- Email: [email protected]
Age at first intercourse is inversely related to female cortisol stress reactivity
Psychoneuroendocrinology 27 (2002) 933–943 www.elsevier.com/locate/psyneuen
Age at first intercourse is inversely related to female cortisol stress reactivity Stuart Brody a,b,∗ a
Center for Psychobiological and Psychosomatic Research, University of Trier, Karl-Marx Strasse 94, 54290 Trier, Germany b Institute of Medical Psychology, University of Tu¨bingen, Germany Received 9 August 2001; received in revised form 15 January 2002; accepted 15 January 2002
Abstract The relationship between age at first sexual intercourse and salivary cortisol stress reactivity (to the Trier Social Stress Test; TSST; consisting of public speaking and mental arithmetic) was examined in healthy subjects (43 females and 36 males; ages 19–38). Women reporting earlier first intercourse had less intense cortisol increases in response to the stressor (a nonsignificant trend was observed for males), and faster recovery from the stressor. Results were not confounded by age, oral contraceptive use, depression scores, smoking status, or body mass index. It is concluded that earlier first intercourse is associated with less reactivity to and faster recovery from stress as indexed by this endocrine measure. Results are discussed in terms of genetic and psychological influences on first intercourse and implications for coping with interpersonal stress. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Coitus; Sexual behavior; Cortisol; Stress reactivity; Stress recovery
1. Introduction Age at first sexual (penile–vaginal) intercourse (AFI) has a significant genetic loading (Martin et al., 1977; Dunne et al., 1997), and part of this mechanism appears
∗
Center for Psychobiological and Psychosomatic Research, University of Trier, Karl-Marx Strasse 94, 54290 Trier, Germany. Fax: +1-561-431-3114. E-mail address: [email protected] (S. Brody). 0306-4530/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 3 0 6 - 4 5 3 0 ( 0 2 ) 0 0 0 0 7 - 0
934
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
to involve genes for dopamine D1 and D2 receptors (Miller et al., 1999). Women’s earlier AFI was predicted by better motor skills at age 5, a domineering and mature personality at age 9 (Udry et al., 1995), earlier menarche (Udry and Cliquet, 1982), adolescent perception of earlier autonomy and physical maturity, and lack of restraint (Rosenthal et al., 1999). AFI is associated with lifelong sexual benefits as well: women’s earlier AFI is associated with greater coital orgasmic capacity (Raboch and Bartak, 1983), and in one sample (of unknown representativeness) of 64-year-old men, the sexually functional were differentiated from the nonfunctional by earlier AFI (Vallery-Masson et al., 1981). In several studies, AFI was associated inversely with adult intercourse frequency (reviewed in Brody (1997). Both AFI (in some studies of women; Paul et al., 2000) and cortisol response to stress (Seeman, 1995) have been found to be related inversely to self-esteem. The domineering and autonomous features noted above to be associated with earlier AFI have a parallel in the finding that social dominance and internal locus of control were associated with lower cortisol reactions (Pruessner et al., 1997) to the Trier Social Stress Test (TSST). In addition to genetic influences and behavioral manifestations of dopaminergic tone influencing cortisol response to social stress, cortisol responses may influence dopamine function. In laboratory animals, treatment with cortisol (at levels simulating a prolonged stress response) impaired dopamine-dependent prefrontal cortical functions (Lyons et al., 2000). Similarly, chronic psychosocial stress reduced the density of dopamine transporters in the caudate nucleus and putamen, and also reduced serum testosterone and testicular weight (Isovich et al., 2000). One may conjecture that chronic cortisol elevations associated with greater stress response could also interfere (or have interfered) with the dopamine-related initiation of sexual behavior. There are several pathways by which younger AFI might reduce stress response: earlier AFI could be a marker of a genetic predisposition against stress response, earlier AFI could have interpersonal and intrapsychic consequences leading to less stress response, and/or earlier AFI could be a precursor of more frequent intercourse, which results in less stress response. For the current study, the relationship between AFI and cortisol response to psychological stress was examined in a sample of healthy adults. To investigate one possible mechanism for differences in cortisol response (modification of adrenal sensitivity), the cortisol response to ACTH stimulation was also examined.
2. Method 2.1. Participants The study was embedded in a randomized placebo-controlled clinical trial of highdose ascorbic acid on stress response (Brody et al., 2002). Approximately half of the subjects in the current report consumed ascorbic acid supplements for the clinical trial, but trial group had no effect on either ACTH stimulation or overall cortisol
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
935
stress response (Brody et al., 2002), nor did it have any significant effect in any of the models herein, so to preserve degrees of freedom this variable was not included in the present analyses. Subjects were recruited for the trial from advertisements posted at the university, at a technical college, and at various shops and sports centers, and screened by telephone per the exclusion criteria (current medication except oral contraception; vitamin supplement use during the preceding three months; current illnesses or pregnancy; prior experience with the TSST stressor; age over 40 or under 19; history of psychiatric disturbance; ulcer, cancer, asthma, cardiovascular, endocrine, or neurological disorders; body mass index ⬎ 34 kg/m2, or indications of enhanced study protocol noncompliance risk), and then told the general details of the study. The subjects received a second screening as part of the brief medical examination at the first laboratory session. In the clinical trial in which this study was embedded (Brody et al., 2002), 12 of the original 120 subjects were excluded: four who reported unusual stress earlier on the TSST day (thereby precluding valid resting baseline values), four who were known to members of the TSST panel (thereby modifying the stress), one who was found to have recently taken corticosteroids as part of another study, one for suspected substance use the day of the TSST, one for noncompliance with the clinical trial dosing protocol, and one (in the placebo group) for fainting during the TSST. For the present study, subjects were also excluded if they scored at or above five (the 86th percentile in our sample) on the Lie scale of the German version of the Eysenck Personality Inventory (Eysenck, 1974). Excluding subjects with a tendency to misrepresent behavior that may be seen as socially undesirable is methodologically crucial (Brody, 1997; Brody et al., 2000). As per the recommendation of an anonymous reviewer, eight male and three female outlier subjects displaying a large (ⱖ6 nmol/L, see below) drop in salivary cortisol directly after the stressor (or from before to the time of the stressor) were excluded from the present analyses. The characteristics of the analyzed subjects by AFI group memberships are presented in Table 1. The overall mean and median AFI was 18 (range 14–33; plus two female virgins). Males reported an older mean AFI than females (19.3 vs. 17.2, t for unequal variances = 3.2, p = .003), were older (mean age 25.7 vs. 23.8, t = 2.0, p = .048), and had greater body mass index (23.8 vs. 22.4, t = 2.3, p = .02), but did not differ on the other variables listed in the Table. Using the median split of AFI (on intrasex comparisons), males with younger AFI were younger (mean age 23.7 vs. 27.4, t = 2.7, p = .01), reported more intercourse days per month (transformed from the 14 day diaries; 7.8 vs. 5.0, t = 2.5, p = .016), and both males and females with younger AFI were more likely to report being current tobacco smokers (both chi-square = 4.4, p = .04). AFI correlated significantly inversely with the number of penile–vaginal intercourse days in the previous 14 days for females (r = –.33, p = .03) but only marginally for males (r = –.29, p = .09).
