Safer Sex: Information for Counselors (Part VII)

References - Condoms

Knowledge, Attitudes, Practice

(Arranged Chronologically)

Condom beliefs in urban, low income, African American and Hispanic youth.

Norris AE. et al.
Health Educ Q 1994 Spring;21(1):39-53

This article focuses on the condom beliefs of low income, urban African American and Hispanic youth living in the Midwest. The condom beliefs under investigation were derived from prior research with members of this population and through consultation with African American and Hispanic youth and service providers. Significant gender, ethnic, and acculturation differences were found among beliefs related to frequency of condom use in the past year (p < .05). These differences indicated that women, African American respondents, and Hispanic respondents high in acculturation tended to have more neutral or more positive views about condoms than other types of respondents.

Gender differences in AIDS-relevant condom attitudes and condom use.

Sacco WP. et al.
AIDS Educ Prev 1993 Winter;5(4):311-26

Two studies, conducted approximately one year apart, examined gender differences in AIDS-relevant condom attitudes, condom use behaviors, and relationships among attitudes and condom use behaviors. Subjects (N = 248, N = 528) were undergraduates, primarily heterosexual. Females reported more favorable attitudes, with the exception of greater inhibition about buying and possessing condoms. Men engaged in preliminary condom use behaviors (carrying and keeping condoms at home) substantially more often than did women. Preliminary condom use behaviors predicted past and intended condom use more consistently for men than for women. Relationships between condom attitudes and condom use behaviors were generally similar for both sexes, with poorer self-control explaining the most variance in past and intended condom use. These results, interpreted from the perspective of Eagly's (1987) gender role theory, suggest that although females may indirectly influence condom use decisions, providing condoms is the expected role of males, infusing them with greater control over the interpersonal process.

Acculturation and gender differences in sexual attitudes and behaviors:Hispanic vs non-Hispanic white unmarried adults.

Marin BV. et al.
Am J Public Health 1993 Dec;83(12):1759-61

Understanding how acculturation and gender affect Hispanics' sexual behavior is needed to prevent infection with the human immunodeficiency virus. We examined differences in and correlates of condom use among 398 Hispanics and 540 non-Hispanic Whites in San Francisco who were part of a random probability sample of unmarried adults. Hispanic women reported fewer sexual partners than all other groups. Condom use was low in all groups, but Spanish-speaking Hispanic women reported lower condom use than White women. Hispanics, generally, had poorer attitudes toward condoms and were less likely than non-Hispanic Whites to believe they could avoid acquired immunodeficiency syndrome. Hispanics need targeted prevention interventions.

HIV infection and the meaning of condoms.

Pivnick A
Cult Med Psychiatry 1993 Dec;17(4):431-53

Present day meanings associated with condom use among drug-using women and their long term sexual partners include: loss of male protection; violations of constructions of intimacy, fidelity, conjugal bonding, and female identity; illness; and death. Efforts to change condom use patterns must be based on increased vocational, educational, and social opportunities for women. HIV prevention efforts might profit from the design of safer sex interventions informed by the above meanings; intensified mass media campaigns; and a re-contextualization of attitudes, behaviors, and beliefs.

Can condom users likely to experience condom failure be identified?

Steiner M. et al.
Fam Plann Perspect 1993 Sep-Oct;25(5):220-3, 226

A study based on a convenience sample of 177 couples who each used 11 condoms found that 103 condoms (5.3%) broke before or during intercourse and 67 condoms (3.5%) slipped off during sex. Couples who had not used a condom in the past year were almost twice as likely to experience condom failure as were couples who had used at least one during that period (p < .001). Of the couples who had used a condom in the previous year, the failure rate among those who reported at least one condom break during that period was more than twice the failure rate among those who reported no breaks (p < .001). Among couples who had used condoms in the past year without breaking any, those who did not live with their partner and those who had a high school education or less were at increased risk of condom failure (adjusted odds ratios of 3.2 and 2.7, respectively).

The theory of reasoned action and cooperative behaviour: it takes two to use a condom.

