Introduction

From its first use in the treatment of depression in 1948 [1], testosterone (T) has a long clinical history in the application for psychological disorders. Between 1950 and 2016, a recent systematic review identified 45 randomized controlled trials (RCT) investigating the effect of T treatment on personality, well-being or mood, but without any consistent effect on psychological outcomes [2].

The empirical evidence for a link between T and psychological traits, such as optimism and pessimism, is only weak. Initial support for a potential hormonal basis of psychological traits relies on findings from two different strands of literature. First, research on prenatal T provides tentative evidence that traits are—at least partially—biologically determined [3]. Secondly, experimental research in economic decision-making suggests a potential link of T with risk-taking, overconfidence and optimism [4]. However, the exact association between T and psychological traits is still elusive [5]. Therefore, we investigated the association of T with optimism and pessimism using cross-sectional and longitudinal data from a large patient-based sample of men and women.

Methods

Results

Sex-specific baseline characteristics of the study population are presented in Table 1. Cross-sectional analyses showed no association of T with optimism, pessimism or total LOT-R score among men and women (Table 2). Assessing a continuum of T concentrations and LOT-R scores, we observed no graded slope between T and optimism or pessimism (Fig 1). Longitudinal analyses of optimism, pessimism, and total LOT-R score yielded similarly non-significant results (multivariable-adjusted β-coefficients per one unit T change for LOT-R: men, -0.01 (95% CI: -0.24–0.22); women, 0.08 (-0.03–0.20)). Finally, one-year change in T was not related to optimism, pessimism or total LOT-R at four-year follow-up. None of the performed sensitivity analyses changed the revealed findings.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Optimism and pessimism by baseline testosterone among men.

Linear regression for baseline optimism (green) and pessimism (blue) by baseline total testosterone concentrations. The green and blue lines represent the fitted regression line and the grey areas the 95% confidence intervals.

https://doi.org/10.1371/journal.pone.0207870.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Sex-specific baseline characteristics of the study population.

https://doi.org/10.1371/journal.pone.0207870.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Sex-specific associations of total testosterone with optimism, pessimism, and total LOT-R.

https://doi.org/10.1371/journal.pone.0207870.t002

Discussion

To the best of our knowledge, this is the first observational study to investigate the association of T with optimism and pessimism in a large-scale sample of men and women. Taken together, cross-sectional and longitudinal analyses showed no association of T with dispositional optimism and pessimism, assessed by the LOT-R.

The tentative role of T as hormonal correlate of psychological traits stems from research in prenatal T exposure and economic decision-making. Using the right hand 2D:4D digit ratio (length of the index finger to the length of the ring finger) as putative marker for relative prenatal T exposure, the 2D:4D ratio has been shown to correlate with traits such as altruism [9], cooperation [10], and cognitive reflection [11]. Another line of research in economic decision-making suggests that T is linked to optimism and risk-taking [4], trust, and self-confidence [12]. But given their small scale, strongly selected samples, and narrow focus, the results of these studies are of low generalizability and do not allow for definite conclusions about a potential link between T and personality traits.

The absent association of endogenous T with dispositional optimism and pessimism in the present study provides a rationale for the lack of interventional evidence reporting no conclusive effect of exogenous T on personality, psychological well-being or mood in clinical trials [2]. Although men under T treatment occasionally show slightly better mood and lower severity of depressive symptoms, as reported by the long-awaited T-trials [13], the statistical significance of these findings does not compensate for their small effect size and unknown clinical significance. Also the suggested dose-response relationship between T and several domains of psychological well-being did not stand up to closer scrutiny. Exemplarily, a RCT among 44 healthy older men receiving five graded T doses over 20 weeks showed not effect on two measures of mood, including Hamilton's Depression Inventory and Young's Mania Scale [14].

Alternatively, change in T itself has been suggested as a relevant predictor of adverse health outcomes [15], risk taking behavior [16], and depressive symptoms [8]. However, on the level of psychological traits, the absence of an association of T change with optimism and pessimism in the present study is in line with synthesized evidence of a systematic review including 27 observational studies among healthy male adolescents, experiencing marked T changes during puberty, reporting no association of T change with behavior [17].

Although our findings agree with previous data showing low correlations between optimism and multiple laboratory parameters in the DETECT study, it is important to mention differences in the analyses of the LOT-R. While some studies investigated optimism and pessimism as two independent constructs [18], others assessed optimism as a continuous measure [19]. To address this methodological heterogeneity, we analyzed both: two-dimensional optimism and pessimism variables, as well as a continuous LOT-R score as one-dimensional trait. However, both analytical strategies yielded non-significant results.

The interpretation of the observational evidence presented here warrants consideration of the following strengths and potential limitations. Strengths of our study include the large sample size, repeated follow-up measures, and the comparative assessment of baseline T vs. T change in men as well as in women. Limitations may arise from the questionnaire-based assessment of LOT-R, relying on subjective and potentially inaccurate patient responses. Furthermore, our study is limited by the lack of data regarding additional sex hormones and alternative measures of the optimism-pessimism continuum, such as the Unrealistic Optimism Scale, the Hope Scale or the Optimism-Pessimism Scale [20].

In summary, the observational data reported here shows no association of T with optimism, pessimism or total LOT-R score, suggesting further research to elucidate potential direct and indirect associations between individual hormonal status and psychological traits.

Acknowledgments

DETECT (Diabetes Cardiovascular Risk-Evaluation: Targets and Essential Data for Commitment of Treatment) is a cross-sectional and prospective-longitudinal, nationwide clinical epidemiological study. DETECT is supported by an unrestricted educational grant of Pfizer GmbH, Karlsruhe, Germany. Members of the DETECT-Study group include: Principal investigator: Professor Dr. H.-U. Wittchen; Staff members: Dr. L. Pieper, Dr. J. Klotsche, Dr. T. Eichler, Dr. H. Glaesmer, E. Katze. Steering Committee: Professor Dr. H. Lehnert (Lübeck), Professor Dr. G. K. Stalla (München), Professor Dr. A. M. Zeiher (Frankfurt); Advisory Board: Professor Dr. W. März (Heidelberg/Graz), Professor Dr. S. Silber (München), Professor Dr. Dr. U. Koch-Gromus (Hamburg), Professor Dr. D. Pittrow (München/Dresden), Professor Dr. M. Wehling (Mannheim), Dr. D. Leistner (Frankfurt), Professor Dr. H. J. Schneider (München), Dr. C. Sievers (München). The current data analysis was conducted within BMBF Grants 01ER1303 & 01ER1703.