Correlates of HIV Risky Sexual Behaviours in an Era of Antiretroviral Therapy Scale ‐ Up : A Cross-Sectional Study among the Adult General Population in Nasarawa State , Nigeria

Corresponding Author: Dr Ishaku Ara Bako, Department of Epidemiology and Community Health College of Health Sciences Benue State University, Makurdi, Nigeria. Email: bakoiara@yahoo.com Analyses of HIV incidence in Nasarawa State Nigeria estimate that most of the new HIV infections occur among persons who reported low HIV risk including couples. The study was aimed at identifying the factors that predict risky sexual behavior among the general population in Nasarawa state, Nigeria. Data analysis was carried on a total of 801 respondents sampled from the general population of Nasarawa State, Nigeria. The original sample was obtained through a two-stage cluster sampling technique using probability proportionate to size. The primary outcome variable was risky sexual behavior. Chi-square and logistic regression analyses were used to determine the association between the outcome and selected Sociodemographic and other independent variables. Females were 54.2% of the total sampled population analysed, the mean age of the respondents was 29.8 years (SD: 10.3). About two-third of the respondents engaged in risky sexual behaviours (65.9%) but only 4.7% considered themselves to be at high risk of HIV. The multivariable regression analysis showed that factors associated with risky sexual behaviour included : been male sex [OR: 0.63; 95% CI: 0.436-0.915], married [OR: 0.26: 95% CI: 0.163 0.419], rural resident [OR: 1.20; 95% CI: 0.775 to 1.871 ], age 20-24 [OR: 1.93, 95% CI: 1.113 3.360] and 25-29 years [OR: 2.34; 95% CI: 1.267-1.308]; and knowledge of HIV [OR: 1.49; 95% CI: 1.056-2.108].There is a need to urgently intensify media campaigns, community-based interventions including one on one communications to reduce risky sexual behaviours.


INTRODUCTION
ersons who engage in risky sexual behaviours are at risk of contracting HIV/AIDS and other sexually transmitted infections.The huge difference in the HIV prevalence between sub-Saharan Africa and developed countries has been partly attributed to differences in the prevalence of risky sexual behaviors, especially if such are done outside of stable/regular relationships. 1 A study in Uganda showed that prevalence of HIV (33.4% among men and women 50.2%) was significantly associated with not using a condom during sex and having two or more sexual partners. 2Sexual behaviors of a people have been attributed, to some extent, to the moral and cultural values within the community.Accordingly, in communities where sexual permissiveness is less tolerated, risky sexual behaviors are less prevalent. 3With increasing globalization and socioeconomic modernization, however, there is an increasing tendency for individuals to become more oriented towards autonomous decisions and hence more likelihood to break these societal and cultural inhibitions. 4However, cultural inhibitions could also prevent people from using preventive measures such as the inability of the female sexual partners to initiate condom use during sex. 5 Reduction of risky sexual behavior' is a target of many HIV and STI interventions.However, HIV and STI epidemiology varies widely.Several studies have been conducted among adolescent and young persons' sexual behaviours, but not much has been done on risk taking behaviours at the general population level.Studies across Africa showed that sexual behaviour vary with different populations..It has therefore been suggested that the epidemiology of HIV and other sexually transmitted infections vary, necessitating the need for local evidences to guide prevention interventions. 9,10t is imperative that we understand the predictors of risky sexual behavior among the general population in Nasarawa state.Understanding this would enable both policy makers and HIV prevention programme implementers to prioritize their interventions.This study was therefore aimed at determining the prevalence and factors that predict risky sexual behavior which consists of an aggregate of early sexual debut, transactional sex and multiple sexual partnering among the general population in Nasarawa state, Nigeria.

Study Design
This cross-sectional analytical study was conducted in 2011 among the general population in Nasarawa State Nigeria.