Kashima Y. et al.
Br J Soc Psychol 1993 Sep;32 ( Pt 3):227-39

The applicability of the Theories of Reasoned Action and Planned Behaviour to the cooperative behaviour of condom use were examined. Seventy-one male and 78 female students, all sexually active unmarried heterosexuals aged 17 to 21 years, gave information about their intentions for the next sexual encounter, as well as their attitude, subjective norm, expectancy-value attitude and subjective norm (including normative beliefs for their sexual partner), and their past behaviour with respect to condom use. After their next sexual encounter, they completed a questionnaire on their actual condom use. Results indicated that when behavioural conditions including the availability of a condom and an agreement with the partner to use it were satisfied, intention interacted with past behaviour to predict actual behaviour. These results imply that intentions which are consistent with past behaviour are stable enough to be carried out in the face of the interpersonal dynamics of a sexual encounter. Further, normative belief for the sexual partner had a direct influence on attitudes, subjective norm and intention. Neither the Theory of Reasoned Action nor the Theory of Planned Behaviour can fully explain these results, which point to the need for further theoretical inquiry into the dynamics of cooperative behaviour.

U.S. Young adults have lots of sex, lots of partners, yet refuse to use condoms: the National AIDS Behavioral Surveys.

Binson D. et al.
Int Conf AIDS. 1993 Jun 6-11;9(2):690 (abstract no.

OBJECTIVE: To assess prevalence and demographic correlates of high risk sexual behavior among young heterosexual adults in the USA. METHODS: The National AIDS Behavioral Surveys (NABS: N = 10,630) is a national probability sample of US residents (response rate = 70%) and a probability sample of urban residents in 23 major cities with high AIDS prevalence (High Risk Cities sample: response rate = 65%). We examined prevalence and demographic correlates of AIDS-related risk behaviors among young adults. 18 to 25 years of age, (n = 1,201) in the High Risk Cities (54% white, 24% African American, 17% Latino American, 5% other ethnic groups). RESULTS: 30% of young adults engaged in risky behavior over the last 12 months and only 20% of these used condoms consistently. 65% engaged in risky behavior over the last 5 years. The breakdown by risk factor (behaviors assessed for last year) for men and women at risk is shown below: TABULAR DATA, SEE ABSTRACT VOLUME. Men, the unmarried, and those 18 and 19 were the most likely to have multiple sex partners. Among Whites and African Americans, young people with education beyond high school were more likely to have multiple partners than those with less education. Of those with multiple sex partners, approximately 40% never used condoms with either primary or secondary partners. As the number of secondary partners increased, condom use decreased. CONCLUSION: In the US, prevention proposals targeted at young adults typically tie protection from AIDS to monogamous commitment. These fail to capture the realities of young adults and are bound to fail. Even worse, these messages are misleading because monogamy for young adults is short-lived.

Widespread condom use by seropositive injecting drug users with non-injector sexual partners.

Friedman SR. et al.
Int Conf AIDS. 1993 Jun 6-11;9(2):691 (abstract no.

OBJECTIVE: To assess whether condom use by injecting drug users (IDUs) differs by HIV serostatus and/or by sex partner (SP) characteristics. METHODS: 293 street-recruited New York IDUs (69% male) were HIV tested and asked about their relationships with up to 10 SPs each. Analyses focus on condom use during the prior 30 days in 386 sexual partnerships. 28 linked pairs of IDUs who are each others' SPs were compared for data validation. RESULTS: 35% of respondents were seropositive. 25 of 28 (89%) linked pairs agreed on the proportion of sex acts in which they used condoms. Always using condoms is reported in 33% (128/386) of relationships. The table shows proportions of relationships in which condoms are always used: TABULAR DATA, SEE ABSTRACT VOLUME. In multivariate logistic regression controlling for gender, closeness of relationship, and other variables, condoms are always used more in relationships between seropositive IDUs and non-IDU SPs than in other relationships (odds ratio = 4.59; 95% CI = 1.57, 13.41). CONCLUSIONS: Condom use by IDUs in New York, with its relatively mature epidemic, is concentrated where it may most reduce the spread of HIV to non-IDU heterosexuals--in relationships between infected IDUs and non-IDU partners. This may reflect IDUs' altruism and/or non-IDU SPs' pressure; prevention projects should build upon both. Mathematical modelers should incorporate differential condom use by serostatus and partner's drug use.

Employing the health belief model to predict condom use.

Toepell AR
Int Conf AIDS. 1993 Jun 6-11;9(2):952 (abstract no.

