Introduction

AIDS is one of the major public health challenges across the world (1). According to a report by the Joint United Nations Program on HIV-AIDS (UNAIDS), approximately 34 million people are infected with HIV worldwide (2). Studies indicated that almost five million new cases of HIV infection occur annually (3) and based on the current trend, HIV/AIDS is estimated to become the third leading cause of mortality in low- and middle-income countries by 2030 (4). Unlike most regions of the world, HIV epidemic in the North African and Middle Eastern countries such as Iran is on a growing trend (5).

Based on the reports of medical universities and health services until March 21, 2015, 29414 HIV patients are living in Iran, 88% of whom are male and 12% female (6). In the recent years, there has been some evidence regarding the growing role of sexual transmission in the spread of HIV in Iran, and the rate of recorded cases attributed to sexual transmission is steadily growing (7); moreover, the prevalence of HIV among female sex workers is much higher than the rest of female population (3).

According to the data obtained from 14 Asian countries, HIV prevalence in females aged 15-49 years was 0.18%, while it was 5.2% in sex workers, indicating that the risk of having HIV is 29 times greater in this population. The data demonstrated that female sex workers are at particularly high risk of infection (8).

In a study by Mirzazadeh performed in 14 cities of Iran, HIV prevalence in female sex workers was estimated to be 4.5% (5). Due to prevalence of HIV in this population, AIDS 2012 Conference highlighted the need for particular attention to planning interventions for female sex workers (9).

Centers for disease control and prevention interventions such as HIV testing and counseling were established to enhance perception of HIV risk (10).

Risk perception is a complex multi-factorial concept built on the experiences a person undergoes during the course of life, which are both directly or indirectly influenced by socioeconomic, political, and cultural factors (2). Behavioral models suggest that behavioral change is associated with perception of the risk of disease infection and its serious consequences (11). Thus, for individuals who exhibit high-risk behaviors, perceiving susceptibility is necessary for changing high-risk behaviors; otherwise, explaining benefits of safe sexual behavior would be ineffective (12).

Perceived risk of HIV infection can be considered as a strong predictor of low level of risk behaviors associated with HIV/AIDS (13), and low HIV risk perception is related to high-risk sexual behaviors (14, 15). Although risk perception is not sufficient for behavioral change, it is the first step toward modifying unhealthy behaviors to healthy ones (11, 16, 17). Darbes (2008) performed a review of behavioral interventions for diminishing HIV risk behaviors, 15 out of 38 studies accentuated the importance of examining or enhancing perceived risk and emphasized that development of AIDS risk perception can be considered as a preventive strategy (10)(18).

Thato (2003) found that men with low AIDS risk perception were more likely to commit high-risk sexual behaviors (19). Luna stated that vulnerable and marginalized groups do not have enough control over their lives, and even feel the need to numerous things for survival, which may lead to low HIV risk perception (20).

Given the reports regarding the need for evaluating awareness of women with high-risk behaviors and scarcity of studies on this issue in Iran (14,15), we aimed to evaluate AIDS risk perception and its related factors in females with high-risk behaviors in Mashhad.

Materials and Methods

This descriptive study was performed in 58 women, who were detained on the charge of prostitution in the Central Prison of Mashhad, Iran, 2013. The inclusion criteria were as follows: having Iranian nationality, giving written consent to participate in the study, having basic reading and writing skills, having no psychological disorders (e.g., bipolar disorder, attention deficit, and depression) according to their health records. The exclusion criteria included unwillingness to continue participating in the study.

The data collection instrument  was demographic and sexual behaviors information questionnaire consists
of three parts as follows: part I: demographic information including eight questions (age, educational level, income level, marital status, history of  opioid addiction, alcohol consumption, and escape from home, and age at first drug use); part II: information regarding sexual behaviors and AIDS, which included eight questions (age at the first voluntary sexual intercourse, the number of intercourses to earn income during a week, contraceptive method, the rate of condom use during intercourse, history of abortion, history of rape, as well as voluntary HIV testing and follow-up); and part III: HIV/AIDS knowledge was assessed in this part using six questions; one point was given to correct answers, while wrong answers and “I did not know” scored no points. The minimum and maximum possible scores in this section were zero and six, respectively.

