- Email: [email protected]
Oral contraceptives decrease saliva testosterone but do not affect the rise in testosterone associated with athletic competition
Hormones and Behavior 56 (2009) 195–198
Contents lists available at ScienceDirect
Hormones and Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y h b e h
Oral contraceptives decrease saliva testosterone but do not affect the rise in testosterone associated with athletic competition David A. Edwards ⁎, J. Laurel O'Neal Department of Psychology, Emory University, Atlanta, GA 30322, USA
a r t i c l e
i n f o
Article history: Received 30 October 2008 Revised 10 January 2009 Accepted 14 January 2009 Available online 31 January 2009 Keywords: Saliva testosterone Oral contraceptives Athletic competition Hormone contraception
a b s t r a c t Women athletes from intercollegiate soccer, volleyball, and softball teams, and women skaters from a team competing in an amateur roller derby league, contributed saliva samples before warm-up and immediately after the completion of one or more sanctioned competitions. Women using oral contraceptives (OCs, n = 29) had a significantly lower mean level of saliva testosterone (T) than non-users (n = 51). Thus, OCs contribute predictable variation to individual differences in saliva T, and OC use is likely to contribute to individual differences in measures of psychological processes and/or behavior which are causally related to individual differences in circulating testosterone. Most of the women (n = 68) played during one or more of the competitions for which they contributed saliva samples. Whether for soccer, volleyball, softball, or roller derby, competition was associated with a robust increase in saliva T. Although OC users had significantly lower saliva T levels than non-users before and after-competition, both users and non-users showed virtually the same increase in saliva T over the course of competition. While the most proximal cause of this increase is not known, it is probably not the result of an increase in gonadotropin (GTH) secretion since an increase in GTH secretion would presumably be prevented by OC use. © 2009 Elsevier Inc. All rights reserved.
Introduction Testosterone is secreted primarily in the testes of males and the ovaries of females. In men (Bhasin et al., 2006) and, less certainly, women (Bhasin, 2005; Santoro et al., 2005), this hormone plays important roles in health and well-being, the maintenance of muscle mass and bone density, sexual desire and other psychological processes. The ability to measure testosterone non-invasively in saliva has made it possible to repeatedly sample individuals in widely varying settings, allowing for studies relating testosterone to such things as mood (van Honk et al., 1999), clothing choice (Grammer et al., 2004), cognitive abilities (Burkitt et al., 2007), sexual activity (van Anders et al., 2007), financial risk-taking (Coates and Herbert, 2008), attraction to masculine faces (Welling et al., 2007), relationship status (van Anders and Watson, 2006a), dominance/personality (Mazur and Booth, 1998; Schultheiss et al., 2005; Sellers et al., 2007), competitive response to defeat (Mehta and Josephs, 2006), athletic competition (e.g., Bateup et al., 2002; Kivligan et al., 2005) and status with teammates (Edwards et al., 2006). Despite a number of caveats (see the thorough review by Granger et al., 2004), it seems certain that measurement of testosterone from saliva samples will continue to be widely used in studies looking to understand how testosterone mediates psychological processing and behavior in men and women.
⁎ Corresponding author. Fax: +1 404 727 0372. E-mail address: [email protected] (D.A. Edwards). 0018-506X/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.yhbeh.2009.01.008
Oral contraceptives (OCs) are currently used by over 100 million women world-wide (Hatcher and Nelson, 2004). Modern formulations available in the United States contain a combination of a synthetic estrogen (either ethynyl estradiol or mestranol) and a progestin (Hatcher and Nelson, 2004). Their effectiveness is derived, at least in part, by blocking secretion of gonadotropins (FSH and LH) responsible for follicular development and ovulation (Lobo, 1994; Speroff and Darney, 2005). OCs in effect substitute low doses of exogenously-administered synthetic hormones for the naturally occurring levels of estrogen and progesterone that would otherwise vary systematically as part of a normal menstrual cycle. It is now abundantly clear that OC use decreases serum testosterone (e.g., Wiegratz et al., 1995; Coenen et al., 1996; Spona et al., 1996; Aden et al., 1998; Thorneycroft et al., 1999). It is perhaps therefore surprising that, with a few notable exceptions (e.g., Schultheiss et al., 2003; van Anders and Watson, 2006b; Wirth and Schultheiss, 2007), studies of reproductive-age women that assay testosterone from saliva samples often do not inquire about/analyze for effects of OC use by participants. Two computerized searches (PubMed and PsycINFO) of the English language research literature using “saliva and testosterone” as the search phrase turned up 35 different studies published in the 5-year span from 2003–2007 (inclusive) with women of reproductive age and presumed heterosexual orientation. For 12 of these studies, women using hormone contraception were either not a part of or were explicitly excluded from research participation. For 18 of the remaining 23 studies, experimenters either did not ask about participant use of hormone
196
D.A. Edwards, J.L. O'Neal / Hormones and Behavior 56 (2009) 195–198
contraception (9 studies) or asked but did not analyze for the possible effect of hormone contraception use on saliva testosterone (9 studies). Only 5 studies asked about hormone contraception use by participants and analyzed for effects on testosterone level — in four of these (Schultheiss et al., 2003, 2005; van Anders and Watson, 2006b; Wirth and Schultheiss, 2007) OC users had significantly lower saliva testosterone levels than nonusers. From 1999 through 2007, one of us (DE) supervised the conduct of five different studies with women athletes. As part of the consent process participants were specifically asked to indicate their use (or not) of OCs and/or other forms of hormone contraception. Procedures for saliva collection were the same for each of the studies. In all cases, samples were obtained in the afternoon or early evening and all samples were assayed for testosterone in the same laboratory using the same radioimmunoassay procedures. Team athletic competition is typically associated with an increase in saliva testosterone (e.g., Bateup et al., 2002; Edwards et al., 2006, 2007). For this article, we combined the results for these studies to give a sufficient number of women to analyze (and here report) the relationship between OC use and saliva testosterone with special attention to the possible effects of OC use on competition-related changes in levels of this hormone. Method Participants Women contributing saliva samples were participants in one of five studies conducted from 1999 through 2007. Four of the studies were conducted with varsity athletes from Emory University ranging in age from 18 to 22 years and included members of the 1999 women's soccer team (n = 16), 2002 and 2005 women's volleyball team (n = 14 and n = 19, respectively), and the 2004 women's softball team (n = 16). A fifth study was conducted with the members (n = 15) of the Toxic Shocks, one of four Atlanta-area teams in an amateur roller derby league. Roller derby is a contact sport, and competitions involve two teams (5 to a side) rollerskating against each other around an oval track. The Toxic Shocks ranged in age from 22 to 37 years. Procedures for each study were approved by the Emory University Institutional Review Board. All individuals were informed about the purpose of the study and about the method and frequency of saliva sampling for the study in which they would be participating. Each woman gave written informed consent. Although not queried about their undoubtedly varied use of prescription/non-prescription medications, as part of the consent procedure women were specifically asked about hormone contraceptive use. For the study with soccer players each woman was asked to respond “yes” or “no” to the following question: “Are you currently using oral contraceptives?” For all the other studies, women were asked to respond “yes” or “no” to this question and another: “Are you currently using any injected, implanted, or patch-delivered hormone contraceptive?”