This study tested a series of hypotheses related to the use of condoms as a means of preventing HIV-infection among 100 adult male prisoners at two correctional facilities in Toronto. The hypotheses were derived from the Health Belief Model (Rosenstock, Becker), a model chosen because of its disease-avoidant orientation. Condom use prior to incarceration was investigated. Each of the model's six variables (knowledge, perceived benefits, barriers, susceptibility and severity, and cues to action) were tested in an attempt to predict condom use. Categorized sexual partner types included: steady, repeated, casual and anonymous partners. Condoms were generally used with casual and anonymous partners, and not with steady partners. The model proved only to be useful in predicting condom behaviour after the factor of partner type was included in the analyses (a variable not accounted for in the original model). The Health Belief Model, on its own, was not a vigorous test for predicting condom use. It was dependent on strongly influential factors unrelated to its own measured variables and could not rely on its established variables to explain or predict such behaviour. Also, the model was limited in its ability to predict health preventive behaviour adequately. This model must be continuously adapted to reflect changing attitudes and beliefs held concerning health and health preventive behaviour. Further, the inclusion of measures of self-efficacy and intention of behaviour would improve the predictive capabilities of the Health Belief Model.

Perceptions of condom use among STD clinic clients by gender U.S. 1991.

Walsh C. et al.
Int Conf AIDS. 1993 Jun 6-11;9(2):951 (abstract no.

OBJECTIVE: To examine perceptions of condom use among STD clinic clients. METHODS: An interviewer-delivered survey was conducted among 1701 STD clinic clients in three U.S. cities. To maximize the chance of identifying clients with risk behaviors of interest, an initial self-administered risk assessment questionnaire was used. Targeted risk behaviors were: 1) men with multiple partners, 2) men with crack-using partners, 3) men who used cocaine, 4) women with multiple partners, and 5) women who used cocaine. From that assessment, eligible clients participated in an in-depth interview. RESULTS: Of the 967 respondents with self-reported risk behaviors of interest, 647 (67%) were male and 320 (33%) were female, 75% were African American, 16% were white, and 6% identified as Hispanic. Only 29% (93/316) of the women replied that it was true or somewhat true that men liked using condoms, while men replied that 39% (249/636) of the men they knew liked using condoms. Fifty-six percent (177/316) of the women believed that using a condom is a turn-off for their partners, while only 30% (293/636) of the men believed this is true for their partners. Sixty percent (189/317) of women stated that it was true or somewhat true that men they knew thought it was a good idea to use a condom with a new sex partner. Seventy percent (448/640) of the men in the survey responded that men they knew agreed with that statement. In fact, 71% (222/314) of women believed it was difficult to find a sex partner who wants to use a condom, while only 53% (341/646) of men believed that to be the case. However, when only men were questioned, 91% (587/646) replied that using condoms showed respect for their partners. CONCLUSION: Women and men view each other's perceptions of condom use differently: women believe that men are less willing to use condoms than men report they are. Therefore, HIV prevention counseling messages should be tailored to take advantage of these perceived gender-specific differences, i.e. that women may be in a stronger position to negotiate condom usage with their sex partners and that men should be encouraged to make their willingness to use condoms more evident to their sex partners.

Trust as a barrier to condom use with steady vs. casual partners among New York City IDUs & sex partners.

Wolfe H. et al.
Int Conf AIDS. 1993 Jun 6-11;9(2):824 (abstract no.

OBJECTIVES: To examine context-specific factors associated with condom use among injecting drug users (IDUs) & sexual contacts. METHODS: The 460 heterosexually active subjects analyzed herein (26% female, 22% Caucasian, 36% African-American, 40% Latino, 20% HIV+, mean age 38) were recruited in late 1992 through 3 NYC methadone clinics & a community outreach program. Trained interviewer-phlebotomists drew blood & interviewed subjects regarding risk behavior, & condom use in their most recent sexual encounter with a) a steady partner, b) a casual partner. RESULTS: With steady partners, 32% of men & 28% of women used a condom; with casual partners, 58% of men & 55% of women. Partner status is more important than gender in predicting condom use. The most often- cited reason for not using a condom with steady partners: for men & women, trust in partner (55% & 53%). With casual partners, the top reason cited by men was condom dislike (36%), by women, forgetting or condom non-availability (48%). The second most often-cited reason by women with steady & casual partners was partner dislike of condoms (19% & 14%). Condom use did not differ by race, except that Latino men were less likely to use condoms with steady partners. CONCLUSIONS: Less than one-third of subjects in this high-risk population report using condoms with steady partners. Reasons for condom non-use differ significantly by gender and partner status. Condom promotion efforts potentially violating relational trust are bound to fail; a universal precautions approach de-emphasizing trust & betrayal is needed to promote condom use in couples. Study of the relationship between dyadic trust & sexual behavior is underway.

