Aids And Behavior
1 like1,831 views
This study evaluated risk behaviors in 153 HIV-positive adolescents receiving care in Bucharest, Romania. While most acquired HIV perinatally during Nicolae Ceausescu's regime, 20% reported sexual acquisition, suggesting sexual transmission is increasing. Participants showed deficits in HIV knowledge, attitudes, norms, control, and intentions regarding risk behaviors. Many reported unprotected sex with both HIV-positive and negative partners, despite access to treatment. Results imply Romania's HIV epidemiology is changing, with growing sexual transmission among adolescents. Discussion centers on interpreting these patterns and improving HIV prevention for young people.
![718 AIDS Behav (2007) 11:716–725
questionnaires individually and privately, but could ask Measures
for support from any of four practitioners who were
on-site (e.g., to aid in understanding some of the The inventory inquired about a broad array of health
questions or instructions). Every third patient on the and psychological dimensions; those not indicated
daily list of appointments was recruited; the children’s below were merely for future study development and
social workers or psychological case workers offered not directly relevant to the current study. All mea-
each prospective participant a modest sum (the equiv- sures were screened through a back-translation tech-
alent of $5 in local currency) to complete a question- nique sensitive to cultural differences and respectful
naire concerning ‘‘health behaviors of HIV-positive of the participants’ potential past actions (e.g., the
youth.’’ Nearly all (99%) of the youth informed about questions never implied that such actions should be
the study agreed to participate. dismissed as unacceptable). A team member in Roma-
nia first translated the English versions into Romanian;
Procedure a native-Romanian speaker in the US provided a
back-translation into English. For several items it was
Similar to other Eastern European countries (Goodwin, necessary to repeat the process until all disagreements
Kozlova, Nizharadze, & Polyakova, 2004), Romanian were resolved.
society associates significant stigma with being infected
with HIV in Romania and regards HIV-risk behaviors Mode of Transmission and Status on ART
such as injection drug use and sexual interactions
outside marriage as sinful. Approximately 90% of the Participants were asked ‘‘How do you think you
population in Romania is Christian Orthodox, and got HIV?’’ (multiple responses were possible and
family education places a great emphasis on religious included: having sex; sharing a needle; blood transfu-
aspects, which increases the feeling of shame within the sion; had it at birth) and other details about their
families of people who are infected with HIV.1 With this medication status. Specifically, they were asked if they
pretext in mind, we designed an information sheet for are taking any sort of medication for HIV (Yes; No),
the participants that emphasized not only how the for how long they had been taking medication for
responses they provided could help to understand the HIV, and for how many months they might have been
health of HIV-positive youth but also their freedom not taking any type of combination of different medica-
to answer any item they wished not to answer; they were tions for HIV. All of these participants had current
also assured of the confidentiality of their responses access to antiretroviral therapy (ART; see Kozinetz,
(e.g., ‘‘all of your responses are kept strictly PRI- ˘ ¸
Matusa, & Cazacu, 2001).
VATE’’). The information sheet and questionnaire
explicitly requested that no names or other identifying Risk Deficit Survey
information be provided. Participants then completed
the paper-and-pencil inventory in a private room. After Participants were then asked to select a phrase that
completing the inventory, they placed it in a sealed best describes their living situation [living on the street;
envelope, mixed it with other such envelopes in a box, living in an abandoned building; living in a homeless
were thanked and dismissed. shelter; living in a rehabilitation home or half-way
The procedures in the study were approved both by house (for drug abuse); living in a assisted living
the Institutional Review Board at the University of program for people with medical problems; living with
Connecticut and its counterpart at the University of parents; living with family members (other than par-
Bucharest. ents); living with friends; living in a house, apartment,
or condominium that you rent or own]. A series of
questions about risk behavior followed; participants
were asked if they ‘‘shoot drugs (inject drugs with a
1
During the Sunday services in the Orthodox Church, priests needle) to get high,’’ have ever had a sexual relation-
teach about this message, based both on the Bible and on the
ship with someone, and if they are currently in a sexual
writings of the Apostles. In recent years, the proportion of youth
participating in the church has increased steadily. Although there relationship with someone. Because no individual
is no research to document the implementation of the religious indicated having ever injected drugs, the remainder
messages in the people’s behavior, in Romania there is indeed a of this paper will focus on sexual-risk, which was a far
mentality that associates HIV with sin. It is important to note
more frequent behavior.
