Prevalence of and Factors Influencing Parent–Child Communication About HIV/AIDS, and Sexual and Reproductive Health Issues in Nigeria

This study investigated the prevalence of and factors that determine parent–child communication about HIV/AIDS and sexual and reproductive health (SRH) issues in Nigeria. The study used data from the National HIV/AIDS and Reproductive Health Survey (NARHS). The data are nationally representative and offers sociodemographic information on women aged 15 to 49 years and men aged 15 to 64 years. Prevalence of parent–child communication about HIV/AIDS, family planning, and contraception was generally low at 37.4%, 32.5%, and 9.5%, respectively. Determinants of discussions with children on HIV/AIDS, family planning, and contraception were respondents’ zone, economic status, and educational attainment. Parents held more discussions with female children than with the males. Parents aged 50 to 64 years, compared with the younger parents, were double likely (odds ratio [OR] = 2.04; 1.52-2.73) to discuss HIV/AIDS with their wards older than 13 years. Male parents, parents living in urban areas, and parents with richer economic status had higher odds of discussing HIV/AIDS and SRH issues with their wards.

with the children is one of the fundamental responsibilities of the family. Nundwe (2012) supports this reality as the author explains that communication within the family system affects the sexual behavior of the child given the fact that communication about sexual behavior and general reproductive health issues within the family is crucial during the adolescent years (Nundwe, 2012). Research has shown that open communication between parents and children has a strong correlation with a reduction in negative sex behaviors by the children when they become sexually active (Leeds et al., 2014;Rosenstock, Strecher, & Becker, 1994;Viner et al., 2012). These studies confirm that the family is very critical to the sexual socialization of children and adolescents. The vital information and messages between parents and children could affect risky sexual behavior among adolescents (Dilorio, Pluhar, & Belcher, 2003). Parent-child communication about sexuality, and the understanding of how parents influence their children's feelings, concerns, and decision making have a long history dating back to at least 1965 (Dubbe, 1965). However, during the past decades, parent-child communication studies have expanded beyond sexuality to include content, predictors, and outcomes associated with these discussions (Bastien et al., 2011;Jejeebhoy & Santhya, 2011;Mpondo et al., 2018;Obono, 2012;Soon et al., 2013;Yadeta, Bedane, & Tura, 2014). One reason for this increased interest is the HIV epidemic, which continues to affect millions of adolescents among its victims (Dilorio et al., 2003).
A meta-analysis of 95 studies on parent-child communication about sexuality by Dilorio et al. show that most parents reported discussing "sex" in general with their children; the most frequently discussed topics include menstruation, reproduction, pregnancy, birth, HIV/AIDS, and sexual values (Dilorio et al., 2003). Many parents, however, confessed to having faced difficulties when communicating sex and sexuality with their children. Some of these problems include embarrassment, difficulty in acknowledging and accepting adolescent sexuality, and the use of communication styles with which adolescents were less likely to feel comfortable. This phenomenon notwithstanding, the study notes an apparent correlation between parent-child communication about sexuality and adolescent sexual behavior (Dilorio et al., 2003). Similarly, the perception of adolescents about parentchild communication on sexuality are that such communication is affected by low parental care, difficulty in talking to their parents about problems, and valuing their friends' opinions in making serious decisions. These were significantly associated with compromised behavioral and emotional health (Ackard, Neumark-Sztainer, Story, & Perry, 2006). Given this fact, parent-child communication of HIV/AIDS and SRH issues in the family is an important factor that deserves the attention of all stakeholders in public health. The extent to which the parents openly share information about health issues with their children determines the children's health behavioral patterns. Adolescent sexual behaviors are partly influenced by family members and peers with whom they interact (Christopher, Johnson, & Roosa, 1993;Pistella & Bonati, 1998). Besides, parents play a critical role in the socialization of their children, of which sexuality is critical. Mpondo et al. (2018) buttress this reality when they say that "sexual health communication that is provided by the parents is considered very important," This fact is further corroborated by Soon et al., when they explained that parents play important roles in influencing adolescent sexual decision-making and access to information about HIV and general SRH issues (Soon et al., 2013).

