The protective relationship of working memory on sexual initiation was mediated through acting-without thinking and temporal discounting. Kahn also found the meditational effects of discounting on relationship between adolescent-parent relationship quality and risk behavior, which were moderated by self-control. Higher delay discounting was related to higher levels of risky sexual behavior for the low self-control group but not the high self-control group. The indirect effect of parent–adolescent relationship quality on risky sexual behavior was significant for the low self-control group. Jones and Sullivan found some evidence of a developmental period of importance in the relationship of discounting to sexual risk behavior. Among MSM between aged 18-24, but not older men, higher monetary discounting was associated with more reports of condomless anal intercourse. Many of the studies investigated discounting and sexual behavior in the context of substance use. Alcohol and other substance use are reliable predictors of high–risk sexual behavior, and often interact with other factors to increase risk. There is also a strong relationship between substance use and discounting, leading researchers to explore discounting among substance users and also as a potential mediating variable between substance use and sexual risk behavior. Studies that examined drug and alcohol users as a subgroup of interest found largely consistent relationships between discounting and sexual behavior. Finding included that discounting was associated with HIV risk behaviors among high-risk drug users, history of high-risk sex among cocaine users,cannabis cultivation technology and percentage of alcohol-related condomless sex . Celio found differential effects of alcohol expectancies based on the degree to which individuals discount delayed rewards, meaning that those who believed they would engage in sex while drinking were more likely to have done so in the past .” In another study among heavy or problem drinkers, monetary delay discounting, but not probability discounting, was associated with increased frequency of unprotected sex with a non-steady partner while drinking.
Higher discounting of condom protected sex in the SDT was associated with substance use and sex under the influence of substances among MSM and interacted with overvaluation of alcohol, to account for both increased sexual risk behavior and alcohol consumption among sexually active college drinkers. Brodbeck et al. was the only study to examine cannabis use, testing whether the connection between sexual behavior and cannabis use was causal or could be attributed to an underlying risk propensity making people predisposed to both risk behaviors. They found a weak meditational relationship between the measure of risk preference/hedonism and the relationship of cannabis use and unprotected sex. Not all studies found evidence of effect. In an experimental study that randomized college students to drink alcohol or placebo, Wray et al. found no relationship between intoxication and change in experiential discounting task scores, or between EDT and sexual arousal or intention to have unprotected intercourse. The literature focusing on pregnancy was smaller. None of the studies directly examined unintended pregnancy. Chesson et al. included history of pregnancy among the surveyed outcomes. They found that higher discount rates were associated with being pregnant before or at the time of the study, or impregnating someone. Jarmolowicz assessed sexual risk behavior with 23-items that included history of pregnancy. Using estimates for indifference points for delay of both sex and money, they found significant correlations between area under the curve for sexual discounting and the number of times both men and women reported having a pregnancy scare. In contrast to the other work, Schmidt’s study focused on fertility and timing of first birth. She hypothesized effect modification by age on the relationship between risk tolerance and fertility timing, noting that more risk tolerant young women may experience earlier birth than do less risk tolerant young women, as risk tolerance would be related to ineffective use of contraception.
Conversely, later in life, risk tolerance may be associated with delayed fertility. Women who are more tolerant to risks may be confortable postponing childbearing until later ages when risks of infertility rise. She found evidence of the positive relationship between fertility timing and risk tolerance among married and unmarried women under 20. She also found that among unmarried, college educated women, risk tolerance appears to delay first births at the end of the fertile period. Several studies focused on men only, or women only. Others reported stratified results that indicated these relationships may be different among men and women. Jarmolowicz, for example, found that delay discounting was associated with number of pregnancy scares for both genders, but the relationship of discounting with total number of lifetime sexual partners, number of instances of having sex while under the influence of alcohol and/or drugs was only significant for women. Their work suggests women may devalue delayed sex less rapidly than men. Johnson and Bruner note that a trend of gender differences in the relationship of sexual discounting and sexual behavior, but do not have the power to examine differences. Collado specifically focused on gender, where males showed higher sexual delay discounting relative to females.The studies in this review revealed consistent associations between economic preference measures and sexual and reproductive behavior. In particular, the literature on temporal preferences suggests that high rates of discounting serve as an important marker of sexual risk behavior. Preferences shifted from condom protected sex to unprotected sex as the delay to sexual behavior increased, suggesting that discounting may relate to willingness to engage in unprotected sex despite overall preference for condoms. This is a similar preference shift from long-term health and disease avoidance to short-term reward as the delay horizon increases, as seen in other discounting and heath literature. The review also supports the conclusion that discounting interacts with substance use, particularly alcohol use, to produce heightened risk for impulsive sexual behavior. Another consistency across studies was the finding that domain specific discounting better predicted sexual behavior than the monetary discounting measure .