that the Church neither explicitly approves nor dismisses
condom use, whereas government policies have consistently The questionnaire was modeled on that utilized by
favored condom use. Kozal et al. (2004). Eleven items assessed knowledge
123](https://image.slidesharecdn.com/aidsandbehavior-091110182751-phpapp02/85/Aids-And-Behavior-3-320.jpg)
Aids And Behavior
- 1. AIDS Behav (2007) 11:716–725 DOI 10.1007/s10461-006-9202-7 ORIGINAL PAPER HIV-positive Romanian Adolescents: Acquisition Routes, Risk Behaviors, and Psychological Correlates Blair T. Johnson Æ Doru Buzducea Published online: 13 February 2007 Ó Springer Science+Business Media, LLC 2007 Abstract This research evaluated risk behaviors in Introduction HIV-positive adolescents (N = 153) receiving routine HIV medical care from their providers at a large The HIV/AIDS pandemic continues to grow, with medical center in Bucharest, Romania. Although so-called hot spots appearing in many African and the participants were of an appropriate age Asian countries as well as many Eastern European (M = 15.94 years) to have acquired HIV as one of the countries (EuroHIV, 2005). Policies of Nicolae large numbers of pediatric AIDS cases during Ceausescu’s Communist regime in Romania, which ¸ Ceausescu’s Communist regime, many (20%) reported ¸ ended in 1989, strongly promoted population growth, acquiring HIV sexually, relatively recently. The sample outlawing contraception and abortion (Hersh et al., as a whole exhibited significant deficits in HIV knowl- 1993; Simon et al., 1999). The result was that many edge, attitude, subjective norm, perceived control, and infants were placed in orphanages; others suffered intention; significant risk behavior also appeared in malnourishment even while living with their natural terms of reports of unprotected intercourse with both parents. Vitamin supplements were often administered HIV-positive and -negative partners. These results along with micro transfusions of whole blood. Unfortu- imply a change in the epidemiological model for HIV nately, a single syringe was often used repeatedly transmission in Romania, with increased sexual trans- without replacement or cleansing and blood products mission of HIV. Discussion centers on interpretations were not at that time screened for HIV. Worsening the of these patterns and drawing implications for HIV risk problem, the Ceausescu government labeled HIV a ¸ reduction for young people in Romania. ‘‘capitalist disease’’ and therefore defined it as not existing in Romania; consequently HIV testing was Keywords Pediatric AIDS cases Á Romania Á deemed unnecessary as well (Danziger, 1996). When Secondary HIV prevention Á Sexual transmission Ceausescu and his government fell, such practices were ¸ of HIV Á Knowledge, attitudes, and behavior ˘ ˘ corrected (Chisevescu, Mihailescu, Mihailescu, & Pasat, 1998), but not before many thousands of children were infected with HIV in medical care, generating intense media scrutiny. Romania has the largest number of pediatric AIDS cases in Europe, over 50% of its cases to date, and among the largest numbers in the world (Danziger, 1996). The cohort of children born between B. T. Johnson (&) Center for Health, Intervention, and Prevention, University 1987 and 1989 suffered the most infections. Unfortu- of Connecticut, 2006 Hillside Road, Unit 1248, Storrs, nately, another legacy of the Ceausescu era is that the ¸ CT 06269-1248, USA country has only recently attempted sex education or e-mail: blair.t.Johnson@uconn.edu sexual health promotion (Bencomo, 2006; Danziger, D. Buzducea 1996; McNeil, 2004). Fortunately, all Romanian chil- University of Bucharest, Bucharest, Romania dren known to be infected with HIV have access 123
- 2. AIDS Behav (2007) 11:716–725 717 to state-paid retroviral treatment (Bencomo, 2006), that he or she controls the behavior in question. This including clinical monitoring and an individualized dimension is assumed to reflect the obstacles that one treatment schedule that includes recent therapeutic encountered in past behavioral performances. Further, advances (e.g., triple drug antiretroviral therapy). intention is one’s willingness to perform the behavior. Owing to regular medical care, many of these The theory holds that intentions are based on attitudes, individuals have survived and are now reaching ado- subjective norms, and perceived behavioral control and lescence and young adulthood. As McNeil (2004) that behavior is based on both intention and per- discussed, Romanian public health officials fear ceived behavioral control. An extensive meta-analysis increased infections because the children infected in concluded that the theory of planned behavior provides the late 1980s are now reaching sexual maturity. Yet, good explanation of condom use behaviors (Albarracın,´ little is known about them or their HIV-risk-related Johnson, Fishbein, & Muellerleile, 2001). Moreover, practices and about how best to prevent them from important theories of HIV prevention have utilized transmitting HIV to others. Due to limited formalized assumptions of the theory of planned behavior (Fisher HIV prevention training for medical professionals who & Fisher, 2000). have contact with this group, it is conceivable that Because injection drug use in Romania is relatively these adolescents’ behaviors may inadvertently infect rare (Dehne, Grund, Khodakevich, & Kobyshcha, others, resulting in a spiraling increase of HIV infec- 1999) we expected to see more risk behaviors along tions. The scarcity of sex education and sexual health sexual than drug dimensions. A secondary and final promotion activities suggests that adolescents’ knowl- purpose of our research was to confirm how these edge of HIV will be low. Finally, little is known about individuals contracted HIV. Although it is believed interventions focused explicitly on HIV-positive ado- that these individuals all contracted HIV as pediatric lescents, even in Western contexts. A recent meta- cases during the Ceausescu regime, other means of ¸ analysis reviewed HIV prevention interventions for acquiring HIV are possible and such cases would adolescents but found no studies with HIV-positives provide valuable information about the state of the and few outside of the US (Johnson, Carey, Marsh, epidemic in Romania. Levin, & Scott-Sheldon, 2003). Similarly, a recent meta-analysis of HIV prevention for those who know they are HIV positive found no studies with adoles- Method cents and none outside the US (Johnson, Carey, Chaudoir, & Reid, 2006). Thus, it is not known Participants and Recruitment whether Western HIV prevention technology will transfer to HIV-positive Romanian adolescents. We recruited a sample of 153 HIV-positive adolescents To maximize the impact of future prevention efforts, (87 