Samples were assayed in either duplicate or triplicate and the mean of the values for any given sample was taken as the value to represent T level for that sample. Intrassay and interassay coefficients of variation varied across the five studies, but in every study were less than 10% and 20% respectively. Oral contraceptive use and the effect of athletic competition on saliva T Each of the studies was designed to determine the effect of team athletic competition on saliva T in women. To this end, saliva was collected before and immediately after one or more coach-selected home competitions. Collection before each of the competitions was done immediately prior to warm-up which began approximately 1 h before the start of play. This procedure was repeated immediately after the end of the competition so that samples were received within 15 min after the completion of the game or, in the case of volleyball players, a best-of-5-games match. For purpose of the present study, we took the T value for each individual's before warm-up saliva sample to represent the beforecompetition value for that individual for that competition. For soccer players we used values from samples associated with a single 1999 intercollegiate game played in mid-November — a 5–0 victory. For the 2002 volleyball players we averaged values for saliva samples for two intercollegiate matches played on consecutive days in mid-October — each a 3-games-to-1 victory. For the 2005 volleyball players we averaged sample values for three different intercollegiate matches played over the course of a weekend in late October — each one a loss. For the softball players we averaged sample values for five intercollegiate games (3 wins and 2 losses) played from late February through early April of the 2004 season. For the roller-derby team we used samples obtained for a single league competition (a loss) in late July, 2007. In women, team athletic competition is associated with an increase in saliva T (Edwards et al., 2006, 2007). To determine how this effect might be influenced by OC use, we first selected all of the individuals who played in the competitions for which we had saliva samples (see above). Next, we calculated the mean before-competition T value and the mean after-competition T value for those competitions in which the individual played, and took these values as the best estimate of the individual's before- and after-competition saliva T level. We then used these values to determine the effect of OC use on the hormonal response to intercollegiate competition. Statistical analyses The SPSS statistical package for personal computer was used for the calculation of Chi-square test of independence, independent and correlated t-tests (two-tailed), and one-way and mixed-design analysis of variance (ANOVA). In all cases, P b 0.05 was required for statistical significance. Results
Saliva samples and hormone assay
Oral contraceptive use by participants
For every study, participants were provided with a piece of sugarfree gum (Trident®, original flavor) to stimulate saliva production and a 20 ml plastic vial which they were asked to fill to a 5 ml line marked on the side. Samples were stored at − 20 °C within 30 min after collection and the frozen samples were later sent to the Biomarkers laboratory of the Yerkes Primate Center in Atlanta, Georgia for hormone assay. Samples were obtained at different times for different studies between noon and 2100. Saliva testosterone was assayed using a modification (Granger et al., 1999) of the Diagnostic Systems Laboratory (DSL, Webster, TX) radioimmunoassay kit with a range of analytical sensitivity of 2–500 pg/ml for test volumes of 200 μl.
Twenty-nine of the 80 participants reported using oral contraceptives; none reported any injected, implanted, or patch-delivered hormone contraception. Oral contraception use across the five teams ranged from 31–44% of the players and was not significantly different according to team (χ2 (4) = 0.54, ns). Oral contraceptive use and before- and after-competition testosterone levels Before-competition saliva T values were within the range expected for reproductive age women (e.g., Dabbs et al., 1995). For every team,
D.A. Edwards, J.L. O'Neal / Hormones and Behavior 56 (2009) 195–198
women using oral contraceptives had, on average, lower beforecompetition T values than women not using oral contraceptives (Fig. 1), but differences between users and non-users analyzed separately for each team were not significantly different for any team. Differences in mean before-competition T levels for the different teams were not significant by one-way ANOVA, so for purposes of analysis and presentation we combined values for the five teams creating two groups — women using oral contraceptives (n = 29) and women not using oral contraceptives (n = 51). Beforecompetition saliva T means for these two groups are shown in Fig. 1 (insert). Mean saliva T for OC-users is significantly lower than for non-users (t(78) = 2.3, P b 0.03.). Sixty-eight of the 80 women included in the analysis above played during one or more of the competitions for which they contributed saliva samples. Whether for soccer, volleyball, or softball, intercollegiate competition was associated with a robust increase in saliva T (Edwards et al., 2006, 2007) which was not affected by the outcome of the competition (see Edwards et al 2007 for more discussion). A similar competition-related rise in saliva testosterone was seen in roller derby skaters (unpublished observations). To determine if this rise in T level is influenced by OC use, we considered only the 68 women who played in at least one game (soccer and softball) or match (volleyball and roller derby). Before- and after-competition T means are shown for OC users and non-users in Fig. 2. Among the women who played, OC-users had significantly lower before- (t(66) = 2.0, P b 0.05) and after-competition (t(66) = 2.3, P b 0.03) T values than non-users. Mixed-design ANOVA with OC use as the between-groups variable and saliva T (before-competition and after-competition) as the within-groups (repeated measures) variable gives a significant effect of competition on T level (F(1,66) = 54.9, P b 0.01) and a significant effect of OC use (F(1,66) = 5.6, P b 0.03) on saliva T. The interaction between OC use and testosterone is not significant owing to the fact that in both OC users and non-users, saliva T increased from before- to after-competition by approximately the same amount (see Fig. 2). For a supplemental analysis of OC use as it might affect the competition-related increase in T level, we calculated the percent change in T level relative to pre-competition baseline for each player. From before warm-up to the end of competition, saliva T levels increased by an average of 50.5% for women not using oral contraceptives and by an average of 70.1% for
Fig. 1. Mean before-competition testosterone levels for women oral contraceptive (OC) users and non-users according to individual team and (insert) all teams combined. Sample sizes (n) for each group are shown. When teams are combined for analysis, the mean before-competition testosterone level for OC users is significantly lower than the comparable mean for non-users.