Theoretical Framework
The Health Belief Model (HBM) was adopted as the theoretical framework for this study. The model, being one of the earliest theories of health behavioral change communication, operates on the tenets that people's decisions and communication about issues concerning their health are always influenced by some key variables: perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cue to action, self-efficacy, and demographic variables (Abraham & Sheeran, 2015;Taylor et al., 2007). For instance, if people believe that they are susceptible to a certain health condition, they are likely to discuss the health matter and adopt new health behavior. Also, if people believe that the health condition has severe consequences, they are likely to take actions that prevent such negative consequences. People would take action if they believe that adopting the new health behavior would mitigate their susceptibility to the negative health condition, and if they perceive higher benefits of adopting the new health behavior in relation to the costs to be incurred, they are likely to discuss or adopt the new health behavior. Furthermore, individuals are likely to discuss and adopt the health behavior if they are exposed to factors (e.g., communication campaigns, television adverts, a reminder from their doctors, and encouragement from partners or peers) that prompt or trigger discussion and action. Finally, people are likely to act if they believe that they are capable of performing the prescribed action successfully (Champion & Skinner, 2008;Rosenstock, Strecher, & Becker, 1988) However, since the HBM is conceptualized on the principle of a value-expectancy relationship (Brewer & Rimer, 2008), whether or not the foregoing would happen is also influenced by demographic variables such as age, gender, location, and social status. Information that adolescents share among their peers, who may or may not be well informed, often form the major source of the scanty knowledge that most adolescents have on reproductive health issues (Yadeta et al., 2014). In this present study, we investigated how demographic variables, as one of the key constructs of HBM, influence patterns and frequency of parent-child communication of HIV/AIDS and SRH issues in Nigeria. We predicted that how Nigerian parents communicated health issues with their children would be shaped by several demographic factors.

Method and Materials
The data for this study are from the 2012 NARHS Plus II (FMoH, 2013). It is a nationally representative cross-sectional data that contained information on knowledge, attitude, practices, and perceptions on HIV/AIDS, reproductive history, sexual characteristics of women aged 15 to 49 years and men aged 15 to 64 years selected using multistage cluster sampling technique. The respondents were drawn from households in the rural and urban areas in all the 36 states and the Federal Capital Territory (FCT) Abuja, Nigeria.
Urban and rural Local Government Areas (LGAs) were selected in each state after stratifying the LGAs into rural and urban localities at Stage 1. Thirty clusters each were then randomly selected from each of the selected LGAs at Stage 2. Selection of households from the clusters and 32 individuals from the selected households was carried out at Stages 3 and 4, respectively. The surveys used pre-tested structured questionnaires by interviewers trained in questionnaire administration. The analysis in this study was based on data from 10,639 respondents who had children older than 10 years.
From the constructs of the HBM (Abraham & Sheeran, 2015;Taylor et al., 2007), we selected the demographic variables to guide our construction of the dependent and independent variables. We examined how demographic characteristics influenced communication about HIV/AIDS and SRH issues between parents and their children. These variables and the relationship between them are explained as follows:

Dependent Variable
The outcome variable in this study is whether a parent having a child older than 10 years discussed (a) sexual relationship, (b) HIV/AIDS, and (c) Family Planning and Contraception with the child within 12 months preceding the survey.

Independent Variables
The independent variables used in this study include "sex of the respondents," "age of respondents," "educational attainment," "religion," "marital status," "wealth status," "location of residence," and "geo-political zones."