females and 66 males) aged between 13 and the first step is to learn the extent to which Romanian 22 years through cooperation with the Matei Bals ¸ HIV-positive adolescents exhibit risk deficits. Thus, the Institute in Bucharest (from October, 2003 until main goal of the current research is to document risk January 2004). This institute is the largest infectious behaviors and psychological risk deficits in a sample disease hospital in Romania, with ~1,400 HIV-positive drawn from this population. Accordingly, we examine adolescents registered (roughly 23% of the total in the adolescents’ knowledge of HIV and HIV transmis- Romania as a whole). According to the caregivers, all sion risk as well as the adolescents’ self-reports of participants were in medical care at this institution for sexual and drug-use behaviors that might put others at 10 or more years; thus, all patients have a medical and risk of acquiring HIV. Consistent with dominant psychological record in the hospital, describing their theories of psychological risk and in particular the medical and psycho-social history, but these histories theory of planned behavior (Ajzen, 1991), we examine were not available for the current study. Attending whether the adolescents’ attitudes, intentions, subjec- the hospital every 2 months for tests and specialized tive norms, and perceived behavioral control suggests medical investigations, the adolescents were recruited the potential for performing such risky behaviors. by a team of four practitioners (three psychologists and Within this framework, Attitude is the degree to which one social worker) from the two wings of the hospital one has a positive versus a negative evaluation of the that were most relevant to the population in question. behavior, subjective norm is the perception that impor- One of these wings is dedicated to serving disadvan- tant other people think that one should or should not taged HIV cases; the other has special clinics for perform the behavior in question, and perceived children with HIV, but this area also serves children behavioral control is the degree to which one perceives with other illnesses. The participants completed the 123
- 3. 718 AIDS Behav (2007) 11:716–725 questionnaires individually and privately, but could ask Measures for support from any of four practitioners who were on-site (e.g., to aid in understanding some of the The inventory inquired about a broad array of health questions or instructions). Every third patient on the and psychological dimensions; those not indicated daily list of appointments was recruited; the children’s below were merely for future study development and social workers or psychological case workers offered not directly relevant to the current study. All mea- each prospective participant a modest sum (the equiv- sures were screened through a back-translation tech- alent of $5 in local currency) to complete a question- nique sensitive to cultural differences and respectful naire concerning ‘‘health behaviors of HIV-positive of the participants’ potential past actions (e.g., the youth.’’ Nearly all (99%) of the youth informed about questions never implied that such actions should be the study agreed to participate. dismissed as unacceptable). A team member in Roma- nia first translated the English versions into Romanian; Procedure a native-Romanian speaker in the US provided a back-translation into English. For several items it was Similar to other Eastern European countries (Goodwin, necessary to repeat the process until all disagreements Kozlova, Nizharadze, & Polyakova, 2004), Romanian were resolved. society associates significant stigma with being infected with HIV in Romania and regards HIV-risk behaviors Mode of Transmission and Status on ART such as injection drug use and sexual interactions outside marriage as sinful. Approximately 90% of the Participants were asked ‘‘How do you think you population in Romania is Christian Orthodox, and got HIV?’’ (multiple responses were possible and family education places a great emphasis on religious included: having sex; sharing a needle; blood transfu- aspects, which increases the feeling of shame within the sion; had it at birth) and other details about their families of people who are infected with HIV.1 With this medication status. Specifically, they were asked if they pretext in mind, we designed an information sheet for are taking any sort of medication for HIV (Yes; No), the participants that emphasized not only how the for how long they had been taking medication for responses they provided could help to understand the HIV, and for how many months they might have been health of HIV-positive youth but also their freedom not taking any type of combination of different medica- to answer any item they wished not to answer; they were tions for HIV. All of these participants had current also assured of the confidentiality of their responses access to antiretroviral therapy (ART; see Kozinetz, (e.g., ‘‘all of your responses are kept strictly PRI- ˘ ¸ Matusa, & Cazacu, 2001). VATE’’). The information sheet and questionnaire explicitly requested that no names or other identifying Risk Deficit Survey information be provided. Participants then completed the paper-and-pencil inventory in a private room. After Participants were then asked to select a phrase that completing the inventory, they placed it in a sealed best describes their living situation [living on the street; envelope, mixed it with other such envelopes in a box, living in an abandoned building; living in a homeless were thanked and dismissed. shelter; living in a rehabilitation home or half-way The procedures in the study were approved both by house (for drug abuse); living in a assisted living the Institutional Review Board at the University of program for people with medical problems; living with Connecticut and its counterpart at the University of parents; living with family members (other than par- Bucharest. ents); living with friends; living in a house, apartment, or condominium that you rent or own]. A series of questions about risk behavior followed; participants were asked if they ‘‘shoot drugs (inject drugs with a 1 During the Sunday services in the Orthodox Church, priests needle) to get high,’’ have ever had a sexual relation- teach about this message, based both on the Bible and on the ship with someone, and if they are currently in a sexual writings of the Apostles. In recent years, the proportion of youth participating in the church has increased steadily. Although there relationship with someone. Because no individual is no research to document the implementation of the religious indicated having ever injected drugs, the remainder messages in the people’s behavior, in Romania there is indeed a of this paper will focus on sexual-risk, which was a far mentality that associates HIV with sin. It is important to note more frequent behavior. that the Church neither explicitly approves nor dismisses condom use, whereas government policies have consistently The questionnaire was modeled on that utilized by favored condom use. Kozal et al. (2004). Eleven items assessed knowledge 123