936
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
Table 1 Characteristics of subjects by age of first intercourse (age-specific median split:⬍18 vs. ⱖ18 or never female, ⬍19 vs. ⱖ19 or never male) group Younger age at first intercourse (⬍19 male, ⬍18 female)
Older age at first intercourse (ⱖ19 or never male, ⱖ18 or never female)
Age (mean, SD)
23.7, 3.4 male 22.7, 2.1 female
27.4, 4.5 male 25.1, 5.7 female
Sex
16 male 24 female
20 male 19 female
Education
11 Some university, 1 University graduate, 4 other male; 3 High School (Abitur), 20 some university, 1 University graduate female
1 High School (Abitur), 13 some university, 5 University graduate, 1 other male; 1 High School (Abitur), 14 some university, 3 University graduate, 1 other female
Current tobacco smokers
12 male 14 female
8 male 5 female
Oral contraceptive users
17
12
Body mass index (kg/m2; mean, SD)
22.9, 2.4 male 22.1, 2.3 female
24.4, 3.1 male 22.7, 2.8 female
Age at first intercourse (mean, SD)
16.9, 1.1 male 15.9, 0.9 female
21.4, 3.6 male 18.8, 0.9 female
Beck Depression Score (mean, SD)
7.4, 6.5 male 4.5, 2.8 female
5.9, 4.2 male 5.8, 5.5 female
Eysenck Extraversion score (mean, SD)
14.6, 3.1 male 14.3, 3.7 female
12.4, 4.8 male 14.2, 4.9 female
Eysenck Neuroticism score (mean, SD)
8.0, 5.6 male 8.9, 4.9 female
8.3, 4.7 male 9.6, 4.4 female
Menstrual phase (N)
days 1–7: 7 days 8–20: 11 days 21–27: 6 missing: 0
days 1–7: 7 days 8–20: 4 days 21–27: 7 missing: 1
2.2. Measures 2.2.1. Cortisol Salivary cortisol was collected by means of chewing on plastic-sheathed cotton, which was returned to a sterile plastic holder (Hellhammer et al., 1987), and analyzed with a time-resolved immunoassay (Dressendo¨ rfer et al., 1992).
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
937
After the blood for plasma cortisol response to ACTH provocation was drawn, it was placed on ice and then centrifuged, and the plasma was pipetted and frozen at –20 ° C. Plasma cortisol levels were determined by chemoluminescence using a commercial kit (Nichols Cortisol 100T kit; San Juan Capistrano, California). 2.2.2. Psychosocial measures Subjects completed questionnaires including the German versions of the Eysenck Lie Scale (Eysenck, 1974) and Beck Depression Inventory (Hautzinger et al., 1994) and our own questionnaire which included questions on AFI (defined as penis in vagina), tobacco smoking status, menstrual cycle phase (subjects were asked the date their last menstruation began), and oral contraceptive use. The subjects also completed a 14-day diary, which included whether penile–vaginal intercourse had occurred on each day. 2.3. Laboratory tasks The psychological stressor consisted of the TSST (Kirschbaum et al., 1993), a standardized psychological stress-induction technique consisting of being told (following a “–10 min” salivary cortisol measurement) that one will have to give a five-minute speech to an unknown panel on personal suitability for a job in the subject’s field of interest, followed by five minutes of mental arithmetic performed out loud. After solitary preparation, the “–1 min” measurement is taken (reflecting the stress of preparation and anticipation of the actual stressor), and the subject enters the TSST room containing the panel of observers (one of each sex) and an obvious video camera and microphone. Pauses during the speech are dealt with by being reminded of remaining time. After 5 min, the task shifts to performing serial subtractions of 17 starting at 2023, with errors resulting in being required to return to the beginning. Immediately afterwards, the “+1 min” measurement is taken, and the subject leaves to the waiting area, where the “+10, +20, +30, and +40 min” measurements are taken in turn. 2.4. Procedure All laboratory sessions were conducted in the mid-afternoon, at least one hour after eating (there was no eating or smoking during the laboratory sessions). After completing questionnaires for at least 10 min, the subjects participated in the TSST. After the last (“+40 min”) measurement, subjects had a cannula inserted, blood was drawn for baseline plasma cortisol levels, followed by an injection of 1 µg of ACTH (the 1 µg dose is more sensitive and at least as safe as higher doses; Beishuizen et al. (2000)bib:Beishuizen, 2000; Tordjman et al. (2000). After 45 and 60 min, blood was again drawn for determination of plasma cortisol response to the ACTH provocation. The cannula was removed, and the subjects departed. The study was approved by the State Medical Ethics Committee and by the University of Trier Ethics Committee. The study conformed to the revised Declaration of Helsinki principles. All subjects gave informed consent, and were informed of
938
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
their ability to discontinue participation at any time. All data was coded by number for confidentiality and anonymity. 2.5. Data analysis Repeated measures analysis of variance (ANOVA) models (with the betweensubjects factor AFI (sex specific median split of AFI: ⬍ age 19 for males and ⬍ age 18 for females), and the within-subjects factor TIME (at TSST –10, –1, +1, +10, +20, +30, and +40 min) were used separately for each sex. Covariate candidates that were tested (and excluded from the model if non-significant) were age, tobaccosmoking status, body mass index (kg/m2), menstrual cycle phase, and oral contraceptive use. Plasma cortisol response to ACTH provocation (at ACTH provocation –1, +45, and +60 min) was analyzed analogously. All analyses were two-tailed with an alpha level of .05. Greenhouse–Geisser corrected p values with uncorrected degrees of freedom are reported for variables with more than two levels. A second analytic technique used with the cortisol data was multiple regression. The baseline-adjusted area under the curve was the dependent variable, and the predictor variables were AFI (as a continuous variable), age, tobacco-smoking status, body mass index (kg/m2), menstrual cycle phase, Beck depression inventory score, intercourse frequency, and oral contraceptive use. Both stepwise and forced entry models were used. Only observed (no imputed) values were analyzed. Data were analyzed with SPSS statistical software. Following data entry, the data were checked twice by two other persons.