Statistical Analyses
First, we used descriptive statistics to show the distribution of the respondents who had children older than 10 years and the prevalence of discussions with their wards on (a) sexual relationship, (b) HIV/AIDS, and (c) family planning and contraception with the child within 12 months preceding the survey. We computed and showed a pattern of the prevalence of discussions on either or both HIV/AIDS and family planning and contraception (Table 2). Bivariate analyses were performed to determine the significance of the association between dichotomous dependent and independent variables using the Pearson Chi-square (χ 2 ) test of association (Table 1). We used logistic regression to model the relationship between the dependent and independent variables at the bivariate level (Table 3). The significant independent variables in the bivariate analysis were adjusted for multiple logistic regression models. Logistic regression models are suitable where the variable of interest has binary outcomes. They determine the association between a dichotomous dependent variable and independent variables by converting the dependent variable to probability scores taking on values between zero and one as shown here: where y i refers to different levels of the dependent variable for the ith observation and x ij is the jth independent variable (j = 1,2,...k) for that observation. The parameter β j is the jth coefficient of covariate x ij , and it indicates the effect of the covariate x ij in the fitted model.

Results
Most of the respondents (54%) were between 35 and 49 years of age, and the proportion of male respondents was 48%. Table 1 shows the distribution of discussion of various types of reproductive health issues considered in this study, which respondents have had with their male and female wards. Discussion of sexual unions with male and female wards was higher among older parents than among younger parents. Fathers appeared to discuss sexual unions with their female child (38%) more than a male child (34%) while the levels of discussion on HIV/AIDS with female and male children were close at 35% versus 34%. Similarly, mothers discussed sexual unions more with their female child (40%) than with the male child (30%) and had more discussions on family planning and contraception with the female child (10%) than with the male child (7%). Generally, parents held more discussions with the female child than with the male child, for instance, 39% versus 32% on sexual unions, 33% versus 31% on HIV/AIDS, and 10% versus 7% on family planning and contraception. There were significant differences in parent-child communication on all topical issues considered both by sex of the children and zone of residence of the parents (Table 1 and Figure 1). In Table 2, we showed levels of discussion with children irrespective of the sex of the children. Generally, only 37% of parents discussed sexual union with their children, compared with 33% discussing HIV/AIDS and less than 10% who discussed family planning and contraception with their wards within 12 months preceding the survey. Only one third (33%) of the parents discussed either HIV/AIDS or family planning with their wards while only 9% discussed both with their children. Discussions on the reproductive health issues with wards appeared higher among better-educated parents than less educated, higher among parents from households with higher economic status than those from poorer economic status. Discussion of family planning and contraception was higher among other Christians (12%) and Catholics (11%) than among the Islam believers (6%).
At the bivariate level, the odds of discussing HIV/AIDS with wards increased with age of parents. Parents aged 50 to 64 years were double (OR = 2.04; 1.52-2.73) likely to discuss HIV/AIDS with their wards older than 13 years than the parents aged 20 to 24 years. Male parents and parents living in urban areas had higher odds of discussing HIV/ AIDS with their wards. Parents living in the South West were more than three times (OR = 3.26; 2.82-3.77) likely to discuss HIV/AIDS and SRH issues with their wards. Parents with richer economic status were almost four times (OR = 3.56; 2.95-4.06) more likely to discuss family planning and contraception with their wards than were those with poorer economic status. While adjusting for other independent variables, respondents' age, zone, economic status, educational attainment, and religious affiliation significantly determined whether a parent discussed HIV/AIDS with their wards. Determinants of discussions with children on family planning and contraception were respondents' zone, economic status, educational attainment, and religious affiliation at the multivariate level. The odds of parents with higher education discussing family planning and contraception with wards tripled (adjusted OR [aOR] = 3.08; 2.40-3.96) the odds among parents with no formal education (Table 3).