- 4. AIDS Behav (2007) 11:716–725 719 about HIV and HIV transmission with options for is it that THEY could get HIV from YOU?’’ (very ‘‘true,’’ ‘‘false,’’ and ‘‘don’t know’’ (a table presented likely, likely, neither likely nor unlikely, unlikely, very in the results lists each item). For those that answered unlikely). at least one knowledge item (N = 122), proportion of Those who indicated sexual experience were asked correct responses served as an index of knowledge with several more questions about the nature of their reasonable reliability (Cronbach’s a = 0.73); ‘‘don’t experiences, including whether they commonly get know’’ answers were treated as incorrect responses. high on drugs or alcohol when having sex, their feelings To assess attitudes about safe sex, participants were about using condoms, their number of sexual partners, asked about their feelings regarding ‘‘always using a whether their partners were steady or casual, the condom during sex’’ with six different types of sexual serostatus of their sex partners (including separate partners, in turn: HIV-positive STEADY partners, questions about each of the three serostatus types), and HIV-positive CASUAL partners, HIV-negative the frequency of sexual interactions with each serosta- STEADY partners, HIV-negative CASUAL partners, tus-type partner. The questions also probed about the STEADY partners whose HIV status is unknown, or consistency of using condoms. As mentioned, partici- CASUAL partners whose HIV status is unknown. pants could choose not to answer the questions they Responses to these items were averaged in a scale with found not applicable in their case. good reliability (Cronbach’s a = 0.80). A single addi- tional attitude item asked about the desirability of ‘‘taking a chance and occasionally having unprotected Results sex with a really hot (sexy) partner.’’ For each of the seven attitude items, the five Likert-type response Description of the Sample choices were very good, good, neither good nor bad, bad, and very bad. To assess subjective norm, three The sample (N = 153) consisted of 87 females and 66 items asked whether steady partner(s), casual part- males and ranged in age from 13 to 22 years (M = ner(s), and ‘‘most people who are important to me,’’ 15.94; SD = 2.45); they reported having been in school respectively, ‘‘think we should always use condoms for 7.76 years (SD = 2.86). Most of the respondents during sex’’ (very true, true, neither true nor false, were living with their parents (68%) or other family false, very false). These three items were highly members (20%); only 5% were living in a home that correlated (a = 0.85), so were averaged to form an they owned. They generally knew (53%) that they had index of subjective norm. HIV for 5 years or less; more surprising is that nearly To assess perceived control over condom use behav- 30% knew for a year or less, with only 8.45% indicating iors, participants were asked ‘‘how hard would it be for that they had it ‘‘as long as I can remember.’’ Neither you to always use condoms with’’ as a stem, completed age nor gender were correlated with length of time in succession by the six partner types listed for the knowing HIV status, rs(141) = -0.10 and –0.04, respec- attitude items (choices were very hard to do, hard to tively, ns. Most (85%) indicated that they take med- do, neither hard nor easy to do, easy to do, very easy to ication for HIV, although as we noted all have access do); the items were highly correlated (a = 0.93) so to the requisite medication regimen. Of the 129 who were averaged into a single measure of perceived responded, 32 (25%) indicating ever having had sex behavioral control. Five items assessing intentions to with someone, and of the 137 who responded, 14 engage in safe sex mirrored those for attitudes using (10.22%) indicated having a current sexual partner. the question stem ‘‘In the future, I will use condoms every time I have sex with,’’ completed in succession HIV Acquisition Trends by the first four partner types listed above. These Likert-type items also had five possible responses Partially confirming our expectations, most respon- (strongly agree, agree, neither agree nor disagree, dents indicated that they had acquired HIV through disagree, strongly disagree), and collectively they infected needles (56%) or blood transfusion (17%) in exhibited high reliability (a = 0.80), so they were their first years of life. Surprising were the numbers averaged. The attitudes, subjective norm, perceived indicating sexual (20%) or neonatal (5%) acquisition control, and intentions items mapped well onto those (one person indicated more than one transmission desired in tests of the theories of reasoned action and route, including sexual transmission, and 43 individuals ´ planned behavior (see Albarracın et al., 2001). Finally, did not make a response). Consistent with past participants were asked, ‘‘If you had unprotected sex research, individuals who indicated sexual acquis- with a person whose HIV status is unknown, how likely ition were older (M = 20.18, SD = 1.22) than those 123
- 5. 720 AIDS Behav (2007) 11:716–725 who indicated other transmission routes (M = 15.16, too few responses to examine the amount of unpro- SD = 1.35), F(1,107) = 253.26, p < 0.001, d = 3.77. No tected intercourse with casual partners in the 3-month one under age 17 indicated acquiring HIV sexually. period, Table 1 shows these data regarding steady partners. Intercourse was unprotected a significant Sexual Risk Behavior Occurrence number of times both overall and with HIV-positive partners. The pattern was not significant for HIV- As we noted in the Method, no respondent indicated negative partners, but the direction of the effect was having used injection drugs. Consequently, the rem- the same. On the average for a sexually active ainder of this paper focuses on sexual risk behavior. respondent, 9.31 of the 31.46 acts (30%) over Of the 32 respondents who reported any sexual 3 months were unprotected; with HIV-positive part- intercourse, 22 (69%) reported activity with a single ners, 11.33 of 28.08 acts (40%), and with HIV- steady partner in the last 3 months and six (19%) negative partners, 2.2 of 7.2 (31%). Only 6 (46%) of reported activity with a casual partner in the same the 13 individuals who reported overall frequencies period. Those with any sexual activity were older maintained perfect condom