3. Results The ANOVA using a sex-specific median split had a significant interaction of Time with median split AFI in females (F(6, 246) = 2.95, p = .048, Greenhouse– Geisser epsilon = .408; see Fig. 1), but not in males (F(6, 204) = 2.0, p = .12, Greenhouse–Geisser epsilon = .537). For the females, there was no significant baseline difference, but the AFI ⱖ median group had a more intense cortisol reaction to the TSST than did the ⬍ median group (at time +20 min [t = 3.2; p⬍.01]; as well as a possible anticipation effect at time TSST +1 min [t = 2.9; p⬍.01]). It should be noted that the AFI below median group had a milder, but not a completely “blunted” cortisol reaction (levels did increase a statistically significant and meaningful 50%). A separate analysis excluding the virgins obtained very similar results. A separate analysis using the criterion of age at first intercourse before age 17 obtained a stronger effect for females (F(6, 246) = 5.4, p ⬍ .003, Greenhouse–Geisser epsilon = .434; there were also significant differences at more time points), but not males. In stepwise multiple regression analysis, the baseline-adjusted area under the cortisol curve for females was predicted by AFI (r = .41; F = 8.26, p = .006). No other variables entered the equation (the closest were smoking status: t = 1.9, p = .06, and oral contraceptive use: t = 1.6, p = .11; all others were t ⬍ 0.8, p ⬎ .45). In the
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
939
Fig. 1. Salivary cortisol (nmol/l) response to psychological stress (TSST) in males and females by sexspecific median split ages at first intercourse (AFI; ⬍19 vs. ⱖ19 or never male, ⬍18 vs. ⱖ18 or never female).
forced entry model, female area under the curve was predicted by AFI (t = 2.1, p = .048; all other predictors were t ⬍ 1.4 and p ⬎.19), with the overall model only marginally significant (p = .08) due to degradation of degrees of freedom. No variables predicted area under the curve for males in either analysis. As depicted in Fig. 2, there were no significant effects of AFI (as indexed by the median split) for either sex on the time course of plasma cortisol response to ACTH stimulation (both F ⬍ 0.6 and p ⬎ .5). 4. Discussion For women but not men, earlier AFI (as indexed by the median split) was associated with less intense cortisol reaction to the TSST stressor. Similarly, multiple regression analysis revealed that AFI was the only significant contributor to the variance in female baseline-adjusted cortisol area under the curve. In contrast, AFI had no significant effect on the cortisol response to low-dose ACTH stimulation. The non-significance of AFI for the cortisol response of males (despite a similar trend observable in the Figures) appears due not only to the smaller male sample
940
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
Fig. 2. Cortisol (µg/dl) response to 1 µg of ACTH (baseline, 45 min, 60 min) by sex-specific median split ages at first intercourse (AFI; ⬍19 vs. ⱖ19 or never male, ⬍18 vs. ⱖ18 or never female) and sex.
size, but also to a smaller effect. The basis for the sex difference is unknown, but possible explanations include hormonal factors and sociobiological influences (younger women desirous of intercourse may more readily fulfill their desires than younger men). Although both males and females with younger AFI were more likely to report being current tobacco smokers, smoking status was not quite significant in the stress response analyses (this is in contrast to the findings of Kirschbaum et al., 1993, who found that smoking status was related to stress responses). Menstrual phase, oral contraceptive use, age, and body mass index were also non-significant covariates in the stress response analyses. In previous studies, social dominance and internal locus of control were associated with lower cortisol reactions to the TSST (Pruessner et al., 1997), but only when averaged over multiple stress testing sessions. When single sessions were used, prior studies did not obtain significant associations of cortisol reactivity or recovery with personality measures (Roy et al., 2001). This makes AFI either a unique trait psychological–behavioral predictor, or a physiological predictor for stress responses. Persons who felt autonomous (and therefore more adult) at an earlier age might be more likely both to engage in intercourse earlier (Reiss, 1967; Reiss and Miller,
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
941
1979) and to be less physiologically stressed by the interpersonal nature of the stressor used in the instant study. It is also possible that some sexual correlates of AFI mediate the stress–AFI relationship: given that earlier AFI is associated with women’s greater coital orgasmic capacity (Raboch and Bartak, 1983) as well as men’s intercourse capability and frequency in old age in one sample (Vallery-Masson et al., 1981), it may be that those with earlier AFI also benefit more from the stressreducing effects of intercourse, an effect which carries over to other interactions in life. However, it may also be that genetic factors strongly influence dopamine tone, which affects AFI, intercourse frequency, orgasmic capacity, and stress response (or that genetic and/or early experiential factors lead to chronic cortisol stress hyperreactivity, which inhibits some aspects of sexuality). In the present investigation, there were no differences between the earlier and later coitarche groups with regard to the major dimensions of personality, Extraversion and Neuroticism, or with regard to Beck Depression scores. It is possible that more specific psychosocial measures (such as novelty or sensation seeking) might have correlated with AFI. Intercourse frequency in the preceding 14 days was significantly negatively associated with female AFI (the marginal significance of the male correlation, despite a similar correlation coefficient, is likely due to the smaller male sample size leading to inadequate statistical power). A shortcoming of the study is that free cortisol was used for the TSST response, but total cortisol for the ACTH response data, thereby limiting direct comparisons. It is not certain what the exact physiological mechanisms for the stress-protective effects are (nor is it certain from the extant literature that reduced cortisol stress reactivity has long-term benefits). However, the lack of difference between groups in cortisol response to ACTH provocation excludes the possibility that it is due to modification of adrenal responsiveness. Similarly, the non-significance of smoking status, oral contraceptive use, and body mass index exclude those factors. However, as in any epidemiologic study, unmeasured lifestyle and/or genetic factors may underlie the primary finding. This study does not provide direct support for a pathway through more frequent intercourse (which is associated with many health benefits; reviewed in Brody, 1997, see also Brody et al., 2000), because although AFI was associated with intercourse frequency, on regression analysis only AFI (and not intercourse frequency) predicted female area under the cortisol curve. However, it may be that a substantially longer intercourse frequency recording period might have yielded significant results (shorter periods are not only less representative, but also increase the number of zero values). In addition, the literature does not offer much support for a link through earlier puberty leading to variations in cortisol stress response, given that in children, pubertal status did not affect cortisol response to stress (Buske-Kirschbaum et al., 1997). As noted in the results section, exclusion of the virgin subjects would not have changed the results. It should be noted that there were no cases of pre-adolescent first intercourse, and results might have differed with such subjects. There were several advantages to the sample. There was no confounding of AFI by race (Brody, 1997) and the subjects were not recruited for a “sex study” (which tends to result in biased samples; Brody (1997)). In addition, the study was conducted in a country
942
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
that is less reproachful of sexuality than the USA (in which the majority of studies of AFI are conducted). The condemnation of sexuality in the USA (Posner and Silbaugh, 1996) may lead to different results in such studies, because the condemnation and/or illegality of earlier intercourse may lead to associations with conduct disorder. Future research on AFI and stress response might examine direct indices of central dopaminergic activity, genetic markers, and investigate the possible role of corticotropin-releasing factor, which not only increases cortisol levels, but also inhibits sexual behavior (Heinrichs et al., 1997).