197
Fig. 2. Mean before- and after-competition testosterone levels for women oral contraceptive (OC) users and non-users. Groups only include individuals who played in the competitions for which they contributed saliva samples. Sample sizes for each group are shown (n). As noted in the figure, mean before- and after-competition testosterone levels for OC users are significantly lower than the comparable means for non-users.
women using oral contraceptives. These values are not significantly different (t(66) = −0.72). Discussion Saliva T is typically correlated with total serum T in blood (Granger et al., 2004). In the present study, OC users had significantly lower saliva T values than non-users which we presume reflect their lower serum levels of this same hormone. This result is consonant with observations that OC use is associated with lower serum levels of free and total T (e.g., Wiegratz et al., 1995; Coenen et al., 1996; Spona et al., 1996; Aden et al., 1998; Thorneycroft et al., 1999), and with other reports that OC users have lower saliva T levels than non-users (e.g., Schultheiss et al., 2003, 2005; van Anders and Watson, 2006b; Wirth and Schultheiss, 2007). This effect presumably results from decreased ovarian secretion of T owing to OC inhibition of hypophyseal secretion of FSH and LH. Moreover, OCs typically elevate serum sex hormone binding globulin (SHBG), whose binding action would effectively decrease free testosterone (e.g., Wiegratz et al., 1995; Coenen et al., 1996; Aden et al., 1998; Thorneycroft et al., 1999). The molecular size of SHBG is too large to allow entry into the salivary glands and it is therefore commonly supposed that what is measured in saliva reflects serum free testosterone. While this notion has been questioned (see Stanczyk, 2006), any reduction in serum free testosterone is likely to be reflected in reduced testosterone level in saliva. That women using OCs have lower T levels than non-users suggests the importance of treating OC use as an independent analytical variable in studies relating T levels to behavioral and/or psychological processes (e.g., Grammer et al., 2004; van Anders and Watson, 2006b; Schultheiss et al., 2003). Although this analytical “problem” can be prevented by excluding women using OCs and other forms of hormonal contraception from participant populations, such a practice is not practical in studies using small, well-defined groups such as athletic teams because OC use is common enough that eliminating users would reduce the power of statistical procedures to the point that it would be difficult to document clear and meaningful hormone/behavior relationships. Treating OC use as an independent analytical variable explicitly acknowledges the possible moderating effects of OC use on T levels and allows for an analysis of the potential
198
D.A. Edwards, J.L. O'Neal / Hormones and Behavior 56 (2009) 195–198
contribution of OC use to variability in the behavioral/psychological variables of interest in the study. In our view this should be a routine part of the design and analysis of any study of reproductive-age women where OC use by some participants is likely. Team athletic competition is associated with an increase in saliva T which presumably reflects an increase in the blood level of this hormone. This effect was clearly evident in each of the studies comprising this report (Edwards et al., 2006, 2007, and unpublished observations), and is equally evident when results for these studies are combined (Fig. 2). The effect is pronounced in OC users and nonusers (Fig. 2) and, whether analyzed by absolute change in T level or by percent change from before-competition value, is not significantly different for users and non-users. This effect is in accordance with what has been reported for college-age women rugby players (Bateup et al., 2002). A simulated team handball match does not increase saliva T (Filaire and Lac, 2000), suggesting (by comparison to the results reported here) that competition that lacks special meaning to the participants may be relatively ineffective in elevating T level. One recent report (Kivligan et al., 2005) failed to find an increase in T after a brief (6–8 min) head-to-head rowing competition in women, perhaps owing to the brevity and/or individual nature of the competitions. Schultheiss et al. (2003, p. 300) caution that because OCs decrease steroid release in the ovaries, OC use may interfere with normal endocrine responses to motivationally arousing stimuli. This will only be true in instances where the endocrine response is mediated by the secretion of pituitary gonadotropins, which seems not to be the case for the competition sports studied here. Indeed, although OC users typically have lower before- and after-competition T levels than nonusers, competition increases T in both users and non-users by nearly the same absolute amount. While the most proximal “cause” of this competition-related increase in T is not known, it is probably not the result of an increase in gonadotropin secretion since such an increase would presumably be prevented by OC use. Activation of the ovary via direct projections from the central nervous system (e.g., Gerendai et al., 2002) is a possible, although yet unproven, means by which psychological processes having to do with competition may influence the secretion of testosterone and (possibly) other ovarian hormones. The present study used the data provided by women soccer, volleyball, and softball players, and roller derby skaters. Considering teams individually, although OC use was associated with lower levels of saliva T for women in each sport, differences between users and nonusers were not statistically significant. When teams are combined in a single analysis, saliva T for users was significantly lower than for nonusers — an effect that clearly depends upon the increased statistical power provided by the greater number of participants included in the analysis. While studies with relatively small samples may not always show significant differences between OC users and non-users, our results suggest that OC use will contribute predictable variation to individual differences in saliva T, and OC use is likely to contribute to individual differences in measures that are causally related to T level. References Aden, U., Jung-Hoffman, C., Kuhl, H., 1998. A randomized cross-over study on various hormonal parameters of two triphasic oral contraceptives. Contraception 58, 75–81. Bateup, H.S., Booth, A., Shirtcliff, E.A., Granger, D.A., 2002. Testosterone, cortisol, and women's competition. Evol. Hum. Behav. 23, 181–192. Bhasin, S., 2005. Female androgen deficiency syndrome — an unproven hypothesis. J. Clin. Endocrinol. Metab. 90, 4970–4972. Bhasin, S., Cunningham, G.R., Haynes, F.J., Matsumoto, A.M., Snyder, P.J., Swerdloff, R.S., Montori, V.M., 2006. Testosterone therapy in adult men with androgen deficiency syndromes: an endocrine society clinical practice guideline. J. Clin. Endocrinol. Metab. 91, 1995–2010.