use over that period. (M = 19.06, SD = 3.04) than those with none Eight individuals responded to the same items with (M = 15.12, SD = 1.40), F(1,151) = 114.42, p < 0.001, regard to partners whose serostatus was unknown and d = 2.12; similarly, those who expressed sexual activity reported no sexual acts. were far more likely to have indicated that they had acquired HIV sexually, v2(1, N = 109) = 73.90, d = HIV Knowledge 3.40. The steady partners were reported more often to be HIV-positive (79%) rather than HIV-negative Table 2 displays the knowledge items and their (21%) or HIV-unknown (0%). Although there were response profiles. Of the 122 who responded to at Table 1 Protected and unprotected condom use with steady partners Dimension and/or item N reporting activity M total acts M acts with a condom t d Total acts 13 31.46 (14.13) 22.15 (13.77) 2.48* 0.64 Acts with an HIV-positive partner 12 28.08 (18.55) 16.75 (16.10) 2.47* 0.66 Acts with an HIV-negative partner 10 7.20 (12.90) 5.00 (9.72) 1.52 0.44 Values in parentheses are standard deviations. The effect size (d) is calculated as the standardized mean difference (Johnson, 1993) * p < 0.05 Table 2 Responses to the HIV-knowledge items, rank-ordered by percentage correct from worst to best Item Responses Valid Don’t Choose correct know item A complete cure for HIV will be found very soon 120 51 (43%) 5 (4%) Combination drug therapy is pretty much a cure for HIV 117 33 (28%) 5 (4%) Someone who is willing to have unprotected sex with me is probably already HIV positive 119 50 (42%) 34 (29%) If a person is hanging out in places like shooting galleries or sexual ‘‘cruising’’ areas, you can be 118 47 (40%) 37 (32%) pretty sure that they are HIV positive Withdrawing the penis before the man ‘‘cums’’ (ejaculates) makes sex safe (T) 118 42 (35%) 59 (50%) Oral sex CANNOT spread HIV (T) 116 28 (24%) 63 (54%) If an HIV positive person has sex with another HIV positive person, they don’t need to use 121 11 (12%) 64 (53%) condoms (T) If two HIV positive people have sex and don’t use a condom, they can give each other new and 120 29 (24%) 79 (66%) different kinds of HIV (T) Having sex just once won’t spread HIV (T) 120 22 (18%) 79 (66%) If an HIV positive person is taking his or her pills (meds) or is on combination therapy, he or she 118 21 (18%) 82 (69%) cannot give HIV to anybody (T) If two HIV positive people have sex and don’t use a condom, they can give each other sexually 119 19 (16%) 91 (76%) transmitted diseases other than HIV (T) Mean total score of valid items (Cronbach’s a = 0.73) 122 NA 44.56% (SD = 22.74%) Items were scored as false except for items 8 and 9. T = item implies something about transmission-related risk 123
- 6. AIDS Behav (2007) 11:716–725 721 Table 3 Attitude, subjective norm, perceived behavioral control, and intentions to use condoms Index or item Valid Reliability M SD Unsafea (%) Attitude Attitude toward using condoms (index) 79 0.80 1.62 0.78 1.27 Taking a chance and occasionally having unprotected 75 – 3.80 1.38 20.00 sex with a really hot (sexy) partner Subjective norm (index) 78 0.85 1.53 0.89 10.26 Perceived behavioral control (index) 139 0.93 3.61 1.15 24.46 Intentions 146 0.80 1.44 0.65 4.11 Each item was assessed on 5-point scales ranging from 1 to 5. Low numbers on attitude, subjective norm, and intention items indicate lower risk; high numbers on perceived behavioral control indicate less risk, more control over condom use. Reliability figures are internal consistency estimates based on Cronbach’s a a The percentage of the sample expressing a response on the side of the scale implying increased risk (i.e., high scores on attitudes, subjective norm, or intentions; low scores on perceived behavioral control) least one of the nine items,2 only one person scored norm, and positive intentions to use condoms in sexual 100% correct; most (52%) got fewer than four correct; encounters, either with main or with casual partners. five (4%) got none correct. Consistent with these Nonetheless, there were non-trivial numbers of indi- overall patterns, an examination of the individual items viduals who gave responses on the unsafe ends of the suggests that the respondents incorrectly believed that scales. In particular, as the right-most column in Table 3 combination drug therapy is a cure for HIV and that a shows, profiles of unsafe responding were especially complete cure would soon be found for HIV. likely on perceived behavioral control, followed by Responses were better but not perfect regarding items subjective norm, intentions, and attitudes. Although that were more about the transmission of HIV (e.g., general responding on the attitude scale indicated that sexually transmitted diseases are spread when strong support for condom use, the item, ‘‘taking a condoms are not used). Roughly one-third of the chance and occasionally having unprotected sex with a sample did not realize that unprotected sex between really hot (sexy) partner’’ suggested the potential for HIV-positives can result in a different kind of HIV risk behaviors, with nine (12%) indicating that it would infection. Analyses of the total knowledge scores be ‘‘very good’’ to do so, six (8%) saying ‘‘good,’’ and revealed one other significant pattern: older respon- seven (9%) saying ‘‘neither good nor bad.’’ dents scored better than younger respondents, r(121) = 0.30, p < 0.001. Length of time knowing their Route of Transmission HIV status, gender, and sexual experience did not significantly relate to HIV knowledge. Route of We then examined whether route of HIV transmission transmission was related to HIV knowledge, a result was related to (a) the risk-related judgments from the we report below in the sub-section on Route of preceding sub-section, (b) knowledge, and (c) the item Transmission. about using drugs or alcohol when having sex. Because the number of comparison cases varied quite widely Safer Sex Attitudes, Perceived Control, Subjective across these dimensions, an inspection of standardized Norms, and Intentions difference effect sizes is particularly informative. Spe- cifically, as the right-most column in Table 4 shows, Overall Patterns those with sexual transmission had higher risk on all of the risk-related dimensions except knowledge. Yet, Mean responses to the questions assessing attitudes, perhaps in part to lowered statistical power available perceived control, subjective norm, and intentions on some of the dimensions, only 2 of the individual toward safe sex were all near the positive end of the patterns emerged as significant: compared to those scale. As Table 3 reveals, the respondents had positive with other means of transmission, those who reported attitudes, perceived control, supportive subjective sexual transmission were (a) more knowledgeable about HIV and (b) thought that important others were 2 less supportive of using condoms (less supportive The portion of the sample that did not reply was significantly subjective norm). Although very few participants younger (M = 14.31 years) than those who did reply (M = 16.28), p < 0.001, suggesting that the younger respondents did not see answered the question about being high on drugs or the relevance of this section of the questionnaire. alcohol while having sex, the response profile is still 123