Acknowledgements I thank the study participants and TSST members, and acknowledge the technical contributions and assistance of (in alphabetical order): Alexandra von HabsburgLothringen, Ragnar Preut, Kerstin Schommer, Thomas H. Schu¨ rmeyer and Larissa Tscherenkova.
References Beishuizen, A., van Lijf, J.H., Lekkerkerker, J.F.F., Vermes, I., 2000. The low dose (1 µg) ACTH stimulation test for assessment of the hypothalamo-pituitary-adrenal axis. Neth. J. Med 56, 91–99. Brody, S., 1997. Sex at risk: Lifetime number of partners, frequency of intercourse, and the low AIDS risk of vaginal intercourse. Transaction, New Brunswick, NJ. Brody, S., Preut, R., Schommer, K., Schu¨ rmeyer, T.H., 2002. A randomized controlled trial of high dose ascorbic acid for reduction of blood pressure, cortisol, and subjective responses to psychological stress. Psychopharmacology 159, 319–324. Brody, S., Veit, R., Rau, H., 2000. A preliminary report relating frequency of vaginal intercourse to heart rate variability, Valsalva ratio, blood pressure, and cohabitation status. Biol. Psychol 52, 251–257. Buske-Kirschbaum, A., Jobst, S., Wustmans, A., Kirschbaum, C., Rauh, W., Hellhammer, D., 1997. Attenuated free cortisol response to psychosocial stress in children with atopic dermatitis. Psychosom Med 59, 419–426. Dressendo¨ rfer, R.A., Kirschbaum, C., Rohde, W., Stahl, F., Strasburger, C.J., 1992. Synthesis of a cortisolbiotin conjugate and evaluation as a tracer in an immunoassay for salivary cortisol measurement. J. Steroid Biochem. Mol. Biol. 43, 683–692. Dunne, M.P., Martin, N.G., Statham, D.J., Slutske, W.S., Dinwiddie, S., Bucholz, K., Madden, P., Heath, A., 1997. Genetic and environmental contributions to variance in age at first sexual intercourse. Psychol. Sci. 8, 211–216. Eysenck, H.J., 1974. Eysenck Perso¨ nlichkeits Inventar. Verlag fu¨ r Psychologie, Go¨ ttingen. Hautzinger, M., Bailer, M., Worall, H., Keller, F., 1994. Beck-Depressions-Inventar (BDI). Huber, Bern. Heinrichs, S.C., Min, H., Tamraz, S., Carmouche, M., Boehme, S.A., Vale, W.W., 1997. Anti-sexual and anxiogenic behavioral consequences of corticotropin-releasing factor overexpression are centrally mediated. Psychoneuroendocrinology 22, 215–224. Hellhammer, D.H., Kirschbaum, C., Belkien, L., 1987. Measurement of salivary cortisol under psychological stimulation. In: Hingtgen, J.N., Hellhammer, D.H., Huppmann, G. (Eds.), Advanced Methods in Psychobiology. Hogrefe, Toronto, pp. 281–289. Isovich, E., Mijnster, M.J., Flu¨ gge, G., Fuchs, E., 2000. Chronic psychosocial stress reduces the density of dopamine transporters. Eur. J. Neurosci 12, 1071–1078. Kirschbaum, C., Pirke, K.M., Hellhammer, D.H., 1993. The ‘Trier Social Stress Test’--a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology 28, 76–81.
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
943
Lyons, D.M., Lopez, J.M., Yang, C., Schatzberg, A.F., 2000. Stress-level cortisol treatment impairs inhibitory control of behavior in monkeys. J. Neurosci. 20, 7816–7821. Martin, N.G., Eaves, L.J., Eysenck, H.J., 1977. Genetical, environmental, and personality factors influencing the age of first sexual intercourse in twins. J. Biosoc. Sci. 9, 91–97. Miller, W.B., Pasta, D.J., MacMurray, J., Chiu, C., Wu, H., Comings, D.E., 1999. Dopamine receptor genes are associated with age at first sexual intercourse. J. Biosoc. Sci. 31, 43–54. Paul, C., Fitzjohn, J., Herbison, P., Dickson, N., 2000. The determinants of sexual intercourse before age 16. J. Adolesc. Health 27, 136–147. Posner, R.A., Silbaugh, K.B., 1996. A guide to America’s sex laws. University of Chicago Press, Chicago. Pruessner, J.C., Gaab, J., Hellhammer, D.H., Lintz, D., Schommer, N., Kirschbaum, C., 1997. Increasing correlations between personality traits and cortisol stress responses obtained by data aggregation. Psychoneuroendocrinology 22, 615–625. Raboch, J., Bartak, V., 1983. Coitarche and orgastic capacity. Arch. Sex. Behav. 12, 409–413. Reiss, I., 1967. The social context of premarital sexual permissiveness. Holt, Rinehart and Winston, New York. Reiss, I., Miller, B., 1979. Heterosexual permissiveness: a theoretical analysis. In: Burr, W., Hill, R., Nye, F., Reiss, I. (Eds.), Contemporary theories about the family. Vol. 1. The Free Press. Rosenthal, D.A., Smith, A.M., de Visser, R., 1999. Personal and social factors influencing age at first sexual intercourse. Arch. Sex Behav. 28, 319–333. Roy, M.P., Kirschbaum, C., Steptoe, A., 2001. Psychological, cardiovascular, and metabolic correlates of individual differences in cortisol stress recovery in young men. Psychoneuroendocrinology 26, 375–391. Seeman, T.E., Berkman, L.F., Gulanski, B.I., Robbins, R.J., Greenspan, S.L., Charpentier, P.A., Rowe, J.W., 1995. Self-esteem and neuroendocrine response to challenge: MacArthur studies of successful aging. J. Psychosom. Res. 39, 69–84. Tordjman, K., Jaffe, A.A., Trostanetsky, Y., Greenman, Y., Limor, R., Stern, N., 2000. Low-dose (1 µg) adrenocorticotrophin (ACTH) stimulation as a screening test for impaired hypothalamo-pituitary-adrenal axis function: sensitivity, specificity and accuracy in comparison with the high-dose (250 µg) test. Clin. Endocrinol 52, 633–640. Udry, J.R., Cliquet, R.L., 1982. A cross-cultural examination of the relationship between ages at menarche, marriage, and first birth. Demography 19, 53–63. Udry, J.R., Kovenock, J., Morris, N.M., van den Berg, B.J., 1995. Childhood precursors of age at first intercourse for females. Arch. Sex Behav. 24, 329–337. Vallery-Masson, J., Valleron, A.J., Poitrenaud, J., 1981. Factors related to sexual intercourse frequency in a group of French pre-retirement managers. Age Aging 10, 53–59.