Study Population
Persons eligible to participate in the study were females aged 15 -49 years and males aged 15-64 years and living in Nasarawa State Nigeria.A sample size of 840 was determined using the appropriate formula and using Contraceptive Prevalence rate of 9.7% obtained from the National Demographic Health Survey Nigeria, 2008 and 2.5% level of precision.Probability sampling techniques was used through a two stage sampling.In stage one, 30 Enumeration Team Areas (ETA) were selected using the cumulative measure of size of the Enumeration area in the state.The ETA were listed with their cumulative population sizes and selection was done using random numbers.In stage two, eligible individuals were randomly selected from a list within each cluster.The sample allocation to urban and rural areas was proportional to the rural/urban distribution of the population.A locality with a population of less than 20,000 was classified as a rural area.In this stage, the buildings and households within each selected ETA were identified and members of the households were listed with their names, ages and sexes.All eligible persons were identified and numbered serially.Systematic random sampling procedure was used to select 28 persons (14 males and 14 females) from each cluster.

Study Procedures
Each of the selected respondents was approached and given explanations about the study and then signed an informed consent form.A structured questionnaire was administered to selected individuals.Information was obtained on the background information of respondents, sexual behavior, knowledge and perception of HIV/AIDS, substance use, accessibility and use and access to HIV Counseling and Testing by trained interviewers.Risky sexual behavior was determined by finding the average score for the following (see table 2.1): early sexual initiation (Sexual debut before 15 years), transactional sex in past 12 months and number of sex partners in the past 12 months.Table 2.1 summarized the scoring for each of the variables.The total score for each respondent were added up to get the total score.Total score of 3 and lower were labeled as having low risk while any respondent with a total score of 4 and above was considered to have a high risk sexual behaviour.The exposure variables include the following: sex, age, location, educational attainment, marital status, self-perception of risk of HIV transmission, alcohol and substance use and comprehensive knowledge of HIV.

Data Analysis
Data analysis was done using the STATA package.The level of significance for the analysis was 5% and analysis was done using chi square, crude logistic regression and multivariable regression.

Ethical Consideration
Permission to carry out secondary data analysis was obtained from Society for Family Health (SFH) Nigeria while ethical approval for the original data collection was earlier obtained from the National Health Research Ethics committee of Nigeria (NHREC).The data do not have identifiers, so confidentiality of population that was used for the original data is maintained.

Socio-Demographic Characteristics
Table 3.1 shows the respondents' socio-demographic characteristics.Majority of them were females (54.2%) and aged 25-64 years) 65.8%).The mean age was 29.83 (± 10.3) years, the youngest respondent was 15 years while the oldest was 64 years.The majority of the respondents were ever married (70.7 %), while (29.3%) were never married.Rural residents form the majority of the respondents (83%).Slightly more than a third (35%) never attended school and only 11.2 % had tertiary level education.(Table 3.1)

Sexual behaviours and other selected characteristics
Majority of the respondents, 489 (61.1%) had their first sex at age 18 years or older while 188 (23.5%) had their sexual debut at 15 years or younger.Only 51 (6.4%) had transactional sex in the previous 12 months and 189 (23.6%) had two or more sexual partners in the previous year.Majority had no comprehensive knowledge of HIV (57.8%), ever done HIV test (55.7 %) and had a low self-risk perception of HIV (85.1%).Regular alcohol use and substance use were reported by 9.7% and 5.9% of the respondents respectively.(Table 3.2)

