Examining delay discounting of condom-protected sex among opioid-dependent women and non-drug-using control women

Abstract

Background: Opioid-dependent (OD) women tend to engage in unprotected sex with high-risk partners, placing themselves at elevated risk for sexually transmitted HIV infection. This behavior generally persists after completion of interventions that increase sexual HIV risk reduction knowledge and skills, suggesting that decision-making biases may influence HIV transmission among OD women.

Methods: The primary aim of this report is to examine delay discounting of condom-protected sex among OD women and non-drug-using control women using the novel Sexual Discounting Task (SDT; Johnson and Bruner, 2012). Data were collected from 27 OD women and 33 non-drug-using control women using the SDT, a monetary discounting task, and the Barratt impulsiveness scale (BIS-11).

Results: OD women discounted the value of delayed condom-protected sex more steeply than controls for hypothetical sexual partners in the two sets of paired partner conditions examined. Overall, women discounted condom protected sex more steeply for partners they perceived as being lowest STI risk vs. those they perceived as being highest risk. Steeper discounting of condom-protected sex was significantly associated with higher scores on the BIS-11, but not with discounting of money.

Conclusions: Delay discounting of condom-protected sex differs between OD women and non-drug-using women, is sensitive to perceived partner risk, and is correlated with a self-report measure of impulsivity, the BIS-11. The effect of delay on sexual decision-making is a critical but underappreciated dimension of HIV risk among women, and the SDT appears to be a promising measure of this domain. Further investigation of these relationships is warranted.

Keywords: Delay discounting; HIV/AIDS; Opioid dependence; Women.

Figure 1

Example delay trial (1 hour) from the Sexual Discounting Task showing the hypothetical sexual partner selected by a study participant for the MOST SEX partner condition and the corresponding visual analog scale (VAS) below. The arrow above the VAS and the percentage value to the right indicate the participant’s selected likelihood value (68%).

Figure 2

a. Median standardized likelihood values for each SDT partner condition with best-fit two-parameter hyperbolic functions (V=A/(1+kD)^s) for OD women (n=33) and non-drug-using control women (n=27). b. Individual participant’s standardized area under the curve (AUC) values from the SDT organized according to partner condition (MOST SEX, LEAST SEX, LEAST STI, and MOST STI) and group (OD women vs. non-drug-using controls). Horizontal lines indicate median AUC values for each partner condition/group. Two-way ANOVA indicated that OD women discounted condom-protected sex more steeply than controls for MOST SEX/LEAST SEX partners [F(1,54)=4.11, p=.048] and for LEAST STI/MOST STI partners [F(1,54)=7.50, p<.01]. Overall, women discounted condom-protected sex more steeply for LEAST STI partners than for MOST STI partners [F(1,43)=4.87, p=.03]. There were no overall differences between MOST SEX and LEAST SEX partners [F(1,43)=2.23, p=.14].

Figure 2

a. Median standardized likelihood values for each SDT partner condition with best-fit two-parameter hyperbolic functions (V=A/(1+kD)^s) for OD women (n=33) and non-drug-using control women (n=27). b. Individual participant’s standardized area under the curve (AUC) values from the SDT organized according to partner condition (MOST SEX, LEAST SEX, LEAST STI, and MOST STI) and group (OD women vs. non-drug-using controls). Horizontal lines indicate median AUC values for each partner condition/group. Two-way ANOVA indicated that OD women discounted condom-protected sex more steeply than controls for MOST SEX/LEAST SEX partners [F(1,54)=4.11, p=.048] and for LEAST STI/MOST STI partners [F(1,54)=7.50, p<.01]. Overall, women discounted condom-protected sex more steeply for LEAST STI partners than for MOST STI partners [F(1,43)=4.87, p=.03]. There were no overall differences between MOST SEX and LEAST SEX partners [F(1,43)=2.23, p=.14].

Figure 3

a. Median standardized indifference points from the MDT with best-fit two-parameter hyperbolic functions (V=A/(1+kD)^s) for OD women (n=33) and non-drug-using control women (n=27). b. Individual participant’s standardized area under the curve (AUC) values from the MDT organized by group (OD women vs. non-drug-using controls). Horizontal lines indicate median AUC values for each group. One-way ANOVA on the ranks indicated that OD women discounted the value of money more steeply than control women [F(1,55)=6.66, p=.012].

Figure 3

a. Median standardized indifference points from the MDT with best-fit two-parameter hyperbolic functions (V=A/(1+kD)^s) for OD women (n=33) and non-drug-using control women (n=27). b. Individual participant’s standardized area under the curve (AUC) values from the MDT organized by group (OD women vs. non-drug-using controls). Horizontal lines indicate median AUC values for each group. One-way ANOVA on the ranks indicated that OD women discounted the value of money more steeply than control women [F(1,55)=6.66, p=.012].