Condom characteristics: the perceptions and preferences of men in the United States [see comments]

Grady WR. et al.
Fam Plann Perspect 1993 Mar-Apr;25(2):67-73

Perceptions regarding the consequences of condom use, as well as preferred characteristics of condoms, are examined in a nationally representative sample of 3,321 men aged 20-39. The psychological and interpersonal effect most often cited is that using a condom shows that you are a concerned and caring person. This is particularly the case among black men and men who are young and have low educational attainments. However, the same men usually agree that using a condom sends unwanted messages to one's partner--for example, that doing so makes your partner think that you have AIDS and shows that you think that your partner has AIDS. In contrast, white men and those who are highly educated tend to cite embarrassment when buying condoms as a frequent consequence of condom use. The device-related consequences cited most often, particularly among black, unmarried, young and poorly educated men, are that using a condom results in reduced sensation, that one must be careful during sex or the condom may break and that one must withdraw quickly after sex or the condom may come off. When purchasing condoms, most men look for those that are easy to put on, have the right amount of lubrication and stay on; these preferences are particularly prevalent among black men. Few men identify color, ribbing and partner's preference for condom type as important.

Condom use among U.S. men, 1991 [see comments]

Tanfer K. et al.
Fam Plann Perspect 1993 Mar-Apr;25(2):61-6

A 1991 study of a nationally representative sample of men aged 20-39 finds that 27% of sexually active men had used a condom in the four weeks before interview. Black men are more likely than white men to report condom use (38% vs. 25%), and men younger than 30 are more likely to do so than are those older than 30 (36% vs. 19%). Among white men, condom use increases with years of education; among black men, however, those with 12 years of education are much less likely to report condom use than are those with more or less than 12 years (28% vs. 43-50%). Condom use is positively related to number of partners. Men who have engaged in anal intercourse, those who have had a one-night stand and those who are bisexual or homosexual are also more likely to report condom use. Among those who reported using a condom in the previous four weeks, 55% of whites and 18% of blacks had done so only for birth control and 7% of whites and 9% of blacks had done so only for protection against infection with the human immunodeficiency virus and other sexually transmitted organisms; the remainder had used a condom for both reasons.

Factors associated with condom use in a high-risk heterosexual population.

Weinstock HS. et al.
Sex Transm Dis 1993 Jan-Feb;20(1):14-20

The use of condoms has been advocated as a means of preventing the transmission of the human immunodeficiency virus and other sexually transmitted agents. To better understand factors that may influence condom use, 300 heterosexuals were enrolled in a cross-sectional study of patients attending San Francisco's only public sexually transmitted disease clinic. Interviewer- administered questionnaires were conducted. Condom use at last sexual intercourse was examined by logistic regression analysis. Men who used drugs or alcohol at last intercourse and whose partners did not want to use condoms were less likely to have used them; women who were black or Hispanic, who reported difficulty getting their partners to use condoms, or who reported that condoms decrease sexual pleasure also were less likely to have used them. Efforts to increase condom use in this population should target minorities, assist women to negotiate their use, emphasize the dangers of using alcohol and other drugs with sex, and address the perception that condoms interfere with sexual pleasure.

Condom use in U.S. Hispanic men with multiple female partners.

Gomez CA. et al.
Int Conf AIDS. 1992 Jul 19-24;8(3):168 (abstract no. PuC

OBJECTIVE: To describe condom use and its predictors in Hispanic men with multiple female sexual partners. METHOD: In nine states of the U.S. with high proportions of Hispanics, a stratified clustered random telephone sampling strategy was used to identify Hispanic households. Interviews of 968 Hispanic males, aged 18 to 49 were conducted resulting in 521 (53.8%) men who reported having one female partner and 371 (38.3%) men with more than one female partner in the prior 12 months. Frequency of condom use with steady and causal partners in the 12 months prior to the interview was measured on a 5-point always-never scale. RESULTS: Among Hispanic men with multiple female partners, 60% reported always using condoms with their casual partners while only 18% always used with their steady partner. Carrying condoms, self-efficacy to use condoms, positive attitude towards use, having friends who used, and lack of depression in the two weeks prior to the interview predicted casual condom use (R2 = .35), whereas drug and alcohol use before sex did not. Comfort with sex significantly predicted carrying condoms, attitude toward use, and self-efficacy to use them. CONCLUSION: High rates of condom use with causal partners among high risk Hispanic heterosexual men suggest that programs to promote condom use with casual partners can succeed with this population. These programs should address the specific skills needed to carry and use a condom with a causal partner, and increase comfort with sexuality in general.