- 7. 722 AIDS Behav (2007) 11:716–725 Table 4 Risk-related dimensions broken into those who indicated sexual versus other routes of infection with HIV Risk-Related Dimension Sexual transmission Other transmission routes Comparison N M SD N M SD F p da Knowledge (index) 22 0.57 0.22 70 0.44 0.20 7.17 <0.01 +0.65 Attitude Attitude toward using condoms (index) 22 1.70 0.57 40 1.53 0.73 0.95 ns –0.26 Taking a chance and occasionally having unprotected sex 22 4.05 1.13 37 3.73 1.54 0.70 ns –0.22 with a really hot (sexy) partner Subjective norm (index) 22 2.27 0.99 40 1.20 0.55 29.74 <0.001 –1.43 Perceived behavioral control (index) 22 3.65 0.80 77 3.68 1.27 0.01 ns –0.02 Intentions 22 1.53 0.63 82 1.36 0.60 1.28 ns –0.27 How often do you get high on drugs or alcohol when 20 5.40 1.10 5 6.00 0.00 1.45 ns –0.58 you are having sex? See Table 2 for scoring of knowledge index; other items were assessed on 5-point scales ranging from 1 to 5 except that the drug-alcohol question ranged from 1 (all the time) to 6 (never). Low values on knowledge, attitude, subjective norm, and the intention items indicate lower risk; high numbers on perceived behavioral control and the drug-alcohol dimensions indicate more control over condom use a Standardized mean difference effect sizes (d) are negative for comparisons that show greater risk for the sexual transmission group compared with those who indicated other routes of transmission and positive for the reverse informative: None of those who acquired HIV non- planned behavior’s predictions, intention was signifi- sexually indicated ever using drugs or alcohol while cantly correlated with proportion of protected having sex; 7 of the 20 (35%) sexual-transmission cases intercourse during the last 3 months, r(11) = 0.49, who responded indicated at least some of these pone-tailed = 0.045, and perceived control related mar- practices. One case indicated doing so ‘‘most of the ginally, r(11) = 0.43, pone-tailed = 0.07. time’’; another ‘‘a good bit of the time’’; and five indicated doing so ‘‘a little of the time.’’ Discussion Other Regression Analyses It has been more than 15 years since the Ceausescu-era ¸ Further regression analyses simultaneously assessed pediatric AIDS cases garnered such intense media whether gender, age, or sexual experience, were attention. The current research strove to provide related to the attitude, subjective norm, perceived information about these young people, the survivors control, and intention dimensions; in fact, none of of whom are reaching sexual maturity. Although the these produced significant patterns when controlling strong majority of these young Romanians appear to for route of acquisition. Subsequently, consistent with have acquired their disease as pediatric AIDS cases the theory of planned behavior, the main risk item of during the Communist era, a substantial minority intentions to use condoms was regressed on the two instead appears to have acquired HIV relatively attitude items (the summed scale and the single item recently and via sexual transmission. Indeed, many of regarding ‘‘a really hot (sexy) partner’’) and the other these individuals (20% of our sample) who were two dimensions (subjective norm, perceived control thought to be long-term AIDS cases are instead dimensions); we used one-tailed t-tests given that the short-term cases who probably acquired HIV in their hypotheses were directional. This analysis revealed early to mid-teen years, given the latency required for that the attitude scale had the largest contribution AIDS symptoms to appear. The fact that so many such (b = 0.56, p < 0.001), followed by perceived control cases would appear in the wing of a hospital dedicated (b = –0.31, p < 0.001), and the single attitude item since the earliest days of the pediatric AIDS crisis in regarding unprotected intercourse with a hot partner Romania to providing service to such cases strongly (b = 0.17, p = 0.034). Subjective norm did not signifi- suggests that sexual transmission of HIV should be an cantly relate (b = 0.01, ns). Together, the predictors important concern. Although few if any of these young explained 50% of the variation in condom use inten- people seem to be injection drug users, they are to a tions for the 72 participants whose data were in this growing extent sexually active and reported significant analysis. Finally, in the portion of the sample with numbers of unprotected sexual interactions with part- recent sexual activity and consistent with the theory of ners known to be HIV-negative as well as with partners 123
- 8. AIDS Behav (2007) 11:716–725 723 known to be HIV-positive. Because inconsistent tected sex with a really hot (sexy) partner’’ suggested adherence to antiretroviral therapy is known to create that nearly one-third of the sample would take the new, drug-resistant strains of HIV, protected inter- chance (Table 3). Sample members who indicated course is important, even when the partner is also HIV acquiring HIV sexually were even more likely to show positive. Finally, our questionnaire illustrated that risk deficits than were others, although the compari- many of these HIV-positive adolescents have large sons did not always attain conventional levels of deficits in knowledge and motivation for safer sex; they statistical significance (Table 4). In particular, sexual- may also have serious deficits in skills for safer sex. The route individuals perceived that important oth- results of this research highlight the need for further ers—especially their sexual partners—did not strongly national-level research into the sexual behaviors of the support the use of condoms for every sexual encounter. HIV-positive adolescents. This pattern may well reflect continued sexual inter- The risk survey showed that knowledge of HIV- actions with the very people from whom the infections related risk is relatively low. Indeed, most participants derived, especially given that other responses indicated answered the majority of items wrong. By comparison, that most sexual interactions are with a main rather Ezeanolue, Wodi, Patel, Dieudonne, and Oleske’s than casual partner. Thus, HIV sero-sorting appears to (2006) recent study of 77 HIV-positive adolescents be a likely practice in this sample. and young adults