A self-designed questionnaire was used to measure AIDS risk perception, which consists of eight items including individual feeling about likelihood of getting infected with HIV (ranging from extremely unlikely to extremely likely), concern about getting infected with HIV (ranging from not concerned to extremely concerned), thinking about getting infected with HIV (ranging from never thought about to often think about), individual fear and shame of getting infected with HIV (ranging from very low to very high), unlikely to get infected with HIV (ranging from strongly disagree to strongly agree), chance of getting infected with HIV (ranging from zero to very high), and self-image of getting HIV (ranging from very easy to very hard). The respondents rated the items using a 6-point Likert scale, and the minimum and maximum possible scores were 0 and 40, respectively. Scores ranging from 0 to 13.33 indicated poor risk perception, 13.34 to 26.67 moderate, and 26.68 to 40 appropriate, with higher scores indicating greater HIV risk perception.

The content validities of the AIDS awareness and HIV risk perception questionnaires were confirmed by 10 faculty members of Mashhad and Tehran universities of medical sciences and reproductive health PhD candidates as a group of subject matter experts. Afterwards, they were asked to rate the questionnaires for their relevance to the study objectives using a 4-point Likert type scale, ranging from perfectly appropriate to completely inappropriate.

The content validities of HIV/AIDS knowledge and HIV risk perception were 0.93 and 0.84, respectively. The reliabilities of HIV/AIDS knowledge and HIV risk perception questionnaires were established using the test-retest method, (r=0.83 and r=0.72, respectively). After obtaining approval of the Ethics Committee of Mashhad University of Medical Sciences and being introduced to authorities of Vakil Abad Prison of Mashhad, Iran, the researcher attended the prison during office hours and after explaining the importance, objectives, and how to apply the results to the eligible women, their oral and written consents were obtained. Thereafter, the questionnaires were given to the samples and they were assured of the confidentiality of the personal information.

Data were collected through face-to-face method in the library, which lasted approximately 10-15 minutes. Thereafter, Kolmogorov-Smirnov and Shapiro-Wilk normality tests, Pearson and Spearman coefficient tests, linear regression, One-way ANOVA, independent samples t-test, and Chi-square tests were run, using SPSS version 16.

Results

The results of this study showed that the mean age of the participants was 28.4±5.4 years (age range: 20-44 years). With respect to educational level, 39.7% (n=23) had high school diploma and above, and based on subjective measures of poverty, 44.28% (n=26) had less than sufficient income. Moreover, 74.13% (n=43) were divorcees, and age at the first voluntary sexual intercourse was 15.5±3.1 years, the mean age at the first drug abuse was 17.09±5.5 years, and the mean scores of HIV/AIDS knowledge and HIV risk perception were 3.79±1.56 and 18.43±5.92, respectively. The Pearson correlation test results showed a significant relationship between the level of HIV/AIDS knowledge and HIV risk perception (r=0.82, P=0.005).

Based on the results of independent t-test, there was a statistically significant difference between the mean score of HIV risk perception and history of opioid addiction (P=0.008), alcohol consumption (P=0.04), voluntary testing for HIV (P=0.001), and HIV testing follow-up actions (P=0.009). Thus, participants without history of opioid addiction, alcohol consumption, as well as voluntary testing for HIV and follow-up actions had greater HIV risk perception.

There was a significant relationship between HIV risk perception and contraceptive methods (P=0.01) and the rate of condom use during intercourse (P=0.02). However, Pearson correlation coefficient reflected no significant correlation between age and HIV risk perception (r=0.15, P= 0.73). In addition, according to Spearman correlation coefficient analysis, it was found that age at the first voluntary sexual intercourse was not significantly related to HIV risk perception (r=0.11, P=0.97).

Additionally, the results showed that the overall mean score of risk perception in married participants was higher than single (22.7±8.3 vs. 21.9±11.4) and divorced ones (21.3±9.2), but there was not a statistically significant difference between the mean score for HIV risk perception and marital status (P=0.46).

There was not a statistically significant difference between history of escape (P=0.73), abortion (P=0.68), and rape (P=0.43) as well as number of intercourses to earn income during a week (P=0/53) with mean score of HIV risk perception.