Burkitt, J., Widman, D., Saucier, D.M., 2007. Evidence for the influence of testosterone in the performance of spatial navigation in a virtual water maze in women but not in men. Horm. Behav. 51, 649–654. Coates, J.M., Herbert, J., 2008. Endogenous steroids and financial risk taking on a London trading floor. Proc. Natl. Acad. Sci. 104, 6167–6172. Coenen, C.M.H., Thomas, C.M.G., Borm, G.F., Hollandars, J.M.G., Rolland, R., 1996. Changes in androgens during treatment with four low-dose contraceptives. Contraception 53, 171–176. Dabbs Jr., J.M., Campbell, B.C., Gladue, B.A., Midgley, A.R., Navarro, M.A., Read, G.F., Susman, E.J., Swinkels, L.M.J.W., Worthman, C.M., 1995. Reliability of salivary testosterone measurements: a multicenter evaluation. Clin. Chem. 41, 1581–1584. Edwards, D.A., Wetzel, K., Wyner, D.R., 2006. Intercollegiate soccer: saliva cortisol and testosterone are elevated during competition, and testosterone is related to status and social connectedness with teammates. Physiol. Behav. 87, 135–143. Edwards, D.A., Waters, J., Weiss, A., Jarvis, A., 2007. Intercollegiate athletics: competition increases saliva testosterone in women soccer, volleyball, and softball players. In: Ardis, L.I. (Ed.), Testosterone Research Trends. Nova Science Publishers, Inc., Hauppage, New York, pp. 195–209. Filaire, E., Lac, G., 2000. Dehydroepiandrosterone (DHEA) rather than testosterone shows saliva androgen responses to exercise in elite female handball players. Int. J. Sports Med. 21, 17–20. Gerendai, I., Katalin, K., Halász, B., 2002. Supraspinal connections of the ovary: structural and functional aspects. Microsc. Res. Tech. 59, 474–483. Grammer, K., Renninger, L.-A., Fischer, B., 2004. Disco clothing, female sexual motivation, and relationship status: is she dressed to impress? J. Sex Res. 41, 66–74. Granger, D.A., Schwartz, E.B., Booth, A., Arentz, M., 1999. Salivary testosterone determination in studies of child health and development. Horm. Behav. 35, 18–27. Granger, D.A., Shirtcliff, E.A., Booth, A., Kivlighan, K.T., Schwartz, E.B., 2004. The “trouble” with salivary testosterone. Psychoneuroendocrinology 29, 1229–1240. Hatcher, R.A., Nelson, A., 2004. Combined hormonal contraceptive methods. In: Hatcher, R.A. (Ed.), Contraceptive Technology, eighteenth rev. ed. Ardent Media, New York, pp. 391–460. Kivligan, K.T., Granger, D.A., Booth, A., 2005. Gender differences in testosterone and cortisol response to competition. Psychoneuroendocrinology 30, 58–71. Lobo, R.A., 1994. New knowledge in the physiology of hormonal contraceptives. Am. J. Obstet. Gynecol. 170 (5 pt 2), 1499–1507. Mazur, A., Booth, A., 1998. Testosterone and dominance in men. Behav. Brain Sci. 21, 353–397. Mehta, P.H., Josephs, R.A., 2006. Testosterone change after losing predicts the decision to compete again. Horm. Behav. 50, 684–692. Santoro, N., Torrens, J., Crawford, S., Allsworth, J.E., Finkelstein, J.S., Gold, E.B., Korenman, S., Lasley, W.L., Luborsky, J.L., McConnell, D., Sowers, M.F., Weiss, G., 2005. J. Clin. Endocrinol. Metab. 90, 475–481. Schultheiss, O.C., Dargel, A., Rohde, W., 2003. Implicit motives and gonadal steroid hormones: effects of menstrual cycle phase, oral contraceptive use, and relationship status. Horm. Behav. 43, 293–301. Schultheiss, O.C., Wirth, M.M., Torges, C.M., Pang, J.S., Villacorta, M.A., Welsh, K.M., 2005. Effects of implicit power motivation on men's and women's implicit learning and testosterone changes after social victory or defeat. J. Pers. Soc. Psychol. 88, 174–188. Sellers, J.G., Mehl, M.R., Josephs, R.A., 2007. Hormones and personality: testosterone as a marker of individual differences. J. Res. Pers. 41, 126–138. Speroff, L., Darney, P., 2005. A Clinical Guide for Contraception, 4th ed. Lippincott Williams & Wilkins, Philadelphia, pp. 21–138. Spona, J., Feichtinger, W., Kinderman, Ch., Wünsch, C., Brill, K., 1996. Inhibition of ovulation by an oral contraceptive containing 100 μg levonorgestrel in combination with 20 μg ethinylestradiol. Contraception 54, 299–304. Stanczyk, F.Z., 2006. Measurement of androgens in women. Semin. Reprod. Med. 24, 78–85. Thorneycroft, I.H., Stanczyk, F.Z., Bradshaw, K.D., Ballagh, S.A., Nichols, M., Weber, M.E., 1999. Contraception 60, 255–262. van Anders, S.M., Watson, N.V., 2006a. Relationship status and testosterone in North American heterosexual and non-heterosexual men and women: cross-sectional and longitudinal data. Psychoneuroendocrinology 31, 715–723. van Anders, S.M., Watson, N.V., 2006b. Menstrual cycle irregularities are associated with testosterone levels in healthy premenopausal women. Am. J. Hum. Biol. 18, 841–844. van Anders, S.M., Hamilton, L.D., Schmidt, N., Watson, N.V., 2007. Associations between testosterone secretion and sexual activity in women. Horm. Behav. 51, 477–482. van Honk, J., Tuiten, A., Verbaten, R., van den Hout, M., Koppeschaar, H., Thijssen, J., de Haan, E., 1999. Correlations among salivary testosterone, mood, and selective attention to threat in humans. Horm. Behav. 36, 17–24. Welling, L.L.M., Jones, B.C., DeBruine, L.M., Conway, C.A., Law Smith, M.J., Little, A.C., Feinberg, D.R., Sharp, M.A., Al-Dujaili, E.A.S., 2007. Raised salivary testosterone in women is associated with increased attraction to masculine faces. Horm. Behav. 52, 156–161. Wiegratz, I., Jung-Hoffman, C., Kuhl, H., 1995. Effect of two oral contraceptives containing ethinylestradiol and gestodene or norgestimate upon androgen parameters and serum binding proteins. Contraception 51, 341–346. Wirth, M.M., Schultheiss, O.C., 2007. Basal testosterone moderates responses to anger faces in humans. Physiol. Behav. 90, 496–505.