The relationship of the SDT to sexual behavior was confirmed in several of the other studies reviewed here. This commodity specific discounting has also been seen in substance use, indicating that domain specific measures are more proximal and relevant. While the SDT attempts to model complex situations, it is limited to hypothetical sexual partners and decisions. The literature that uses hypothetical delays states that they are comparable to real rewards in producing estimates of individual differences in outcomes. Future research should expand on this work to model complex choice environments. The literature on pregnancy, while smaller,indoor grow cannabis also suggests an association between risk and temporal preferences and younger age at first birth and number of pregnancy scares. This indicates that HIV and pregnancy risk behavior may share proximate pathways from discounting and risk tolerance to sexual behavior. Further, the literature on sexual behavior provides insight into possible connections with reproductive health that should be further explored.Thus the relationship between substance use and discounting may inform unintended pregnancy interventions. Probability discounting can apply to the literature showing that the perception of pregnancy risk influences behavior and risk of unintended pregnancy. If substantial uncertainty exists surrounding the risk of pregnancy, women may be more likely to prefer current rewards when future consequences from pregnancy. The concept could also interact with individual risk tolerance, as described in the Schmidt article, to produce heterogeneity in pregnancy risk behavior even within groups with the same pregnancy intention and access to resources. Studies that attempted to unpack the relative importance of delay and uncertainty could be expanded to include pregnancy related attitudes. Inclusion criteria for the review required that the study assess a sexual health behavior or outcome directly. As a result, several studies that included only sexual delay discounting as an outcome and not a direct measure of behavior were excluded. For example, one study examined sexual delay discounting among opioid dependent women, compared to control. Opioid dependent women discounted condom protected sex more steeply than controls in the each of the four partnership conditions. This review finds that while there are consistent associations between economic preference measures and sexual and reproductive health behaviors, there is little evidence to establish a causal effect. Many of the studies were correlational in nature and did not include sufficient control of potential confounding variables. As Brodbeck noted, it is unclear whether an underlying risk propensity is a common cause for both sexual risk behavior preference measures, or if preferences could be manipulated to reduce risky behavior. This work shows that heterogeneity in the way people make choices about temporal trade offs and perceptions of risk and uncertainty may provide insight into their reproductive behavior. More work is needed to establish mechanisms. Given that there is a relationship with monetary discounting, as well as sexual discounting, it should be considered that there is a generalized temporal trade off process in sexual behavior. This apparent preference–behavior incongruence is not only explanatory but also offers potential targets of intervention. The results of the SDT papers imply that condoms should be make readily available, as people are more likely to prefer immediate condomless sex under heightened arousal or when using substances. This same conclusion could be applied to contraceptive use, with similar implications for easily available methods or long acting methods among people more likely to discount sex.
Contraceptive use that require daily maintenance or use at the time of sex may be more challenging for high discounters. The discounting and risk tolerance measures could potentially be applied in a clinical context, to identify those at highest risk of engaging in unprotected sex. While the SDT consistently showed a relationship with sexual behavior in the experimental studies, the task may have less utility in a screening setting. It requires more time to administer than other tools, and the hypothetical partner and choice experiments may be less relevant for women making a contraceptive decision. More work is needed to establish if these tools have utility for screening and prevention. While excluded from this review because it did not include a preference assessment measure, a study by Hiel and colleagues gives a potential framework for interventions leveraging discounting. They provided a decision making tool and financial incentives to reduce barriers to contraceptive uptake and continuation among a population at high risk of unintended pregnancy. Their experiment showed marked differences in contraceptive use and continuation, and trends in pregnancy rates, even in a small sample. Even people who are clear on their stated intentions may make different decisions in different settings. Risk reduction interventions may prove difficult because of a disconnect between intentions and what individuals would do in contexts where contraception is unavailable. Understanding how decision-making could be influenced by these other contextual factors would be very important in tailoring intervention strategies. Indeed the studies in this review suggest that situations factors, such as attraction or partners, alcohol and drug use, and may alter preferences and biases. Unintended pregnancy, contraceptive use and pregnancy decision-making result from a multidimensional set of economic, cultural, social, psychological and demographic determinants. I believe that there is rich potential to include risk and temporal preference constructs into social epidemiologic models. The differences in STI incidence, unintended pregnancy, and contraceptive use by race, class and social environment are well established. Yet the pathways through which these structural factors exert influence are less explored. Two of the articles reviewed include a discussion of the contextual factors that produce differences in discounting. They discuss discounting as a potential mediator between social environment and sexual risk behavior. The environmental risk factors that influence discounting and risk behavior are not well explored. Indeed, there is a debate among scholar as to the stability of these preferences and whether they constitute trait or state characteristics. Odum argues that delay discounting is a personality trait that influenced through environmental circumstances. The findings for parent-child relationship quality, also inly that discounting is responsive to social environment.The review leads to several considerations for future research. Pregnancy risk should be explicitly included in research, so that we may understand how the behavioral processes may be different from those involved in STI risk. Certainly, among young adults, the dual risks of STI and pregnancy that are associated with unprotected sex should be further explored.