HIV risk perception was associated with HIV/AIDS knowledge, history of opioid addiction, alcohol consumption, contraceptive method, the rate of condom use during intercourse, as well as voluntary HIV testing and HIV follow-up actions. To control for the aforementioned variables, they were entered into general linear regression model. At the first step, the results of regression analysis showed that history of opioid addiction (P=0.04) and the rate of condom use during intercourse (P<0/001) were significantly able to predict HIV risk perception (R² =0.659).

At the second step, considering the results of regression analysis,  the variables of  opioid addiction history and the rate of condom use during intercourse were found to be predictors of HIV risk perception (R²=0.574). At the third step, which only included the rate of condom use during intercourse, it was observed that this variable was a predictor of the HIV risk perception (R²=0.508).

Review of absolute estimate of effect of the variables showed that the rate of condom use during intercourse was the strongest predictor of HIV risk perception, that is, one unit increase in the rate of condom use during intercourse was associated with a 0.713 unit increase in the HIV risk perception (Table 2).

Discussion

The results of this study demonstrated that despite having high-risk sexual behaviors and unprotected intercourse, the mean score of HIV risk perception was moderate. This population did not have appropriate HIV risk perception, which was similar to findings of studies conducted by Ford (2000) in Indonesia and Tran (2005) in Vietnam (21, 22).

Qualitative studies conducted in high-risk groups such as female prostitutes and drug users in Iran, demonstrated poor protective behaviors in sexual relations and suggested that these groups perceived themselves to be at no risk for AIDS. Therefore, to increase knowledge and improve perceived HIV risk, establishing AIDS Information Programs and training interventions seems to be mandatory (20, 23).

In the present study, among the individual characteristics variables, the only significant difference was observed between HIV risk perception and history of drug and alcohol consumption. In this study, more than half of the participants had history of drug and alcohol consumption. Li (2010) conducted a review on alcohol consumption in female sex workers during 1980-2008, it was concluded that there was an association between alcohol consumption in women and their physical and mental health, violence experience, sexual abuse, and drug exposure (24).

 Center for Disease Control and Prevention stated that there was a strong association between sex work and drug and alcohol use. Drug and alcohol use were reported as coping mechanisms in response to stressful conditions that female sex workers have in their life (25). Drug abuse was both directly and indirectlyrelated to transmission of HIV across the world (26), as the majority of drug users rarely used condom during legitimate and illicit sexual relations (27). In a study by Shiferaw (2014) conducted in Ethiopia to evaluate perception of HIV risk and factors associated with risk perception among university students, it was concluded that early onset  drug and alcohol use was significantly correlated with perception of HIV risk (28).

According to former studies, use of alcohol and drugs such as marijuana is an important risk factor for HIV infection, that is, AIDS is highly prevalent among drug abusers (29). In the current study, there was not a significant relationship between the factors associated with sexual activity characteristics and perception of HIV risk.

However, in a study by Nunn (2011), conducted in a predominantly Afro-American population, it was concluded that individuals with more than five sexual partners were more likely to perceive their HIV risk as zero or low (30). A study by Kalichman (2005) showed that individuals with several sexual partners and unprotected sexual relations during the past three months had significantly lower perception of HIV risk (31). This difference could be related to the larger sample size and classification method based on the number of sexual partners as zero, less than five, and more than five, which caused a significant difference.

The other results of the current study showed that there was a significant association between contraceptive method as well as the use of condom and HIV risk perception, that is, condom users had higher level of HIV risk perception. Several studies reflected that high HIV risk perception was the most powerful predictor of condom use (32).

This finding was in line with the results of Khaniya study (2004), which found that female sex workers with high HIV risk perception were more likely to use condom in their sexual relations (33). This finding is comparable to similar studies (34, 35); the results of regression analysis showed that the rate of condom use during intercourse was the most powerful predictor of HIV risk perception.

In the current study, a significant association was observed between voluntary HIV testing and HIV risk perception. This finding was in line with the results of the Tran (2013) and Dandona (2005) studies (13, 36).