Contents lists available at ScienceDirect
Hormones and Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y h b e h
Oral contraceptives decrease saliva testosterone but do not affect the rise in testosterone associated with athletic competition David A. Edwards ⁎, J. Laurel O'Neal Department of Psychology, Emory University, Atlanta, GA 30322, USA
a r t i c l e
i n f o
Article history: Received 30 October 2008 Revised 10 January 2009 Accepted 14 January 2009 Available online 31 January 2009 Keywords: Saliva testosterone Oral contraceptives Athletic competition Hormone contraception
a b s t r a c t Women athletes from intercollegiate soccer, volleyball, and softball teams, and women skaters from a team competing in an amateur roller derby league, contributed saliva samples before warm-up and immediately after the completion of one or more sanctioned competitions. Women using oral contraceptives (OCs, n = 29) had a significantly lower mean level of saliva testosterone (T) than non-users (n = 51). Thus, OCs contribute predictable variation to individual differences in saliva T, and OC use is likely to contribute to individual differences in measures of psychological processes and/or behavior which are causally related to individual differences in circulating testosterone. Most of the women (n = 68) played during one or more of the competitions for which they contributed saliva samples. Whether for soccer, volleyball, softball, or roller derby, competition was associated with a robust increase in saliva T. Although OC users had significantly lower saliva T levels than non-users before and after-competition, both users and non-users showed virtually the same increase in saliva T over the course of competition. While the most proximal cause of this increase is not known, it is probably not the result of an increase in gonadotropin (GTH) secretion since an increase in GTH secretion would presumably be prevented by OC use. © 2009 Elsevier Inc. All rights reserved.
Introduction Testosterone is secreted primarily in the testes of males and the ovaries of females. In men (Bhasin et al., 2006) and, less certainly, women (Bhasin, 2005; Santoro et al., 2005), this hormone plays important roles in health and well-being, the maintenance of muscle mass and bone density, sexual desire and other psychological processes. The ability to measure testosterone non-invasively in saliva has made it possible to repeatedly sample individuals in widely varying settings, allowing for studies relating testosterone to such things as mood (van Honk et al., 1999), clothing choice (Grammer et al., 2004), cognitive abilities (Burkitt et al., 2007), sexual activity (van Anders et al., 2007), financial risk-taking (Coates and Herbert, 2008), attraction to masculine faces (Welling et al., 2007), relationship status (van Anders and Watson, 2006a), dominance/personality (Mazur and Booth, 1998; Schultheiss et al., 2005; Sellers et al., 2007), competitive response to defeat (Mehta and Josephs, 2006), athletic competition (e.g., Bateup et al., 2002; Kivligan et al., 2005) and status with teammates (Edwards et al., 2006). Despite a number of caveats (see the thorough review by Granger et al., 2004), it seems certain that measurement of testosterone from saliva samples will continue to be widely used in studies looking to understand how testosterone mediates psychological processing and behavior in men and women.
⁎ Corresponding author. Fax: +1 404 727 0372. E-mail address: [email protected] (D.A. Edwards). 0018-506X/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.yhbeh.2009.01.008
Oral contraceptives (OCs) are currently used by over 100 million women world-wide (Hatcher and Nelson, 2004). Modern formulations available in the United States contain a combination of a synthetic estrogen (either ethynyl estradiol or mestranol) and a progestin (Hatcher and Nelson, 2004). Their effectiveness is derived, at least in part, by blocking secretion of gonadotropins (FSH and LH) responsible for follicular development and ovulation (Lobo, 1994; Speroff and Darney, 2005). OCs in effect substitute low doses of exogenously-administered synthetic hormones for the naturally occurring levels of estrogen and progesterone that would otherwise vary systematically as part of a normal menstrual cycle. It is now abundantly clear that OC use decreases serum testosterone (e.g., Wiegratz et al., 1995; Coenen et al., 1996; Spona et al., 1996; Aden et al., 1998; Thorneycroft et al., 1999). It is perhaps therefore surprising that, with a few notable exceptions (e.g., Schultheiss et al., 2003; van Anders and Watson, 2006b; Wirth and Schultheiss, 2007), studies of reproductive-age women that assay testosterone from saliva samples often do not inquire about/analyze for effects of OC use by participants. Two computerized searches (PubMed and PsycINFO) of the English language research literature using “saliva and testosterone” as the search phrase turned up 35 different studies published in the 5-year span from 2003–2007 (inclusive) with women of reproductive age and presumed heterosexual orientation. For 12 of these studies, women using hormone contraception were either not a part of or were explicitly excluded from research participation. For 18 of the remaining 23 studies, experimenters either did not ask about participant use of hormone
196
D.A. Edwards, J.L. O'Neal / Hormones and Behavior 56 (2009) 195–198
contraception (9 studies) or asked but did not analyze for the possible effect of hormone contraception use on saliva testosterone (9 studies). Only 5 studies asked about hormone contraception use by participants and analyzed for effects on testosterone level — in four of these (Schultheiss et al., 2003, 2005; van Anders and Watson, 2006b; Wirth and Schultheiss, 2007) OC users had significantly lower saliva testosterone levels than nonusers. From 1999 through 2007, one of us (DE) supervised the conduct of five different studies with women athletes. As part of the consent process participants were specifically asked to indicate their use (or not) of OCs and/or other forms of hormone contraception. Procedures for saliva collection were the same for each of the studies. In all cases, samples were obtained in the afternoon or early evening and all samples were assayed for testosterone in the same laboratory using the same radioimmunoassay procedures. Team athletic competition is typically associated with an increase in saliva testosterone (e.g., Bateup et al., 2002; Edwards et al., 2006, 2007). For this article, we combined the results for these studies to give a sufficient number of women to analyze (and here report) the relationship between OC use and saliva testosterone with special attention to the possible effects of OC use on competition-related changes in levels of this hormone. Method Participants Women contributing saliva samples were participants in one of five studies conducted from 1999 through 2007. Four of the studies were conducted with varsity athletes from Emory University ranging in age from 18 to 22 years and included members of the 1999 women's soccer team (n = 16), 2002 and 2005 women's volleyball team (n = 14 and n = 19, respectively), and the 2004 women's softball team (n = 16). A fifth study was conducted with the members (n = 15) of the Toxic Shocks, one of four Atlanta-area teams in an amateur roller derby league. Roller derby is a contact sport, and competitions involve two teams (5 to a side) rollerskating against each other around an oval track. The Toxic Shocks ranged in age from 22 to 37 years. Procedures for each study were approved by the Emory University Institutional Review Board. All individuals were informed about the purpose of the study and about the method and frequency of saliva sampling for the study in which they would be participating. Each woman gave written informed consent. Although not queried about their undoubtedly varied use of prescription/non-prescription medications, as part of the consent procedure women were specifically asked about hormone contraceptive use. For the study with soccer players each woman was asked to respond “yes” or “no” to the following question: “Are you currently using oral contraceptives?” For all the other studies, women were asked to respond “yes” or “no” to this question and another: “Are you currently using any injected, implanted, or patch-delivered hormone contraceptive?”
