodic (b)  <Daily (c)

15-24            37.6         49.5          12.7
                 (28.4-47.0)  (39.7-59.4)   (7.4-18.1)
25-44            48.0         42.2           9.8
                 (43.4-52.6)  (37.7-46.7)   (7.2-12.4)
45-64            59.5         33.1           7.4
                 (53.8-65.2)  (27.7-38.5)   (4.2-10.5)
65+              71.7         19.5           8.7
                 (64.2-79.3)  (13.4-25.6)   (2.9-14.5)
P-value for      0.0010       <0.0001       <0.0001
  trend for age
All years        52.1         38.3           9.6
                 (48.9-55.4)  (35.0-41.5)   (7.7-11.5)

(a) 95% confidence interval.
(b) Episodic includes regular supplement taking for a limited time
period.
(c) <Daily includes: more than once per week, once per week, one to
three times per month, less than once per month.

Table 3 Prevalence of any dietary supplement use in the previous 24
hours for New Zealand children

                                                   P for difference
Age (years)  Males % (CI (a))  Females % (CI (a))  between sexes

5-6          7.2 (3.2-11.2)    7.7 (3.2-12.2)      0.87
             (n = 383)         (n = 309)
7-10         6.9 (4.3-9.5)     3.0 (1.2-4.9)       0.02
             (n = 738)         (n = 687)
11-14        4.3 (2.1-6.5)     5.4 (2.8-8.1)       0.51
             (n = 576)         (n = 582)
Total 5-14   5.9 (4.3-7.5)     4.9 (3.3-6.5)       0.36
             (n = 1697)        (n = 1578)

(a) 95% confidence interval.

Table 4 Prevalence of any dietary supplement use in the previous 24
hours, by ethnicity

          Ethnicity      Males % (CI (a))      Females % (CI (a))

Adults    Maori          10.9 (5.3-16.4)       14.3 (10.1-18.6)
                         (n = 268)             (n = 436)
          Pacific         3.8 (0.8-6.8)         5.9 (2.0-9.7)
                         (n = 116)             (n = 191)
          NZ European    19.4 (16.8-21.9)      34.1 (31.3-36.8)
            and others   (n = 1543)            (n = 2082)
          P-values for   M < P, P = 0.0200     M > P, P = 0.0036
            differences  P < NZEO, P < 0.0001  P < NZEO, P < 0.0001
            between      M < NZEO, P = 0.0065  M < NZEO, P < 0.0001
            ethnicities
Children  Maori           1.6 (0.5-2.6)         2.3 (1.0-3.7)
                         (n = 631)             (n = 593)
          Pacific         0.0                   0.1 (0.0-2.0)
                         (n = 514)             (n = 544)
          NZ European     8.2 (5.9-10.5)        6.3 (4.0-8.5)
            and others   (n = 552)             (n = 441)
          P-values for   M > P, P = 0.0037     M = P, P = 0.1253
            differences  P < NZEO, P < 0.0001  P < NZEO, P < 0.0001
            between      M < NZEO, P < 0.0001  M < NZEO, P = 0.0031
            ethnicities

(a) 95% confidence interval.
NZ = New Zealand.

Table 5 The most frequently used supplements in the previous year among
New Zealand adults

                    Multivitamins  Vitamin
% Using             and minerals   C        B-complex  Garlic   Iron

Males (n= 1918)
  15-24 years       17.2           28.0     17.6        8.9      4.2
  25-44 years       16.8           18.2     22.1        9.5      0.8
  45-64 years       11.5           15.2     12.7       13.2      0.3
  65+ years          7.9            6.0      3.0        8.3      1.7
  P for age trend   <0.0001        <0.0001  <0.0001     0.3700   0.0900
  Total             14.3           17.7     16.2       10.2      1.4
Females (n = 2699)
  15-24 years       26.1            3.7     21.4        9.7     13.4
  25-44 years       28.6           19.4     25.2       10.7     14.9
  45-64 years       24.3           16.3     18.2       17.6      6.7
  65+ years         14.7           10.8      9.4       13.5      3.7
  P for age trend   <0.0001        <0.0001  <0.0001     0.6100  <0.0001
  Total             25.3           20.5     20.2       12.7     10.8
P for sex effect    <0.0001         0.1100   0.0205     0.0500  <0.0001

                             Evening
                             primrose            Bee
% Using             Calcium  oil       Selenium  products  Zinc

Males (n= 1918)
  15-24 years        0.2      0.2      0.8        0.6      1.4
  25-44 years        0.8      0.3      2.3        1.2      1.3
  45-64 years        2.1      0.5      3.2        3.2      2.8
  65+ years          1.6      1.3      3.6        5.7      2.5
  P for age trend    0.0700   0.2200   0.1500     0.0020   0.4800
  Total              1.4      0.4      2.5        2.2      1.9
Females (n = 2699)
  15-24 years        5.2      3.6      1.0        0.5      6.5
  25-44 years        4.7      7.2      2.0        1.8      3.1
  45-64 years        6.9     11.1      4.6        3.8      2.2
  65+ years          6.6      2.8      3.0        6.8      2.4
  P for age trend    0.9300   0.2900   0.0700    <0.0001   0.0400
  Total              5.6      6.9      2.6        2.9      3.4
P for sex effect    <0.0001  <0.0001   0.8200     0.2800   0.0500