Discussion
We found a significant relationship between the sociodemographic characteristics and parent-child communication of HIV/AIDS and SRH in Nigeria. Prevalences of parent-child communication on sexual unions, HIV/AIDS, and SRH were generally at 37.4%, 32.5%, and 9.5%, respectively. On all health-related issues that parents discussed with their children, female children got their parents' attention more than the male children at 39% versus 32%. Specifically, fathers discussed sexual unions with their female children (38%) more than they did with their male children (34%). Similarly, mothers discussed sexual unions more with their female children (40%) than they did with their male children (30%), and also had more discussions on family planning and contraception with female children (10%) than they did with their male children (7%). However, with regard to HIV/AIDS, both female and male children received almost equal levels of their parents' attention (35% vs. 34%).
The inference from the foregoing findings is that Nigerian parents seemed to believe that female child deserve more attention than their male counterparts, in terms of sexuality and health communication, perhaps because of the perception that, compared with the male adolescents, the female adolescents are more vulnerable to health risks. This issue is a cultural stereotype in most African countries. As established in literature, African parents tend to give more attention to female children than their male counterparts especially in issues that have to do with risks associated with HIV, sex, and reproductive health (Bastien et al., 2011;Manu et al., 2015;Obono, 2012;Yadeta et al., 2014). For example, as Rivers and Aggleton (1999) explained, in most developing African countries, female adolescents are considered to be at heightened risks of HIV infection more than their male counterparts. Also, the cultural values placed on virginity in most African communities compels parents to give more attention to their female children in matters relating to SRH.
We also found that irrespective of the sex of their children, more parents (37%) discussed with their children issues about sexual union compared with 33% and less than 10% of the parents discussing HIV/AIDS and reproductive health issues, respectively, with their wards. This pattern suggests that parents preferred to discuss sexuality issues more while they gave the least attention to reproductive health matters in their communication with their children. Findings also show that the level of education and religious affiliations influenced the intensity and patterns of HIV/AIDS, and SRH communication in the family. Parents with higher education appeared to discuss reproductive health issues more with their wards than the less educated parents.
At the bivariate level, the probability of parents with higher education discussing family reproductive health (RH) issues with their wards tripled the odds among parents with no higher education. Also, parents with richer economic status were almost four times more likely to discuss family planning and contraception with their wards than those with poorer economic status. It could be inferred from the foregoing findings that the more educated and economically strong the parents are, the higher the likelihood that they would discuss HIV/AIDS and SRH issues with their wards. Our findings here share some level of similarity with a study by Yadeta et al. (2014), which reports that the probability of parent-adolescent discussion on RH issues in Harar, Eastern Ethiopia was found to be significantly higher among parents who had acquired some form of formal education compared with parents with no formal education.
Only 6% of Muslim parents, compared with 12% of Christians and 11% of Catholics, communicated health issues with their wards. Generally, as the findings confirm, economic status, level of education, and religion shape the patterns of discussion of SRH issues in Nigerian households, but religion appears to reduce the intensity of parent-child discussion of HIV/AIDS and SRH in Nigeria more than other factors. This is not surprising given the fact that apart from ethnicity, religion is another strong sociocultural index in Nigeria. A previous study by Mariga et al. (2012) also confirmed that religion, among other factors, was found to prevent effective discussion of HIV/AIDS and SRH issues among sexually active adolescents, especially those from the rural parts of Nigeria. Similarly, in a recent study by Ahmed Note. Discussions were with children aged 11+ years within 12 weeks preceding the survey. FPC = family planning and contraception. *Significant at 5% χ 2 test.
et al. (2018), low social support, gender norms, and HIV stigma were among other factors why people from low-and middle-income countries would not discuss or initiate treatment of HIV/AIDS. Mpondo et al. (2018), in a study they conducted in South Africa, found that parent-child communication on sexual health matters in rural communities was limited to messages that warned against pregnancy, while the messages were also fraught with cultural idioms that could be difficult to comprehend because of their ambiguous meanings. These findings also confirm the findings of our study that sociocultural and demographic factors influenced discussion of HIV/AIDS and SRH issues between Nigerian parents and their children.
Findings of this study have implications for the HBM, which we adopted as the theoretical framework. The core tenet of the theory is anchored on the philosophy that some factors always serve as stimuli that shape people's overall health behaviors. These factors include perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cue to action, self-efficacy, and demographic variables (Abraham & Sheeran, 2015;Taylor et al., 2007). It is interesting to note, as established in this study, that parents across Nigeria discussed HIV/AIDS and SRH issues with their children. However, what is more interesting is that these discussions were influenced by various sociodemographic determinants such as respondents' zone, economic status, educational attainment, and religious affiliation at the multivariate level. We, therefore, consider it relevant to explain how our findings here are relevant to the tenets of the HBM. The model operates on the principle that people's decisions and communication about issues relating to their health are always influenced by some key variables as already identified earlier. In order words, how people modify their behavior and how they discuss a given health issue depend largely on the values they place on some variables that have implications for their total well-being. This principle is what is described as value-expectancy (Brewer & Rimer, 2008). As the HBM provides, whether an individual would discuss some health issues or change his or her behavior in response to the health issues is often influenced by a number of factors (Abraham & Sheeran, 2015;Taylor et al., 2006), one of which is the nature of demographic variables, which our study examined.
Our present study has established that demographic characteristics influenced communication about HIV/AIDS and SRH issues between parents and their children across Nigeria. As the findings showed, patterns and intensity of communication about HIV/AIDS and SRH in Nigerian households vary according to the identified demographic variables. This suggests that, for instance, male parents' perception of, and subsequent discussion about, HIV/AIDS and SRH issues are different from that of female parents. The same phenomenon applies to younger parents versus older parents, rich parents versus poor parents, parents with formal education versus parents without formal education, parents who reside in urban places versus parents from less-urban places, and religious affiliations. Also, Nigerian parents communicate about HIV/AIDS, sexuality, and reproductive issues with their female wards more often than they do with their male children.