Age at first intercourse is inversely related to female cortisol stress reactivity Stuart Brody a,b,∗ a
Center for Psychobiological and Psychosomatic Research, University of Trier, Karl-Marx Strasse 94, 54290 Trier, Germany b Institute of Medical Psychology, University of Tu¨bingen, Germany Received 9 August 2001; received in revised form 15 January 2002; accepted 15 January 2002
Abstract The relationship between age at first sexual intercourse and salivary cortisol stress reactivity (to the Trier Social Stress Test; TSST; consisting of public speaking and mental arithmetic) was examined in healthy subjects (43 females and 36 males; ages 19–38). Women reporting earlier first intercourse had less intense cortisol increases in response to the stressor (a nonsignificant trend was observed for males), and faster recovery from the stressor. Results were not confounded by age, oral contraceptive use, depression scores, smoking status, or body mass index. It is concluded that earlier first intercourse is associated with less reactivity to and faster recovery from stress as indexed by this endocrine measure. Results are discussed in terms of genetic and psychological influences on first intercourse and implications for coping with interpersonal stress. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Coitus; Sexual behavior; Cortisol; Stress reactivity; Stress recovery
1. Introduction Age at first sexual (penile–vaginal) intercourse (AFI) has a significant genetic loading (Martin et al., 1977; Dunne et al., 1997), and part of this mechanism appears
∗
Center for Psychobiological and Psychosomatic Research, University of Trier, Karl-Marx Strasse 94, 54290 Trier, Germany. Fax: +1-561-431-3114. E-mail address: [email protected] (S. Brody). 0306-4530/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 3 0 6 - 4 5 3 0 ( 0 2 ) 0 0 0 0 7 - 0
934
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
to involve genes for dopamine D1 and D2 receptors (Miller et al., 1999). Women’s earlier AFI was predicted by better motor skills at age 5, a domineering and mature personality at age 9 (Udry et al., 1995), earlier menarche (Udry and Cliquet, 1982), adolescent perception of earlier autonomy and physical maturity, and lack of restraint (Rosenthal et al., 1999). AFI is associated with lifelong sexual benefits as well: women’s earlier AFI is associated with greater coital orgasmic capacity (Raboch and Bartak, 1983), and in one sample (of unknown representativeness) of 64-year-old men, the sexually functional were differentiated from the nonfunctional by earlier AFI (Vallery-Masson et al., 1981). In several studies, AFI was associated inversely with adult intercourse frequency (reviewed in Brody (1997). Both AFI (in some studies of women; Paul et al., 2000) and cortisol response to stress (Seeman, 1995) have been found to be related inversely to self-esteem. The domineering and autonomous features noted above to be associated with earlier AFI have a parallel in the finding that social dominance and internal locus of control were associated with lower cortisol reactions (Pruessner et al., 1997) to the Trier Social Stress Test (TSST). In addition to genetic influences and behavioral manifestations of dopaminergic tone influencing cortisol response to social stress, cortisol responses may influence dopamine function. In laboratory animals, treatment with cortisol (at levels simulating a prolonged stress response) impaired dopamine-dependent prefrontal cortical functions (Lyons et al., 2000). Similarly, chronic psychosocial stress reduced the density of dopamine transporters in the caudate nucleus and putamen, and also reduced serum testosterone and testicular weight (Isovich et al., 2000). One may conjecture that chronic cortisol elevations associated with greater stress response could also interfere (or have interfered) with the dopamine-related initiation of sexual behavior. There are several pathways by which younger AFI might reduce stress response: earlier AFI could be a marker of a genetic predisposition against stress response, earlier AFI could have interpersonal and intrapsychic consequences leading to less stress response, and/or earlier AFI could be a precursor of more frequent intercourse, which results in less stress response. For the current study, the relationship between AFI and cortisol response to psychological stress was examined in a sample of healthy adults. To investigate one possible mechanism for differences in cortisol response (modification of adrenal sensitivity), the cortisol response to ACTH stimulation was also examined.
2. Method 2.1. Participants The study was embedded in a randomized placebo-controlled clinical trial of highdose ascorbic acid on stress response (Brody et al., 2002). Approximately half of the subjects in the current report consumed ascorbic acid supplements for the clinical trial, but trial group had no effect on either ACTH stimulation or overall cortisol
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
935
stress response (Brody et al., 2002), nor did it have any significant effect in any of the models herein, so to preserve degrees of freedom this variable was not included in the present analyses. Subjects were recruited for the trial from advertisements posted at the university, at a technical college, and at various shops and sports centers, and screened by telephone per the exclusion criteria (current medication except oral contraception; vitamin supplement use during the preceding three months; current illnesses or pregnancy; prior experience with the TSST stressor; age over 40 or under 19; history of psychiatric disturbance; ulcer, cancer, asthma, cardiovascular, endocrine, or neurological disorders; body mass index ⬎ 34 kg/m2, or indications of enhanced study protocol noncompliance risk), and then told the general details of the study. The subjects received a second screening as part of the brief medical examination at the first laboratory session. In the clinical trial in which this study was embedded (Brody et al., 2002), 12 of the original 120 subjects were excluded: four who reported unusual stress earlier on the TSST day (thereby precluding valid resting baseline values), four who were known to members of the TSST panel (thereby modifying the stress), one who was found to have recently taken corticosteroids as part of another study, one for suspected substance use the day of the TSST, one for noncompliance with the clinical trial dosing protocol, and one (in the placebo group) for fainting during the TSST. For the present study, subjects were also excluded if they scored at or above five (the 86th percentile in our sample) on the Lie scale of the German version of the Eysenck Personality Inventory (Eysenck, 1974). Excluding subjects with a tendency to misrepresent behavior that may be seen as socially undesirable is methodologically crucial (Brody, 1997; Brody et al., 2000). As per the recommendation of an anonymous reviewer, eight male and three female outlier subjects displaying a large (ⱖ6 nmol/L, see below) drop in salivary cortisol directly after the stressor (or from before to the time of the stressor) were excluded from the present analyses. The characteristics of the analyzed subjects by AFI group memberships are presented in Table 1. The overall mean and median AFI was 18 (range 14–33; plus two female virgins). Males reported an older mean AFI than females (19.3 vs. 17.2, t for unequal variances = 3.2, p = .003), were older (mean age 25.7 vs. 23.8, t = 2.0, p = .048), and had greater body mass index (23.8 vs. 22.4, t = 2.3, p = .02), but did not differ on the other variables listed in the Table. Using the median split of AFI (on intrasex comparisons), males with younger AFI were younger (mean age 23.7 vs. 27.4, t = 2.7, p = .01), reported more intercourse days per month (transformed from the 14 day diaries; 7.8 vs. 5.0, t = 2.5, p = .016), and both males and females with younger AFI were more likely to report being current tobacco smokers (both chi-square = 4.4, p = .04). AFI correlated significantly inversely with the number of penile–vaginal intercourse days in the previous 14 days for females (r = –.33, p = .03) but only marginally for males (r = –.29, p = .09).
