First author, year | Study type and population | Study summary | Rates of condom use/unprotected sex | Rates of other sexual risk behaviours |
DiClemente, 201432 | RCT 13–17-year-old African-American females (n=188) | Assessed efficacy of an HIV/STI risk reduction intervention compared with control at 3–6 months after release. | No statistically significant effect of the intervention for consistent condom use, unprotected vaginal sex, or proportion of condom-protected sex acts at the 3- or 6-month assessments, or across the 6-month follow-up period. | No statistically significant effect of the intervention for number of vaginal sex partners at the 3- or 6-month assessments, or across the 6-month follow-up period. |
Freudenberg, 201033 | RCT 16–18-year-old males (n=397) | Compared the efficacy of an intensive 30 hours jail/community-based intervention vs a single jail-based discharge planning session in reducing drug use, ‘sexual risk behaviour’, and criminal activity after release | Measured condom use and having sex while on drugs/alcohol within ‘sexual risk behaviour variable. Rates of condom use not reported; OR for ‘sexual risk behaviour’ variable reported (see next column). | At 1-year post-release, those who participated in intervention sessions and used community-based services were 1/3 as likely to engage in sexual risk behaviour (OR 0.34, p<0.05), compared with participants who only attended intervention sessions in jail. |
Lawrence, 199934 | RCT Adolescent male offenders with a mean age of 15.8 years (n=361) | Evaluated efficacy of an HIV/STI risk reduction training intervention and an anger management intervention (control) at 6 months after release in Mississippi. | Participants in the HIV/STI risk reduction training intervention showed more positive attitudes about condoms and significantly higher condom use skills, compared with the participants in the anger management intervention. | No significant effect of HIV/STI risk reduction intervention on means and repeated measures ANOVA results of sex in combination with alcohol or drugs. Significant decrease in means and repeated measures of ANOVA results of sexual risk behaviours observed in both groups. For example, number of sexual partners in past 3 months decreased from 7.1 to 3.1 in the HIV/STI risk reduction intervention and from 6.9 to 5.2 in the control group (p<0.05). |
Magura, 199440 | RCT 16–19-year-old males (n=157) | Assessed likelihood of HIV risk behaviour after education programme during incarceration and 5 months after release. | Sexual risk behaviours reported at arrest (ie, baseline) for period 6 months prior to arrest, in intervention and control groups:
Rates of sexual risk behaviours after release not reported. Adjusted means of outcome measures after release for general condom use was higher in the education group, compared with control (education 2.0, control 1.3; p=0.002). | Sexual risk behaviours reported at arrest (ie, baseline) for period 6 months prior to arrest in intervention and control groups:
Rates of sexual risk behaviours after release not reported. Adjusted means of outcome measures after release compared with arrest for engaging in sex with high-risk partners was lower in the education group, compared with control (education 0.01, control 0.11; p=0.06). There was no statistically significant difference between the groups for adjusted means of outcome measures for engaging in sex with multiple partners. |
Needels, 200543 | RCT 16–18-year-old males (n=706) | Assessed reduction in HIV risk behaviour at 1 year post-release after use of case management services during discharge planning. | Mixed results observed, with differences by gender. Reported mean number of times having unprotected anal or vaginal sex in past 30 days:
| Mixed results observed, with differences by gender. Percentage of participants reporting engaging in the following sexual risk behaviours (in past 30 days): Had sex four or more times under influence of drugs and/or alcohol:
Had sex with four or more sexual partners:
Gave sex for money/ drugs
|
Ramaswamy, 201044 | RCT 16–18-year-old males (n=397) | Compared the efficacy of intensive jail and community-based intervention vs a single jail-based discharge planning session in reducing drug use, sexual risk behaviour, and criminal after release. | NA | Type of sexual partner 12 months after release: 21.8% short-term sex partner only 41.3% both long- and short-term partners simultaneously Long-term partnership was associated with decreased likelihood of having sex while high on drugs/alcohol (OR 0.14, p<0.001) |
Robertson, 201131 | RCT 12–17-year-old females (n=46) | Assessed efficacy of health education vs HIV risk reduction programme in reducing sexual risk behaviour after release from a Mississippi correctional facility. | There was a statistically significant decrease in frequency of unprotected sex occasions in last 3 months among both groups at 9-month follow-up.
| There was a statistically significant decrease in frequency of engaging in sex under the influence of drugs or alcohol in last 3 months among both groups at 9-month follow-up.
|
Rosengard, 200845 | RCT Adolescents, unspecified age and gender (n=114) | Evaluated impact of motivational enhancement of substance abuse treatment vs relaxation training on frequency of sex without condoms 3 months after release from a correctional facility in the Northeast. | Percentage of participants who reported having sex without a condom in past 90 days prior to incarceration: 52.6%. Percentage of participants who reported having sex without a condom while using alcohol: 53.3% never, 36.7% sometimes, 10% always. Percentage of participants who reported having sex without a condom while using marijuana: 28.3% never, 30% sometimes, 41.7% always. Regression analyses demonstrated that among participants with fewer baseline depressive symptoms, those in the motivation enhancement intervention compared with the relaxation intervention had high rates of condom use overall and lower rates of condom non-use involving marijuana. | Described reports of unprotected sex while using alcohol or marijuana (see column to the left). |
Rowe, 201630 | RCT Adolescents with histories of substance use, unspecified age and gender (n=154) | Evaluated impact of MDFT compared with usual care (ESAU) on reduction of HIV/STI risk behaviour among after release. | Average number of unprotected sex acts in the last 90 days was significantly reduced at both sites for both conditions from intake to 9-month follow-up (Site A: 9-month follow-up mean unprotected sex rate 14.2% for MDFT group and 15.3% in ESAU, compared with 21.8% and 18.6% at baseline). | MDFT group showed a greater reduction in number of sexual acts compared with ESAU group between intake and 9-month follow-up at Site A (Site A: 9-month follow-up mean number of sex acts 19.8% for MDFT group and 19.7% in ESAU, compared with 31.2% and 25.1% at baseline). There was no intervention effect at Site B. |
ESUA, community substance use treatment services; MDFT, multidimensional family therapy; NA, not applicable; OD, odds ratio; RCT, randomised controlled trial; STI, sexually transmitted infection.