The frequencies of supplement types and reasons mentioned by supplement users are given in Figure . The most widely used supplement was vitamin C, used by 41.9% of all athletes in the sample. Vitamin C was taken by 70.4% of athletes who use supplements but further analysis revealed that vitamin C was not taken for performance-related reasons (Tables , , and ). The most popular performance-related supplements used were creatine and whey protein (36.1% and 30.6%, respectively) with slightly less than one third of athletes taking iron (29.8%) and/or caffeine (23.8%). A minority used ginseng (8.3%). The prevalence of creatine, whey protein, iron, caffeine and ginseng use among all athletes responded showed the same pattern (21.5%, 18.2%, 17.7%, 14.2% and 4.9%, respectively).
Relative percentage of congruent answers by 'ability to train longer' and 'doctors' advice' as rationale for supplement use and supplement used by all athletes who reported the use of nutritional supplements (n = 520)
Analyses of the motives for supplement use reveal that of the three performance-related reasons, the desire to maintain strength predominates (38.1%). A notable cohort of 24.2% indicated the reasons for supplement use as doctors' advice and enhancement of endurance (20.0%) with 13.3% listing ability to train longer and a mere 3.8% citing provided by governing body as reason. Unfortunately the data did not contain information on the extent and type of supplements provided by governing bodies and if any, it was suspected to vary across sports.
The prevalence of associations are listed in Tables , , , and , and of the expected associations (11 test pairs), only eight were significant. Although expected, no significant associations were observed between iron and enhancing endurance or between vitamin C and doctor's advice or being able to train longer. The matrix cell for ginseng reflects the anecdotal but not-yet-proven effect of ginseng on performance.
Among reasons for supplement use, maintaining strength seems to be selected by many athletes (190/520), regardless of the sports, whereas athletes were more selective with sports specific reasons such as endurance (104/520) or ability to train longer (69/520) and there was a considerable overlap between the uses of these supplements. Of the 104 athletes who claimed enhancing endurance as a reason for supplement use 87 also selected maintaining strength from the list of reasons. The same is true for vitamin C which was taken by most athletes in the sample (366/520). For example, among those who reported the use of iron (155/520), 130 also used vitamin C; among those who used protein (159/520) 104 also reported the use of creatine.
The strength of these associations, however, varies as seen in Table . Of the 11 pairs with expected associations, three showed an intermediate association: maintaining strength – creatine and whey protein; and ability to train longer – creatine.
Under the assumption of a fully informed choice, it would be expected that the YY (Yes-Yes) cells to be or close to 100%. For example, those who wish to increase strength take the supplement that has this ergogenic effect (i.e. protein). However, with the exception of vitamin C, even the pairs with the strongest association coefficients showed poor results. Of those who take supplements to maintain strength, only 56.1% reported the use of whey protein. Similarly, creatine was taken by 73.9% of those who wish to be able to train longer and by only 62.6% of those who wish to maintain strength. The full matrix of the relative percentages of congruent answers by rationale for supplement use and supplement used by all athletes who reported the use of nutritional supplements are shown in Tables and .
Displaying the strength of association coefficients on a radar plot (Figure ) it is evident that athletes are reasonably informed about supplements that can be used to help maintain strength (creatine and whey protein) but the proportion of congruent answer pairs decrease when the supplements in question are other than creatine and protein. This is not surprising as creatine has proven effects on muscle mass, high intensity exercise capacity [29
] and sprint performance [30
], but not on endurance or prolonged training sessions [31
], and is considered as the most effective ergogenic nutritional supplement currently available to athletes [32
Strength of associations (expressed as phi coefficients) for the selected 9 pairs. Strength of associations are categorized as strong (ϕ > .7), intermediate (.7 < ϕ > .3) and weak (ϕ < .3).