936
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
Table 1 Characteristics of subjects by age of first intercourse (age-specific median split:⬍18 vs. ⱖ18 or never female, ⬍19 vs. ⱖ19 or never male) group Younger age at first intercourse (⬍19 male, ⬍18 female)
Older age at first intercourse (ⱖ19 or never male, ⱖ18 or never female)
Age (mean, SD)
23.7, 3.4 male 22.7, 2.1 female
27.4, 4.5 male 25.1, 5.7 female
Sex
16 male 24 female
20 male 19 female
Education
11 Some university, 1 University graduate, 4 other male; 3 High School (Abitur), 20 some university, 1 University graduate female
1 High School (Abitur), 13 some university, 5 University graduate, 1 other male; 1 High School (Abitur), 14 some university, 3 University graduate, 1 other female
Current tobacco smokers
12 male 14 female
8 male 5 female
Oral contraceptive users
17
12
Body mass index (kg/m2; mean, SD)
22.9, 2.4 male 22.1, 2.3 female
24.4, 3.1 male 22.7, 2.8 female
Age at first intercourse (mean, SD)
16.9, 1.1 male 15.9, 0.9 female
21.4, 3.6 male 18.8, 0.9 female
Beck Depression Score (mean, SD)
7.4, 6.5 male 4.5, 2.8 female
5.9, 4.2 male 5.8, 5.5 female
Eysenck Extraversion score (mean, SD)
14.6, 3.1 male 14.3, 3.7 female
12.4, 4.8 male 14.2, 4.9 female
Eysenck Neuroticism score (mean, SD)
8.0, 5.6 male 8.9, 4.9 female
8.3, 4.7 male 9.6, 4.4 female
Menstrual phase (N)
days 1–7: 7 days 8–20: 11 days 21–27: 6 missing: 0
days 1–7: 7 days 8–20: 4 days 21–27: 7 missing: 1
2.2. Measures 2.2.1. Cortisol Salivary cortisol was collected by means of chewing on plastic-sheathed cotton, which was returned to a sterile plastic holder (Hellhammer et al., 1987), and analyzed with a time-resolved immunoassay (Dressendo¨ rfer et al., 1992).
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
937
After the blood for plasma cortisol response to ACTH provocation was drawn, it was placed on ice and then centrifuged, and the plasma was pipetted and frozen at –20 ° C. Plasma cortisol levels were determined by chemoluminescence using a commercial kit (Nichols Cortisol 100T kit; San Juan Capistrano, California). 2.2.2. Psychosocial measures Subjects completed questionnaires including the German versions of the Eysenck Lie Scale (Eysenck, 1974) and Beck Depression Inventory (Hautzinger et al., 1994) and our own questionnaire which included questions on AFI (defined as penis in vagina), tobacco smoking status, menstrual cycle phase (subjects were asked the date their last menstruation began), and oral contraceptive use. The subjects also completed a 14-day diary, which included whether penile–vaginal intercourse had occurred on each day. 2.3. Laboratory tasks The psychological stressor consisted of the TSST (Kirschbaum et al., 1993), a standardized psychological stress-induction technique consisting of being told (following a “–10 min” salivary cortisol measurement) that one will have to give a five-minute speech to an unknown panel on personal suitability for a job in the subject’s field of interest, followed by five minutes of mental arithmetic performed out loud. After solitary preparation, the “–1 min” measurement is taken (reflecting the stress of preparation and anticipation of the actual stressor), and the subject enters the TSST room containing the panel of observers (one of each sex) and an obvious video camera and microphone. Pauses during the speech are dealt with by being reminded of remaining time. After 5 min, the task shifts to performing serial subtractions of 17 starting at 2023, with errors resulting in being required to return to the beginning. Immediately afterwards, the “+1 min” measurement is taken, and the subject leaves to the waiting area, where the “+10, +20, +30, and +40 min” measurements are taken in turn. 2.4. Procedure All laboratory sessions were conducted in the mid-afternoon, at least one hour after eating (there was no eating or smoking during the laboratory sessions). After completing questionnaires for at least 10 min, the subjects participated in the TSST. After the last (“+40 min”) measurement, subjects had a cannula inserted, blood was drawn for baseline plasma cortisol levels, followed by an injection of 1 µg of ACTH (the 1 µg dose is more sensitive and at least as safe as higher doses; Beishuizen et al. (2000)bib:Beishuizen, 2000; Tordjman et al. (2000). After 45 and 60 min, blood was again drawn for determination of plasma cortisol response to the ACTH provocation. The cannula was removed, and the subjects departed. The study was approved by the State Medical Ethics Committee and by the University of Trier Ethics Committee. The study conformed to the revised Declaration of Helsinki principles. All subjects gave informed consent, and were informed of
938
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
their ability to discontinue participation at any time. All data was coded by number for confidentiality and anonymity. 2.5. Data analysis Repeated measures analysis of variance (ANOVA) models (with the betweensubjects factor AFI (sex specific median split of AFI: ⬍ age 19 for males and ⬍ age 18 for females), and the within-subjects factor TIME (at TSST –10, –1, +1, +10, +20, +30, and +40 min) were used separately for each sex. Covariate candidates that were tested (and excluded from the model if non-significant) were age, tobaccosmoking status, body mass index (kg/m2), menstrual cycle phase, and oral contraceptive use. Plasma cortisol response to ACTH provocation (at ACTH provocation –1, +45, and +60 min) was analyzed analogously. All analyses were two-tailed with an alpha level of .05. Greenhouse–Geisser corrected p values with uncorrected degrees of freedom are reported for variables with more than two levels. A second analytic technique used with the cortisol data was multiple regression. The baseline-adjusted area under the curve was the dependent variable, and the predictor variables were AFI (as a continuous variable), age, tobacco-smoking status, body mass index (kg/m2), menstrual cycle phase, Beck depression inventory score, intercourse frequency, and oral contraceptive use. Both stepwise and forced entry models were used. Only observed (no imputed) values were analyzed. Data were analyzed with SPSS statistical software. Following data entry, the data were checked twice by two other persons.