DISCUSSION
2][13][14][15][16][17] .However most of the previous studies were concentrated on young persons.High level of risky sexual behaviour among the For Reprint Contact: jbrcp.net@gmail.comgeneral population is of great public health concern because of its effect on transmission of HIV and other sexually transmitted infections and unwanted pregnancies.In addition, unprotected sex and other risky sexual practices have long term consequences including other sexually transmitted infections such as syphilis and hepatitis B, cancers associated with sexual exposures such as cervical cancer.Among young persons, complications from illegal abortions are also of great concern to public Health experts and the government.This study has shown that the following factors are associated with risky sexual behaviours: ever married, rural resident, male sex, comprehensive knowledge of HIV, substance use and age group 20-29 years.The finding on effect of gender on risky behaviour is not surprising.Females have been found to be generally less likely to engage in likely sexual behaviours.They are less likely to engage in accidental sex and have tendency to have unprotected sex only in stable, long term relationships. 18,19While men are more easily attracted to physical appearance, women generally enter into relationship based on affection.In many African settings, the society is more tolerant of male engaging in unacceptable sexual practices than they do for females.The disparity in reported risky sexual behavior between males and females could also be due to reporting bias because females are less likely to report risky sexual behaviours.Because majority of the respondents are married, risky behaviours among males implies that the actual risk would be shared among the couples.This is in line with findings from the analyses of new HIV infections in Nasarawa State Nigeria which revealed that close to half of all new HIV infections in the state occur among persons who reported low risk including married partners. 6The male partners often serve as the bridge between the traditional high risk populations such as sex workers, MSM and injecting drug users and the general population.
Our finding on the effect of location of respondents on sexual behaviour is consistent with some of the previous studies.In a previous study carried out among young adults in Nigeria, rural residents were significantly more likely than their urban counterparts to have multiple sexual partners. 20owever, some previous studies showed that persons residing in urban areas were more likely to have multiple sexual partner early sexual debut among other risky sexual behaviours. 11,12ural residents have less access to information on HIV and AIDS as well as other sexually transmitted infections.In addition, the level of socio economic status especially income level is generally lower in the rural areas.Regular alcohol, previous HIV test and self-risk perception were not significantly associated with risky sexual behaviour.Many of these findings are consistent with previous findings.] The data we analysed did not obtain information on quantity of alcohol used but rather on the frequency of use.While frequency of alcohol ingestion is important, combining it with quantity would have provided a better estimation of alcohol use.One important finding in this analysis is that even through risky sexual behaviour was high (65.9%),only 4.7% of the sampled population considered themselves to be at high risk of HIV.Previous studies have shown very varying findings on the association between personal risk perception and risky sexual behaviours. 23,24Individuals who engage in risky sexual practices but yet have low personal risk perception are usually less likely to alter such practices.Research has shown that risk perception have been found to be more important in altering risky behaviours than level of knowledge of HIV transmission.

27
The motivation to reduce the risky behaviours among these respondents is therefore expectedly very low and would need to be taken into consideration by HIV and AIDS control programmers and policy makers.One major strength of this study is that it was conducted among the general population and thus would enable us apply the findings across the board.This has become imperative because the burden of HIV in Nasarawa state Nigeria is high and population based interventions are required to reduce this burden.Many previous studies in the state and elsewhere only looked at key population such as the sex workers, persons who inject drugs and men who have sex with men (MSM).A limitation of this study is the reporting and recall bias.The study relied on recall of exposures and the outcomes.Some respondents may not be able to recall some of their past practices with precision.

CONCLUSION
Close to two third of the respondents engaged in risky sexual behaviours (65.9%) but only 4.7% considered themselves to be at high risk of HIV.The multivariable regression analysis showed that been married, rural resident, male sex, age between 20-29 years, use of substance and surprisingly knowledge of HIV were associated with risky sexual behaviour.

RECOMMENDATIONS
There is need to urgently intensify media campaigns, community based interventions including one on one communications to reduce risky sexual behaviours among the general population.Further studies should be carried out on cultural factors inhibiting positive change in sexual behaviours despite existing interventions.

Table 2 .1:
Scoring of the risky sexual behaviours variables

Table 3 . 3
Association between risky sexual behaviors and selected Characteristics among the general population, Nasarawa State Nigeria, 2014

Table 3 .
4 : Result of bivariable logistic regression model on factors associated with risky sexual behaviours among the general population in Nasarawa state, Nigeria 2014

Table 3 . 5 :
Result of multivariate logistic regression model on factors associated with risky sexual behaviours among the general population in Nasarawa state, Nigeria 2014