Samples were assayed in either duplicate or triplicate and the mean of the values for any given sample was taken as the value to represent T level for that sample. Intrassay and interassay coefficients of variation varied across the five studies, but in every study were less than 10% and 20% respectively. Oral contraceptive use and the effect of athletic competition on saliva T Each of the studies was designed to determine the effect of team athletic competition on saliva T in women. To this end, saliva was collected before and immediately after one or more coach-selected home competitions. Collection before each of the competitions was done immediately prior to warm-up which began approximately 1 h before the start of play. This procedure was repeated immediately after the end of the competition so that samples were received within 15 min after the completion of the game or, in the case of volleyball players, a best-of-5-games match. For purpose of the present study, we took the T value for each individual's before warm-up saliva sample to represent the beforecompetition value for that individual for that competition. For soccer players we used values from samples associated with a single 1999 intercollegiate game played in mid-November — a 5–0 victory. For the 2002 volleyball players we averaged values for saliva samples for two intercollegiate matches played on consecutive days in mid-October — each a 3-games-to-1 victory. For the 2005 volleyball players we averaged sample values for three different intercollegiate matches played over the course of a weekend in late October — each one a loss. For the softball players we averaged sample values for five intercollegiate games (3 wins and 2 losses) played from late February through early April of the 2004 season. For the roller-derby team we used samples obtained for a single league competition (a loss) in late July, 2007. In women, team athletic competition is associated with an increase in saliva T (Edwards et al., 2006, 2007). To determine how this effect might be influenced by OC use, we first selected all of the individuals who played in the competitions for which we had saliva samples (see above). Next, we calculated the mean before-competition T value and the mean after-competition T value for those competitions in which the individual played, and took these values as the best estimate of the individual's before- and after-competition saliva T level. We then used these values to determine the effect of OC use on the hormonal response to intercollegiate competition. Statistical analyses The SPSS statistical package for personal computer was used for the calculation of Chi-square test of independence, independent and correlated t-tests (two-tailed), and one-way and mixed-design analysis of variance (ANOVA). In all cases, P b 0.05 was required for statistical significance. Results
Saliva samples and hormone assay
Oral contraceptive use by participants
For every study, participants were provided with a piece of sugarfree gum (Trident®, original flavor) to stimulate saliva production and a 20 ml plastic vial which they were asked to fill to a 5 ml line marked on the side. Samples were stored at − 20 °C within 30 min after collection and the frozen samples were later sent to the Biomarkers laboratory of the Yerkes Primate Center in Atlanta, Georgia for hormone assay. Samples were obtained at different times for different studies between noon and 2100. Saliva testosterone was assayed using a modification (Granger et al., 1999) of the Diagnostic Systems Laboratory (DSL, Webster, TX) radioimmunoassay kit with a range of analytical sensitivity of 2–500 pg/ml for test volumes of 200 μl.
Twenty-nine of the 80 participants reported using oral contraceptives; none reported any injected, implanted, or patch-delivered hormone contraception. Oral contraception use across the five teams ranged from 31–44% of the players and was not significantly different according to team (χ2 (4) = 0.54, ns). Oral contraceptive use and before- and after-competition testosterone levels Before-competition saliva T values were within the range expected for reproductive age women (e.g., Dabbs et al., 1995). For every team,
D.A. Edwards, J.L. O'Neal / Hormones and Behavior 56 (2009) 195–198
women using oral contraceptives had, on average, lower beforecompetition T values than women not using oral contraceptives (Fig. 1), but differences between users and non-users analyzed separately for each team were not significantly different for any team. Differences in mean before-competition T levels for the different teams were not significant by one-way ANOVA, so for purposes of analysis and presentation we combined values for the five teams creating two groups — women using oral contraceptives (n = 29) and women not using oral contraceptives (n = 51). Beforecompetition saliva T means for these two groups are shown in Fig. 1 (insert). Mean saliva T for OC-users is significantly lower than for non-users (t(78) = 2.3, P b 0.03.). Sixty-eight of the 80 women included in the analysis above played during one or more of the competitions for which they contributed saliva samples. Whether for soccer, volleyball, or softball, intercollegiate competition was associated with a robust increase in saliva T (Edwards et al., 2006, 2007) which was not affected by the outcome of the competition (see Edwards et al 2007 for more discussion). A similar competition-related rise in saliva testosterone was seen in roller derby skaters (unpublished observations). To determine if this rise in T level is influenced by OC use, we considered only the 68 women who played in at least one game (soccer and softball) or match (volleyball and roller derby). Before- and after-competition T means are shown for OC users and non-users in Fig. 2. Among the women who played, OC-users had significantly lower before- (t(66) = 2.0, P b 0.05) and after-competition (t(66) = 2.3, P b 0.03) T values than non-users. Mixed-design ANOVA with OC use as the between-groups variable and saliva T (before-competition and after-competition) as the within-groups (repeated measures) variable gives a significant effect of competition on T level (F(1,66) = 54.9, P b 0.01) and a significant effect of OC use (F(1,66) = 5.6, P b 0.03) on saliva T. The interaction between OC use and testosterone is not significant owing to the fact that in both OC users and non-users, saliva T increased from before- to after-competition by approximately the same amount (see Fig. 2). For a supplemental analysis of OC use as it might affect the competition-related increase in T level, we calculated the percent change in T level relative to pre-competition baseline for each player. From before warm-up to the end of competition, saliva T levels increased by an average of 50.5% for women not using oral contraceptives and by an average of 70.1% for
Fig. 1. Mean before-competition testosterone levels for women oral contraceptive (OC) users and non-users according to individual team and (insert) all teams combined. Sample sizes (n) for each group are shown. When teams are combined for analysis, the mean before-competition testosterone level for OC users is significantly lower than the comparable mean for non-users.