3. Results The ANOVA using a sex-specific median split had a significant interaction of Time with median split AFI in females (F(6, 246) = 2.95, p = .048, Greenhouse– Geisser epsilon = .408; see Fig. 1), but not in males (F(6, 204) = 2.0, p = .12, Greenhouse–Geisser epsilon = .537). For the females, there was no significant baseline difference, but the AFI ⱖ median group had a more intense cortisol reaction to the TSST than did the ⬍ median group (at time +20 min [t = 3.2; p⬍.01]; as well as a possible anticipation effect at time TSST +1 min [t = 2.9; p⬍.01]). It should be noted that the AFI below median group had a milder, but not a completely “blunted” cortisol reaction (levels did increase a statistically significant and meaningful 50%). A separate analysis excluding the virgins obtained very similar results. A separate analysis using the criterion of age at first intercourse before age 17 obtained a stronger effect for females (F(6, 246) = 5.4, p ⬍ .003, Greenhouse–Geisser epsilon = .434; there were also significant differences at more time points), but not males. In stepwise multiple regression analysis, the baseline-adjusted area under the cortisol curve for females was predicted by AFI (r = .41; F = 8.26, p = .006). No other variables entered the equation (the closest were smoking status: t = 1.9, p = .06, and oral contraceptive use: t = 1.6, p = .11; all others were t ⬍ 0.8, p ⬎ .45). In the
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
939
Fig. 1. Salivary cortisol (nmol/l) response to psychological stress (TSST) in males and females by sexspecific median split ages at first intercourse (AFI; ⬍19 vs. ⱖ19 or never male, ⬍18 vs. ⱖ18 or never female).
forced entry model, female area under the curve was predicted by AFI (t = 2.1, p = .048; all other predictors were t ⬍ 1.4 and p ⬎.19), with the overall model only marginally significant (p = .08) due to degradation of degrees of freedom. No variables predicted area under the curve for males in either analysis. As depicted in Fig. 2, there were no significant effects of AFI (as indexed by the median split) for either sex on the time course of plasma cortisol response to ACTH stimulation (both F ⬍ 0.6 and p ⬎ .5). 4. Discussion For women but not men, earlier AFI (as indexed by the median split) was associated with less intense cortisol reaction to the TSST stressor. Similarly, multiple regression analysis revealed that AFI was the only significant contributor to the variance in female baseline-adjusted cortisol area under the curve. In contrast, AFI had no significant effect on the cortisol response to low-dose ACTH stimulation. The non-significance of AFI for the cortisol response of males (despite a similar trend observable in the Figures) appears due not only to the smaller male sample
940
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
Fig. 2. Cortisol (µg/dl) response to 1 µg of ACTH (baseline, 45 min, 60 min) by sex-specific median split ages at first intercourse (AFI; ⬍19 vs. ⱖ19 or never male, ⬍18 vs. ⱖ18 or never female) and sex.
size, but also to a smaller effect. The basis for the sex difference is unknown, but possible explanations include hormonal factors and sociobiological influences (younger women desirous of intercourse may more readily fulfill their desires than younger men). Although both males and females with younger AFI were more likely to report being current tobacco smokers, smoking status was not quite significant in the stress response analyses (this is in contrast to the findings of Kirschbaum et al., 1993, who found that smoking status was related to stress responses). Menstrual phase, oral contraceptive use, age, and body mass index were also non-significant covariates in the stress response analyses. In previous studies, social dominance and internal locus of control were associated with lower cortisol reactions to the TSST (Pruessner et al., 1997), but only when averaged over multiple stress testing sessions. When single sessions were used, prior studies did not obtain significant associations of cortisol reactivity or recovery with personality measures (Roy et al., 2001). This makes AFI either a unique trait psychological–behavioral predictor, or a physiological predictor for stress responses. Persons who felt autonomous (and therefore more adult) at an earlier age might be more likely both to engage in intercourse earlier (Reiss, 1967; Reiss and Miller,
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
941
1979) and to be less physiologically stressed by the interpersonal nature of the stressor used in the instant study. It is also possible that some sexual correlates of AFI mediate the stress–AFI relationship: given that earlier AFI is associated with women’s greater coital orgasmic capacity (Raboch and Bartak, 1983) as well as men’s intercourse capability and frequency in old age in one sample (Vallery-Masson et al., 1981), it may be that those with earlier AFI also benefit more from the stressreducing effects of intercourse, an effect which carries over to other interactions in life. However, it may also be that genetic factors strongly influence dopamine tone, which affects AFI, intercourse frequency, orgasmic capacity, and stress response (or that genetic and/or early experiential factors lead to chronic cortisol stress hyperreactivity, which inhibits some aspects of sexuality). In the present investigation, there were no differences between the earlier and later coitarche groups with regard to the major dimensions of personality, Extraversion and Neuroticism, or with regard to Beck Depression scores. It is possible that more specific psychosocial measures (such as novelty or sensation seeking) might have correlated with AFI. Intercourse frequency in the preceding 14 days was significantly negatively associated with female AFI (the marginal significance of the male correlation, despite a similar correlation coefficient, is likely due to the smaller male sample size leading to inadequate statistical power). A shortcoming of the study is that free cortisol was used for the TSST response, but total cortisol for the ACTH response data, thereby limiting direct comparisons. It is not certain what the exact physiological mechanisms for the stress-protective effects are (nor is it certain from the extant literature that reduced cortisol stress reactivity has long-term benefits). However, the lack of difference between groups in cortisol response to ACTH provocation excludes the possibility that it is due to modification of adrenal responsiveness. Similarly, the non-significance of smoking status, oral contraceptive use, and body mass index exclude those factors. However, as in any epidemiologic study, unmeasured lifestyle and/or genetic factors may underlie the primary finding. This study does not provide direct support for a pathway through more frequent intercourse (which is associated with many health benefits; reviewed in Brody, 1997, see also Brody et al., 2000), because although AFI was associated with intercourse frequency, on regression analysis only AFI (and not intercourse frequency) predicted female area under the cortisol curve. However, it may be that a substantially longer intercourse frequency recording period might have yielded significant results (shorter periods are not only less representative, but also increase the number of zero values). In addition, the literature does not offer much support for a link through earlier puberty leading to variations in cortisol stress response, given that in children, pubertal status did not affect cortisol response to stress (Buske-Kirschbaum et al., 1997). As noted in the results section, exclusion of the virgin subjects would not have changed the results. It should be noted that there were no cases of pre-adolescent first intercourse, and results might have differed with such subjects. There were several advantages to the sample. There was no confounding of AFI by race (Brody, 1997) and the subjects were not recruited for a “sex study” (which tends to result in biased samples; Brody (1997)). In addition, the study was conducted in a country
942
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
that is less reproachful of sexuality than the USA (in which the majority of studies of AFI are conducted). The condemnation of sexuality in the USA (Posner and Silbaugh, 1996) may lead to different results in such studies, because the condemnation and/or illegality of earlier intercourse may lead to associations with conduct disorder. Future research on AFI and stress response might examine direct indices of central dopaminergic activity, genetic markers, and investigate the possible role of corticotropin-releasing factor, which not only increases cortisol levels, but also inhibits sexual behavior (Heinrichs et al., 1997).