Gender differences in condom use following educational interventions among drug users.

McCusker J. et al.
Int Conf AIDS. 1992 Jul 19-24;8(2):C374 (abstract no. PoC

OBJECTIVES: To identify variables which were associated with increased condom use with heterosexual partners following an educational intervention for drug users. METHODS: Participants in a controlled evaluation of several AIDS educational interventions in a drug detoxification program were interviewed during treatment and again at 3 months after discharge, on average. 301 subjects with heterosexual contact both at baseline and follow-up were included in this study. Data were collected on both occasions on drug use and sexual behaviors during the previous 3 months, and on attitudes and beliefs and social variables relating to condom use. RESULTS: At baseline, 34% ever used condoms. Condom use was associated with multiple partners especially among women. Among those who were not always using condoms (n = 271), 21% increased condom use at follow-up. 38% of women and 31% of men reported multiple partners; these percentages did not change at follow-up. Variables associated with increased condom use differed by gender. Among men, these variables included awareness of HIV(+) status, perception of greater self-efficacy to use condoms and benefits of behavior change, social support for condom use, and no use of alcohol or drugs during sex. Among women, these variables included multiple partners and engaging in sex for money or drugs. Type of education received and most demographic variables were not associated with increased condom use. CONCLUSIONS: Attitudes about condoms were generally more strongly associated with increased condom use among men than women, perhaps reflecting men's greater control over whether condoms are used. Women may experience greater difficulty in introducing condoms into a primary relationship, but find this easier with casual partners.

Effects of communication with important social referents on beliefs and intentions to use condoms.

Strader MK. et al.
J Adv Nurs 1992 Jun;17(6):699-703

Data from a 1989 survey using the condom attitude and belief instrument with 310 clients from two sexually transmitted disease clinics identified significant social referents who influence condom-use intentions. They are sexual partner, father and friends. The present study found that communication with these referents had a positive net effect on beliefs about and intentions to use condoms. Implications applicable for intervention programmes to increase condom use are to promote talking about condoms between the sexual partners and important social referents and to develop such communication skills.

Black males who always use condoms: their attitudes, knowledge about AIDS, and sexual behavior.

Johnson EH. et al.
J Natl Med Assoc 1992 Apr;84(4):341-52

One hundred six black males completed a questionnaire concerning attitudes and knowledge about the use of condoms and acquired immunodeficiency syndrome (AIDS). Of the 106 males in the study, 27 (26%) reported that they always used condoms, 31 (29%) did not use condoms and had low intentions of using them, and 48 (45%) reported high intentions to use condoms. Results indicated that knowledge about AIDS was exceptionally high for black males in all three groups. Black males with low intentions to use condoms reported significantly more negative attitudes about the use of condoms (eg, using condoms is disgusting) and reacted with more intense anger when their partners asked about previous sexual contacts, when a partner refused sex without a condom, or when they perceived condoms as interfering with foreplay and sexual pleasure. A significantly larger percentage of low intenders were treated for gonorrhea, syphilis, herpes, and genital warts than males in the other groups. Drug use did not differentiate the three groups, although marijuana was used more often by males in the low-intender group. Finally, a larger percentage of black males in the low-intender group reported experiences with anal intercourse and sex with a prostitute, but considered themselves at lower risk for AIDS than did their high-intender or steady-user counterparts.

Do African-American men and women differ in their knowledge about AIDS, attitudes about condoms, and sexual behaviors?