197
Fig. 2. Mean before- and after-competition testosterone levels for women oral contraceptive (OC) users and non-users. Groups only include individuals who played in the competitions for which they contributed saliva samples. Sample sizes for each group are shown (n). As noted in the figure, mean before- and after-competition testosterone levels for OC users are significantly lower than the comparable means for non-users.
women using oral contraceptives. These values are not significantly different (t(66) = −0.72). Discussion Saliva T is typically correlated with total serum T in blood (Granger et al., 2004). In the present study, OC users had significantly lower saliva T values than non-users which we presume reflect their lower serum levels of this same hormone. This result is consonant with observations that OC use is associated with lower serum levels of free and total T (e.g., Wiegratz et al., 1995; Coenen et al., 1996; Spona et al., 1996; Aden et al., 1998; Thorneycroft et al., 1999), and with other reports that OC users have lower saliva T levels than non-users (e.g., Schultheiss et al., 2003, 2005; van Anders and Watson, 2006b; Wirth and Schultheiss, 2007). This effect presumably results from decreased ovarian secretion of T owing to OC inhibition of hypophyseal secretion of FSH and LH. Moreover, OCs typically elevate serum sex hormone binding globulin (SHBG), whose binding action would effectively decrease free testosterone (e.g., Wiegratz et al., 1995; Coenen et al., 1996; Aden et al., 1998; Thorneycroft et al., 1999). The molecular size of SHBG is too large to allow entry into the salivary glands and it is therefore commonly supposed that what is measured in saliva reflects serum free testosterone. While this notion has been questioned (see Stanczyk, 2006), any reduction in serum free testosterone is likely to be reflected in reduced testosterone level in saliva. That women using OCs have lower T levels than non-users suggests the importance of treating OC use as an independent analytical variable in studies relating T levels to behavioral and/or psychological processes (e.g., Grammer et al., 2004; van Anders and Watson, 2006b; Schultheiss et al., 2003). Although this analytical “problem” can be prevented by excluding women using OCs and other forms of hormonal contraception from participant populations, such a practice is not practical in studies using small, well-defined groups such as athletic teams because OC use is common enough that eliminating users would reduce the power of statistical procedures to the point that it would be difficult to document clear and meaningful hormone/behavior relationships. Treating OC use as an independent analytical variable explicitly acknowledges the possible moderating effects of OC use on T levels and allows for an analysis of the potential
198
D.A. Edwards, J.L. O'Neal / Hormones and Behavior 56 (2009) 195–198
contribution of OC use to variability in the behavioral/psychological variables of interest in the study. In our view this should be a routine part of the design and analysis of any study of reproductive-age women where OC use by some participants is likely. Team athletic competition is associated with an increase in saliva T which presumably reflects an increase in the blood level of this hormone. This effect was clearly evident in each of the studies comprising this report (Edwards et al., 2006, 2007, and unpublished observations), and is equally evident when results for these studies are combined (Fig. 2). The effect is pronounced in OC users and nonusers (Fig. 2) and, whether analyzed by absolute change in T level or by percent change from before-competition value, is not significantly different for users and non-users. This effect is in accordance with what has been reported for college-age women rugby players (Bateup et al., 2002). A simulated team handball match does not increase saliva T (Filaire and Lac, 2000), suggesting (by comparison to the results reported here) that competition that lacks special meaning to the participants may be relatively ineffective in elevating T level. One recent report (Kivligan et al., 2005) failed to find an increase in T after a brief (6–8 min) head-to-head rowing competition in women, perhaps owing to the brevity and/or individual nature of the competitions. Schultheiss et al. (2003, p. 300) caution that because OCs decrease steroid release in the ovaries, OC use may interfere with normal endocrine responses to motivationally arousing stimuli. This will only be true in instances where the endocrine response is mediated by the secretion of pituitary gonadotropins, which seems not to be the case for the competition sports studied here. Indeed, although OC users typically have lower before- and after-competition T levels than nonusers, competition increases T in both users and non-users by nearly the same absolute amount. While the most proximal “cause” of this competition-related increase in T is not known, it is probably not the result of an increase in gonadotropin secretion since such an increase would presumably be prevented by OC use. Activation of the ovary via direct projections from the central nervous system (e.g., Gerendai et al., 2002) is a possible, although yet unproven, means by which psychological processes having to do with competition may influence the secretion of testosterone and (possibly) other ovarian hormones. The present study used the data provided by women soccer, volleyball, and softball players, and roller derby skaters. Considering teams individually, although OC use was associated with lower levels of saliva T for women in each sport, differences between users and nonusers were not statistically significant. When teams are combined in a single analysis, saliva T for users was significantly lower than for nonusers — an effect that clearly depends upon the increased statistical power provided by the greater number of participants included in the analysis. While studies with relatively small samples may not always show significant differences between OC users and non-users, our results suggest that OC use will contribute predictable variation to individual differences in saliva T, and OC use is likely to contribute to individual differences in measures that are causally related to T level. References Aden, U., Jung-Hoffman, C., Kuhl, H., 1998. A randomized cross-over study on various hormonal parameters of two triphasic oral contraceptives. Contraception 58, 75–81. Bateup, H.S., Booth, A., Shirtcliff, E.A., Granger, D.A., 2002. Testosterone, cortisol, and women's competition. Evol. Hum. Behav. 23, 181–192. Bhasin, S., 2005. Female androgen deficiency syndrome — an unproven hypothesis. J. Clin. Endocrinol. Metab. 90, 4970–4972. Bhasin, S., Cunningham, G.R., Haynes, F.J., Matsumoto, A.M., Snyder, P.J., Swerdloff, R.S., Montori, V.M., 2006. Testosterone therapy in adult men with androgen deficiency syndromes: an endocrine society clinical practice guideline. J. Clin. Endocrinol. Metab. 91, 1995–2010.