Acknowledgements I thank the study participants and TSST members, and acknowledge the technical contributions and assistance of (in alphabetical order): Alexandra von HabsburgLothringen, Ragnar Preut, Kerstin Schommer, Thomas H. Schu¨ rmeyer and Larissa Tscherenkova.
References Beishuizen, A., van Lijf, J.H., Lekkerkerker, J.F.F., Vermes, I., 2000. The low dose (1 µg) ACTH stimulation test for assessment of the hypothalamo-pituitary-adrenal axis. Neth. J. Med 56, 91–99. Brody, S., 1997. Sex at risk: Lifetime number of partners, frequency of intercourse, and the low AIDS risk of vaginal intercourse. Transaction, New Brunswick, NJ. Brody, S., Preut, R., Schommer, K., Schu¨ rmeyer, T.H., 2002. A randomized controlled trial of high dose ascorbic acid for reduction of blood pressure, cortisol, and subjective responses to psychological stress. Psychopharmacology 159, 319–324. Brody, S., Veit, R., Rau, H., 2000. A preliminary report relating frequency of vaginal intercourse to heart rate variability, Valsalva ratio, blood pressure, and cohabitation status. Biol. Psychol 52, 251–257. Buske-Kirschbaum, A., Jobst, S., Wustmans, A., Kirschbaum, C., Rauh, W., Hellhammer, D., 1997. Attenuated free cortisol response to psychosocial stress in children with atopic dermatitis. Psychosom Med 59, 419–426. Dressendo¨ rfer, R.A., Kirschbaum, C., Rohde, W., Stahl, F., Strasburger, C.J., 1992. Synthesis of a cortisolbiotin conjugate and evaluation as a tracer in an immunoassay for salivary cortisol measurement. J. Steroid Biochem. Mol. Biol. 43, 683–692. Dunne, M.P., Martin, N.G., Statham, D.J., Slutske, W.S., Dinwiddie, S., Bucholz, K., Madden, P., Heath, A., 1997. Genetic and environmental contributions to variance in age at first sexual intercourse. Psychol. Sci. 8, 211–216. Eysenck, H.J., 1974. Eysenck Perso¨ nlichkeits Inventar. Verlag fu¨ r Psychologie, Go¨ ttingen. Hautzinger, M., Bailer, M., Worall, H., Keller, F., 1994. Beck-Depressions-Inventar (BDI). Huber, Bern. Heinrichs, S.C., Min, H., Tamraz, S., Carmouche, M., Boehme, S.A., Vale, W.W., 1997. Anti-sexual and anxiogenic behavioral consequences of corticotropin-releasing factor overexpression are centrally mediated. Psychoneuroendocrinology 22, 215–224. Hellhammer, D.H., Kirschbaum, C., Belkien, L., 1987. Measurement of salivary cortisol under psychological stimulation. In: Hingtgen, J.N., Hellhammer, D.H., Huppmann, G. (Eds.), Advanced Methods in Psychobiology. Hogrefe, Toronto, pp. 281–289. Isovich, E., Mijnster, M.J., Flu¨ gge, G., Fuchs, E., 2000. Chronic psychosocial stress reduces the density of dopamine transporters. Eur. J. Neurosci 12, 1071–1078. Kirschbaum, C., Pirke, K.M., Hellhammer, D.H., 1993. The ‘Trier Social Stress Test’--a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology 28, 76–81.
S. Brody / Psychoneuroendocrinology 27 (2002) 933–943
943
Lyons, D.M., Lopez, J.M., Yang, C., Schatzberg, A.F., 2000. Stress-level cortisol treatment impairs inhibitory control of behavior in monkeys. J. Neurosci. 20, 7816–7821. Martin, N.G., Eaves, L.J., Eysenck, H.J., 1977. Genetical, environmental, and personality factors influencing the age of first sexual intercourse in twins. J. Biosoc. Sci. 9, 91–97. Miller, W.B., Pasta, D.J., MacMurray, J., Chiu, C., Wu, H., Comings, D.E., 1999. Dopamine receptor genes are associated with age at first sexual intercourse. J. Biosoc. Sci. 31, 43–54. Paul, C., Fitzjohn, J., Herbison, P., Dickson, N., 2000. The determinants of sexual intercourse before age 16. J. Adolesc. Health 27, 136–147. Posner, R.A., Silbaugh, K.B., 1996. A guide to America’s sex laws. University of Chicago Press, Chicago. Pruessner, J.C., Gaab, J., Hellhammer, D.H., Lintz, D., Schommer, N., Kirschbaum, C., 1997. Increasing correlations between personality traits and cortisol stress responses obtained by data aggregation. Psychoneuroendocrinology 22, 615–625. Raboch, J., Bartak, V., 1983. Coitarche and orgastic capacity. Arch. Sex. Behav. 12, 409–413. Reiss, I., 1967. The social context of premarital sexual permissiveness. Holt, Rinehart and Winston, New York. Reiss, I., Miller, B., 1979. Heterosexual permissiveness: a theoretical analysis. In: Burr, W., Hill, R., Nye, F., Reiss, I. (Eds.), Contemporary theories about the family. Vol. 1. The Free Press. Rosenthal, D.A., Smith, A.M., de Visser, R., 1999. Personal and social factors influencing age at first sexual intercourse. Arch. Sex Behav. 28, 319–333. Roy, M.P., Kirschbaum, C., Steptoe, A., 2001. Psychological, cardiovascular, and metabolic correlates of individual differences in cortisol stress recovery in young men. Psychoneuroendocrinology 26, 375–391. Seeman, T.E., Berkman, L.F., Gulanski, B.I., Robbins, R.J., Greenspan, S.L., Charpentier, P.A., Rowe, J.W., 1995. Self-esteem and neuroendocrine response to challenge: MacArthur studies of successful aging. J. Psychosom. Res. 39, 69–84. Tordjman, K., Jaffe, A.A., Trostanetsky, Y., Greenman, Y., Limor, R., Stern, N., 2000. Low-dose (1 µg) adrenocorticotrophin (ACTH) stimulation as a screening test for impaired hypothalamo-pituitary-adrenal axis function: sensitivity, specificity and accuracy in comparison with the high-dose (250 µg) test. Clin. Endocrinol 52, 633–640. Udry, J.R., Cliquet, R.L., 1982. A cross-cultural examination of the relationship between ages at menarche, marriage, and first birth. Demography 19, 53–63. Udry, J.R., Kovenock, J., Morris, N.M., van den Berg, B.J., 1995. Childhood precursors of age at first intercourse for females. Arch. Sex Behav. 24, 329–337. Vallery-Masson, J., Valleron, A.J., Poitrenaud, J., 1981. Factors related to sexual intercourse frequency in a group of French pre-retirement managers. Age Aging 10, 53–59.