living in New Jersey revealed better Other research has shown the importance of knowledge about transmission of HIV than did our ameliorating skills deficits in adolescents as a means current sample. For example, our sample incorrectly to HIV prevention. Johnson and colleagues’ (2003) believed that combination drug therapy is a cure for meta-analytic review of HIV prevention studies in HIV and that a complete cure would soon be found for adolescents found that condom skills training increases HIV (Table 2). Similarly, approximately one-third condom use substantially. The fact that such skills failed to realize that unprotected sex between HIV- training techniques had a big impact strongly suggests positives can result in a different kind of infection. that adolescents generally have a skills deficit, even in Although older participants and those who reported locales such as the US, where safer sex education acquiring HIV sexually scored better on knowledge, programs are much more common, albeit often contro- neither group scored very well. This result could be versial. Although the studies in that review were explained by cultural factors: There has been no conducted mainly in North America, it seems likely ‘‘sexual revolution’’ in Romania such as many Western that the strategy would also work in Romania or other countries experienced and sex-related issues remain a Eastern European countries (for related discussion, see taboo subject for discussions within the family (e.g., Goodwin et al., 2004; Hersh et al., 1993; Wright, 2005). Bencomo, 2006). As we documented, 68% of the Moreover, the possibility of iatrogenic effects of a safe- participants live with their families, and 20% of them sex intervention is remote. Indeed, Johnson and are living with one family member. Nonetheless, the colleagues’ meta-analysis concluded that the reviewed literature contains other examples showing how youn- interventions had reduced rather than increased the ger adolescents have lower HIV-relevant knowledge frequency of sexual interactions (e.g., improved absti- than older adolescents; similarly, it is common to find nence, decreased number of partners, or decreased that sexually active adolescents are more knowledge- number of acts). Smoak, Scott-Sheldon, Johnson, and able than their non-sexually active counterparts (e.g., Carey (2006) reached a similar conclusion in their more Bachanas et al., 2002; Silver & Bauman, 2006). general and larger meta-analysis of safer-sex studies. Nearly all of the participants expressed ‘‘safe’’ One caution that should be noted is the extent to which responses to the questions assessing attitudes, per- HIV/AIDS is stigmatized in Romania, which makes ceived control, subjective norm, and intentions toward being HIV-positive even more of a mental health risk safe sex. The respondents had positive attitudes, than it is normally (Bencomo, 2006; McNeil Jr, 2004). perceived control, supportive subjective norms, and Those who know they have HIV may fear disclosing positive intentions to use condoms in sexual encoun- their status to potential sexual partners out of fear of ters, either with main or with casual partners. Still, rejection. Such individuals may have a particular need some individuals gave responses on the opposite ends for even more skills and resilience training to negotiate of the scales that constitute these indexes. In particular, safe sexual interactions with others. it was common to see profiles of unsafe responding on Although we have been able to form conclusions perceived behavioral control, subjective norms, inten- from the current investigation, it was limited somewhat tions, and attitudes. More pointedly still, answers to the by the fact that so many participants did not answer the item, ‘‘taking a chance and occasionally having unpro- key questions about HIV risk behaviors and related 123
- 9. 724 AIDS Behav (2007) 11:716–725 psychological correlates. Although the HIV transmis- of these young people to their medication regimes; sion and demographic data were available almost only 85% indicated they were taking medications for without exception, unfortunately risk behavior survey HIV/AIDS, whereas all had complete access to anti- information had many missing responses. Our proce- retroviral therapy. To the extent that these individuals dures offered the respondents privacy and reassurances lack adherence, they will of course perpetuate drug- about the confidentiality and anonymity of their resistant strains of HIV, and to the extent that they responses. Because those who completed sexual risk spread these strains, the HIV pandemic will worsen. items tended to be older and to have acquired HIV The current research appears to be the first research to sexually, it is possible that the younger, sexually investigate the risk profile of HIV-positive young inactive participants simply did not see the questions people in Eastern Europe. It is possible that this as relevant to their situation and therefore did not investigation can serve as a model for new studies in answer them; moreover, many of the instructions similar areas in the world. It is our hope that the explicitly gave them the option not to answer. One current study can provide a solid starting point for alternative would be to use electronic means of future research on which to build not only new collecting data from this population. Hand-held com- epidemiological work but also new interventions to puters have proven to be a very reliable means of slow the spread of HIV within Romania and elsewhere. gathering data, and one for which missing responses are not a problem; moreover, these devices have been Acknowledgments We thank Blanche Serban and Simona Sava ¸ for translating and back-translating the materials used in this study used in disadvantaged and stigmatized groups (e.g., and Cristian Anghelina, William D. Barta, Adrian Bidulescu, young drug users, Turner et al., 1998; see also van ˘ ˘ Marcella B. Boynton, Odette Chirila, Carina Jalba, Adela Manea, Griensven et al., 2006). The fact that the devices ˘ Livius Manea, Ioan Marginean, Nicolae Mitrofan, and Iulia reliably take the response from the user and quickly Oprea, who assisted in preparing the questionnaires or collecting, coding, or organizing the study data, and three proceed to the next question means that the respon- anonymous reviewers who provided helpful feedback on prior dent’s private answers are unlikely to be seen by others drafts of this article. This research was supported by University of and therefore encourages honest responding. Connecticut grants to Blair T. Johnson from the Research Due to oversight, our questionnaire did not include Foundation (FRS444150) and the Center for Health, Intervention, and Prevention (2003-01); the