Burkitt, J., Widman, D., Saucier, D.M., 2007. Evidence for the influence of testosterone in the performance of spatial navigation in a virtual water maze in women but not in men. Horm. Behav. 51, 649–654. Coates, J.M., Herbert, J., 2008. Endogenous steroids and financial risk taking on a London trading floor. Proc. Natl. Acad. Sci. 104, 6167–6172. Coenen, C.M.H., Thomas, C.M.G., Borm, G.F., Hollandars, J.M.G., Rolland, R., 1996. Changes in androgens during treatment with four low-dose contraceptives. Contraception 53, 171–176. Dabbs Jr., J.M., Campbell, B.C., Gladue, B.A., Midgley, A.R., Navarro, M.A., Read, G.F., Susman, E.J., Swinkels, L.M.J.W., Worthman, C.M., 1995. Reliability of salivary testosterone measurements: a multicenter evaluation. Clin. Chem. 41, 1581–1584. Edwards, D.A., Wetzel, K., Wyner, D.R., 2006. Intercollegiate soccer: saliva cortisol and testosterone are elevated during competition, and testosterone is related to status and social connectedness with teammates. Physiol. Behav. 87, 135–143. Edwards, D.A., Waters, J., Weiss, A., Jarvis, A., 2007. Intercollegiate athletics: competition increases saliva testosterone in women soccer, volleyball, and softball players. In: Ardis, L.I. (Ed.), Testosterone Research Trends. Nova Science Publishers, Inc., Hauppage, New York, pp. 195–209. Filaire, E., Lac, G., 2000. Dehydroepiandrosterone (DHEA) rather than testosterone shows saliva androgen responses to exercise in elite female handball players. Int. J. Sports Med. 21, 17–20. Gerendai, I., Katalin, K., Halász, B., 2002. Supraspinal connections of the ovary: structural and functional aspects. Microsc. Res. Tech. 59, 474–483. Grammer, K., Renninger, L.-A., Fischer, B., 2004. Disco clothing, female sexual motivation, and relationship status: is she dressed to impress? J. Sex Res. 41, 66–74. Granger, D.A., Schwartz, E.B., Booth, A., Arentz, M., 1999. Salivary testosterone determination in studies of child health and development. Horm. Behav. 35, 18–27. Granger, D.A., Shirtcliff, E.A., Booth, A., Kivlighan, K.T., Schwartz, E.B., 2004. The “trouble” with salivary testosterone. Psychoneuroendocrinology 29, 1229–1240. Hatcher, R.A., Nelson, A., 2004. Combined hormonal contraceptive methods. In: Hatcher, R.A. (Ed.), Contraceptive Technology, eighteenth rev. ed. Ardent Media, New York, pp. 391–460. Kivligan, K.T., Granger, D.A., Booth, A., 2005. Gender differences in testosterone and cortisol response to competition. Psychoneuroendocrinology 30, 58–71. Lobo, R.A., 1994. New knowledge in the physiology of hormonal contraceptives. Am. J. Obstet. Gynecol. 170 (5 pt 2), 1499–1507. Mazur, A., Booth, A., 1998. Testosterone and dominance in men. Behav. Brain Sci. 21, 353–397. Mehta, P.H., Josephs, R.A., 2006. Testosterone change after losing predicts the decision to compete again. Horm. Behav. 50, 684–692. Santoro, N., Torrens, J., Crawford, S., Allsworth, J.E., Finkelstein, J.S., Gold, E.B., Korenman, S., Lasley, W.L., Luborsky, J.L., McConnell, D., Sowers, M.F., Weiss, G., 2005. J. Clin. Endocrinol. Metab. 90, 475–481. Schultheiss, O.C., Dargel, A., Rohde, W., 2003. Implicit motives and gonadal steroid hormones: effects of menstrual cycle phase, oral contraceptive use, and relationship status. Horm. Behav. 43, 293–301. Schultheiss, O.C., Wirth, M.M., Torges, C.M., Pang, J.S., Villacorta, M.A., Welsh, K.M., 2005. Effects of implicit power motivation on men's and women's implicit learning and testosterone changes after social victory or defeat. J. Pers. Soc. Psychol. 88, 174–188. Sellers, J.G., Mehl, M.R., Josephs, R.A., 2007. Hormones and personality: testosterone as a marker of individual differences. J. Res. Pers. 41, 126–138. Speroff, L., Darney, P., 2005. A Clinical Guide for Contraception, 4th ed. Lippincott Williams & Wilkins, Philadelphia, pp. 21–138. Spona, J., Feichtinger, W., Kinderman, Ch., Wünsch, C., Brill, K., 1996. Inhibition of ovulation by an oral contraceptive containing 100 μg levonorgestrel in combination with 20 μg ethinylestradiol. Contraception 54, 299–304. Stanczyk, F.Z., 2006. Measurement of androgens in women. Semin. Reprod. Med. 24, 78–85. Thorneycroft, I.H., Stanczyk, F.Z., Bradshaw, K.D., Ballagh, S.A., Nichols, M., Weber, M.E., 1999. Contraception 60, 255–262. van Anders, S.M., Watson, N.V., 2006a. Relationship status and testosterone in North American heterosexual and non-heterosexual men and women: cross-sectional and longitudinal data. Psychoneuroendocrinology 31, 715–723. van Anders, S.M., Watson, N.V., 2006b. Menstrual cycle irregularities are associated with testosterone levels in healthy premenopausal women. Am. J. Hum. Biol. 18, 841–844. van Anders, S.M., Hamilton, L.D., Schmidt, N., Watson, N.V., 2007. Associations between testosterone secretion and sexual activity in women. Horm. Behav. 51, 477–482. van Honk, J., Tuiten, A., Verbaten, R., van den Hout, M., Koppeschaar, H., Thijssen, J., de Haan, E., 1999. Correlations among salivary testosterone, mood, and selective attention to threat in humans. Horm. Behav. 36, 17–24. Welling, L.L.M., Jones, B.C., DeBruine, L.M., Conway, C.A., Law Smith, M.J., Little, A.C., Feinberg, D.R., Sharp, M.A., Al-Dujaili, E.A.S., 2007. Raised salivary testosterone in women is associated with increased attraction to masculine faces. Horm. Behav. 52, 156–161. Wiegratz, I., Jung-Hoffman, C., Kuhl, H., 1995. Effect of two oral contraceptives containing ethinylestradiol and gestodene or norgestimate upon androgen parameters and serum binding proteins. Contraception 51, 341–346. Wirth, M.M., Schultheiss, O.C., 2007. Basal testosterone moderates responses to anger faces in humans. Physiol. Behav. 90, 496–505.