Voluntary HIV testing is the most important measurement tool that can be used for controlling disease epidemic in different countries. Klein suggested that high level of vulnerability could increase an individual’s chance of testing for HIV (37). In the present study, 60.3% (n=35) of the cases voluntarily tested for HIV, while in studies by Hong (2012) in China , Tran (2013) in Vietnam, and Dandona (2005) in India, 48%, 32.7%, and 8% of female sex workers were voluntarily tested for HIV, respectively (13, 36, 38). This difference could be attributed to the imprisonment of the samples, sexually transmitted disease testing advertisement, which was held in the Triangular Clinic of the prison, and low-cost testing. The current study showed that there was a statistically significant correlation between HIV testing follow-up actions and perceived HIV risk, but this correlation was not observed  in the study by Tran (2013) (13).

HIV testing follow-up actions can help non-infected individuals to have low-risk behaviors and aid HIV positive patients to prevent transmission of the virus to others. 

The results of the present study indicated that HIV/AIDS knowledge score was higher than average. In a study by Qaderi (2011), it was found that female addicts had relatively high knowledge about AIDS and its transmission (39); however, in a study by Valadez (2013) in Libya, female sex workers had poor knowledge about HIV (40). Also, a study by Vian (2012) indicated that female sex workers had low knowledge about prevention and transmission of sexually transmitted diseases than other women did, but this difference was not significant (41). In the present study, there was a significant relationship between HIV/AIDS knowledge and HIV risk perception.

Knowledge is a construct in the model of information, motivation, and behavioral skills to reduce HIV risk behaviors that was proposed by Fisher in 1998. Information about prevention methods and transmission is the primary determinant of sexually transmitted disease/HIV prevention behaviors (42). Studies performed in America showed that education was more effective on health than job and income, and inadequate knowledge could lead to unfavorable attitudes toward health (43).

Studies performed in various parts of Asia indicated that there was an association between high-risk behaviors, inadequate knowledge, and low perception of HIV risk in female sex workers. Thus, training could improve their perception of HIV risk, reduce high-risk behaviors, and result in higher rate of condom use (13).

However, some studies demonstrated that knowledge does not necessarily lead to behavioral change and promotion of risk perception of sexually transmitted diseases. A study by Sutton (2011), carried out in Afro-American students, showed that the majority of students had an average to high HIV knowledge and attended programs related to HIV testing, and perceived themselves to be at low risk for HIV infection. Nonetheless, more than 50% of the students reported two or more partners in the previous 12 months, with inconsistent condom use; this finding suggests a poor relationship between HIV knowledge and increase in HIV risk perception and prevention of high-risk sexual behaviors (44).

The limitations of our study include the total sample size smaller than similar studies and performing the study only on detained female sex workers; thus, the results may be different from those of female sex workers in public. The participant’s emotional and psychological characteristics could affect their perception of risk and the researcher was not able to control this factor. All the confounding variables such as the status of HIV infection in the participants, the availability of condom, and its use method were not measured; hence, our study could not accurately assess all the factors responsible for HIV risk perception. In the current study, the rate of condom use during intercourse was evaluated by a self-designed questionnaire that might influence the results due to personal bias and interests; moreover, the main strength of this study was novelty in research conduct.

 Conclusion

Our findings indicated that there was a significant difference between the variables related to individual characteristics (e.g., knowledge, drug abuse, as well as alcohol consumption) and sexual behaviors (e.g., contraceptive methods, the rate of condom use during intercourse, voluntary HIV testing, and HIV testing follow-up actions) and HIV risk perception.

The rate of condom use during sexual intercourse was an effective predictor of HIV risk perception. Our results accentuate the need for training and prevention programs as well as basic counseling in prisons and for high-risk groups. Therefore, we should take appropriate measures for health promotion by developing appropriate training and interventional approaches to reduce high-risk behaviors in populations without sufficient risk perception. It is recommended to conduct further studies on perceived HIV risk in other populations in different regions of the country, while taking further variables into account

 Acknowledgements

This study was extracted from a master’s thesis approved by Deputy of Research and Technology of Mashhad University of Medical Sciences (MUMS) (approval code: 911145). The authors would like to thank the authorities of MUMS and School of Nursing and Midwifery, Ethics Committee of MUMS, head and staff of the gynecology ward of the Central Prison of Mashhad. We are also grateful to the participants who kindly volunteered to participate in this study.

 Conflicts of Interest

The authors declare no conflicts of inter