preparation of this an item asking directly whether the participants were report was facilitated by U. S. Public Health Service Grant R01- pediatric AIDS cases. The questionnaire did not MH58563 to Blair T. Johnson. include an explicit question on this matter because in Romania it has been presumed that this group of young people was infected during their early childhood References as a result of multiple hospital admissions, through unsafe medical practices, or the lack of appropriate Ajzen, I. (1991). The theory of planned behavior. Organizational medical tools such as we noted in the Introduction. Behavior and Human Decision Processes, 50, 179–211. ´ Albarracın, D., Johnson, B. T., Fishbein, M., & Muellerleile, P. Yet, an official epidemiological investigation was never A. (2001). Theories of reasoned action and planned behav- undertaken to confirm this assumption. All participants ior as models of condom use: A meta-analysis. Psychological in our study have a multiple-year medical record to Bulletin, 127, 142–161. track their health and treatment progression. Although Bachanas, P. J., Morris, M. K., Lewis-Gess, J. K., Sarett-Cuasay, E. J., Sirl, K., Ries, J. K., & Sawyer, M. K. (2002). Predictors the problem is easy to see in retrospect, even the of risky sexual behavior in African American adolescent regular care providers informed our research team that girls: Implications for prevention interventions. Journal of they expected that their patients had acquired HIV Pediatric Psychology, 27, 519–530. as a pediatric AIDS cases during the Ceausescu era. ¸ Bencomo, C. (2006). ‘‘Life Doesn’t Wait’’: Romania’s failure to protect and support children and youth living with HIV. Nonetheless, the fact that so many (20%) respondents Human Rights Watch, 18(6), 1–105. indicated sexual means of acquiring HIV is difficult to ˘ ˘ Chisevescu, D. P., Mihailescu, I., Mihailescu, G. P., & Pasat, L. explain away and should pique interest: It is time to ˆ (1998). Injection practices among nurses—Valcea, Romania, question how AIDS cases are emerging in Romania. It 1998. Morbidity and Mortality Weekly Report, 50(4), 59–61. Danziger, R. (1996). An overview of HIV preventioin in central appears that sexual HIV infection in Romania is a and eastern Europe. AIDS Care, 8, 701–707. bigger problem than previously recognized, mirroring Dehne, K. L., Grund, J. C., Khodakevich, L., & Kobyshcha, Y. so many other Eastern European countries. Further (1999). The HIV/AIDS epidemic among drug injectors in epidemiological research focused on means of trans- Eastern Europe: Patterns, trends and determinants. Journal of Drug Issues, 29, 729–776. mission should be conducted to determine whether EuroHIV (2005). HIV/AIDS Surveillance in Europe: End-year these patterns are limited only to young people. It report 2004, No. 71. Saint-Maurice: Institut de Veille would also be valuable to study the adherence patterns Sanitaire. 123
- 10. AIDS Behav (2007) 11:716–725 725 Ezeanolue, E. E., Wodi, A. P., Patel, R., Dieudonne, A., & ˘ ¸ Kozinetz, C. A., Matusa, R., & Cazacu, A. (2001). The burden of Oleske, J. M. (2006). Sexual behaviors and procreational pediatric HIV/AIDS in Constanta, Romania: A cross-sec- intentions of adolescents and young adults with perinatally tional study. BMC Infectious Diseases, 1:7. Retrieved from acquired human immunodeficiency virus infection: Experi- http://www.biomedcentral.com/1471–2334/1/7. Assessed on ence of an urban tertiary center. Journal of Adolescent April 16, 2002. Health, 38, 719–725. McNeil, D. G., Jr. (2004, February 11). Romania declares victory Fisher, J. D., & Fisher, W. A. (2000). Theoretical approaches to in fight against AIDS. The New York Times, p. A1. individual-level change in HIV risk behavior. In J. L. Silver, E. J., & Bauman, L. J. (2006). The association of sexual Peterson, & R. J. DiClemente (Eds.), Handbook of HIV experience with attitudes, beliefs, and risk behaviors of prevention (pp. 3–55). Dordrecht, Netherlands: Kluwer inner-city adolescents. Journal of Research on Adolescence, Academic Publishers. 16, 29–45. Goodwin, R., Kozlova, A., Nizharadze, G., & Polyakova, G. Simon, C., Evans, K., Calles, N. R., Harris, A. T., Dobmeier, M., (2004). HIV/AIDS among adolescents in Eastern Europe: Moreau, L., et al (1999). A program of acquired immune Knowledge of HIV/AIDS, social representations of risk and deficiency syndrome education for nurses in Romania. sexual activity among school children and homeless adoles- Journal of Pediatric Oncology Nursing, 16, 68–73. cents in Russia, Georgia and the Ukraine. Journal of Health Smoak, N. D., Scott-Sheldon, L. A. J., Johnson, B. T., & Carey, Psychology, 9, 381–396. M. P. (2006). Sexual risk reduction interventions do not Hersh, B. S., Popovici, F., Jezek, Z., Satten, G.A., Apetrei, R.C., inadvertently increase the overall frequency of sexual Beldescu, N., et al (1993). Risk factors for HIV infection behavior: A meta-analysis of 174 studies with 116,735 among abandoned Romanian children. AIDS, 7, 1617–1624. participants. Journal of Acquired Immune Deficiency Johnson, B. T. (1993). DSTAT 1.10: Software for the Syndromes, 41, 374–384. meta-analytic review of research literatures. Mahwah, NJ: Turner, C. F., Ku, L., Rogers, S. M., Lindberg, L. D., Pleck, J. H., Erlbaum. & Sonenstein, F. L. (1998). Adolescent sexual behavior, Johnson, B. T., Carey, M. P., Chaudoir, S., & Reid, A. E. (2006). drug use, and violence: Increased reporting with computer Sexual risk reduction for persons living with HIV: Research survey technology. Science, 280, 867–873. synthesis of randomized controlled trials, 1993 to 2004. van Griensven, F., Naorat, S., Kilmarx, P. H., Jeeyapant, S., Journal of Acquired Immune Deficiency Syndromes, 41, Manopaiboon, C., Chaikummao, S., Jenkins, R. A., Uthaiv- 642–650. orat, W., Wasinrapee, P., Mock, P. A., & Tappero, J. W. Johnson, B. T., Carey, M. P., Marsh, K. L., Levin, K. D., & Scott- (2006). Palmtop-assisted self-interviewing for the collection Sheldon, L. A. J. (2003). Interventions to reduce sexual risk of sensitive behavioral data: Randomized trial with drug use for the Human Immunodeficiency Virus in adolescents, urine testing. American Journal of Epidemiology, 163, 1985–2000: A research synthesis. Archives of Pediatrics and 271–278. Adolescent Medicine, 157, 381–388. Wright, M. T. (2005). Homosexuality and HIV/AIDS preven- Kozal, M. J., Amico, K. R., Chiarella, J., Schreibman, T., tion: The challenge of transferring lessons learned from Cornman, D., Fisher, W. D., Fisher, J. D., & Friedland, G. Western Europe to Central and Eastern European coun- (2004). Antiretroviral resistance and high-risk transmission tries. Health Promotion International, 20, 91–98. behavior among HIV-positive patients in clinical care. AIDS, 18, 2185–2189. 123