Johnson EH. et al.
J Natl Med Assoc 1992 Jan;84(1):49-64

This study identified a sample of young African-American men and women classified as having multiple sex partners or one sex partner. Of the 149 men, 71 (47%) were classified as having multiple sex partners and 78 (53%) as having one sex partner. Of the 165 women, 29 (19%) were classified as having multiple sex partners and 126 (81%) as having one sex partner. Results indicated that the groups did not differ in their knowledge about acquired immunodeficiency syndrome (AIDS). However, attitudes about condom use differed significantly by gender (P less than .01) and by multiple sex status (P less than .001). Angry reactions regarding the negotiation of condom use occurred more with men than with women (P less than .05). Men and members of the multiple sex partners group tended to engage in more risky sexual behavior. These two groups also had a significantly (P less than .001) higher incidence of gonorrhea. While the multiple sex partners group had significantly more smokers (P less than .01), drinkers (P less than .01), and crack users (P less than .05), men were significantly higher consumers of marijuana (P less than .001) and alcohol (P less than .01). Overall, the results indicate that African Americans are knowledgeable about AIDS, but there appears to be a gap between knowledge and risky sexual behaviors. Prospective studies are needed to clarify the factors that determine the relationship (or lack of) between knowledge and risky sexual behaviors. Without such studies, a major part of the foundation that is needed for the development of sensitive and effective AIDS prevention programs for African Americans will be missing.

Attitudes toward condom use and AIDS among patients from an urban family practice center.

Stewart DL. et al.
J Natl Med Assoc 1991 Sep;83(9):772-6

As part of an effort to better educate patients about using condoms, a survey was done to assess the sexual practices, attitudes toward acquired immunodeficiency syndrome (AIDS) and condom use by patients who visit an urban family practice center. A self-administered questionnaire was given to 126 patients of whom the majority were black and single. Seventeen percent indicated they had sex with more than one person in the 3 months before the survey. Within the last 5 years, 10% had sex with an intravenous drug abuser, and 6% with someone of the same sex. Fifty percent believed that condoms decrease sexual pleasure for men and 31% thought condoms made sex inconvenient. Twenty-seven percent of those surveyed believed that a man's penis may be too large for a condom and 18% believed that uncircumcised men could not use condoms. Forty-five percent believed they should be screened for human immune deficiency virus exposure. The results highlight attitudes and beliefs that may function as barriers to condom use and should be addressed when encouraging condom use with this population.

Condom knowledge, history of use, and attitudes among chemically addicted populations.

Kramer TH. et al.
J Subst Abuse Treat 1991;8(4):241-6

The need for behavioral change of risky sexual practices has been of the highest priority since the onset of the AIDS epidemic. The major focus of education for safe sex has been emphasis on condom use. We surveyed 124 individuals applying to treatment for various chemical dependencies and 60 individuals applying for non-chemical-dependency medical treatment on various aspects of condom knowledge, history of use, and attitudes. Respondents reported that AIDS has motivated them to increase their use of condoms, however, only 13.9% always use them. Education is needed in the areas of increasing protection. Along with the use of a condom, the need for a reservoir tip and the risks associated with multiple sex partners should be stressed.

What about condoms?: black, hispanic, and white views from a population-based sample.

Kegeles SM. et al.
Int Conf AIDS. 1991 Jun 16-21;7(2):319 (abstract no.

OBJECTIVE: To examine minority and white heterosexual men's and women's attitudes about condoms, self-efficacy to use condoms, and beliefs about condom efficacy. METHODS: The AMEN (AIDS in Multi-Ethnic Neighborhoods) study is a household probability sample of 1770 unmarried adults who resided in high risk census tracts in San Francisco. The variables below were dichotomized at the median split. RESULTS: Chi-squares were used; differences between minorities and whites are indicated. TABULAR DATA, SEE ABSTRACT VOLUME. CONCLUSIONS: Condoms are poorly regarded in these minority communities, more so than among whites. Culturally sensitive interventions should focus on: 1) developing techniques that make condoms more physically acceptable, and 2) increasing positive attitudes about condoms.

Risk reduction among injecting drug users: changes in the sharing of injecting equipment and in condom use.

Klee H. et al.
AIDS Care 1991;3(1):63-73

In an investigation of risk behaviour among injecting drug users in the North-West of England, information was obtained concerning the sharing of injecting equipment, respondent's sexual partners and the use of condoms. Between six and nine months after the initial contact, 169 respondents (56%) were contacted again. The emphasis in the second phase of the project was on changes, if any, in risk behaviour that had occurred in the intervening period. Significant reductions were found in sharing, mostly in the more indiscriminate use of others' injecting equipment. No reduction was observed in sharing between injecting partners and little in sharing between close friends. The number of sexual partners had decreased and the use of condoms, although it increased among those involved in temporary relationships, remained low. Impediments to further progress in risk reduction are discussed.

[email protected] (Tue Mar 21 23:46:00 1995)