Conclusion
The primary objective of this study was to establish the factors that influenced the patterns and frequency of parentchild discussion about HIV/AIDS and SRH in Nigerian households. This study is different from most of the previous similar studies from Nigeria, which focused on some regions or ethnic constituents of the country. The present study relied on a body of nationally representative cross-sectional data from a NARHS. The implication is that findings of this study have a high generalizability value.
We found that parent-child communication of HIV/ AIDS and SRH issues is generally low in Nigerian households, and the patterns of this prevalence are not even. Different sociodemographic variables such as respondents' zone, economic status, educational attainment, and religious affiliation of respondents influenced this prevalence. For instance, parent-child communication of HIV/AIDS and SRH issues was higher in richer households than in poorer households, higher in families with highly educated parents than with parents with lesser or no formal education, and most prevalent in the Southwestern part of Nigeria but least prevalent in the Northern region of the country. On the same subjects, female adolescents received their parents' attention than their male counterparts did. The fact that Nigeria is a heterogeneous nation with high variations in terms of ethnicity, religion, cultural values, and other demographic variables suggests that patterns and intensity of discussing HIV/AIDS and SRH among the citizens are not expected to be even across the country. Given these variations, health communication scholars and stakeholders should be sensitive to sociodemographic variables that always influence prevalence and patterns of parent communication of HIV/AIDS and SRH issues especially in a setting such as Nigeria with manifest sociocultural heterogeneity. Although the importance of HIV/AIDS communication on controlling the transmission of HIV/AIDS is replete in literature, the findings of our study showed that HIV/AIDS communication among parents and their wards is generally poor in Nigeria, especially among parents with lower educational and poorer economic status. Educational attainment of parents and wealth status of their households are main predictors of the likelihood of parents to engage their wards in HIV/AIDS communication. This would have negative consequences on the awareness of HIV/AIDS, and knowledge about transmission and prevention of HIV/ AIDS and the possibility of adopting good control measures. Given the fact that greater exposure to HIV/AIDS communication could enhance HIV knowledge, encourage condom use, encourage HIV counselling and testing, discourage risky sexual behaviors, and reduce discrimination and stigmatization of people living with HIV/AIDS, there is the need to give urgent and sincere attention to issues that could improve parent-child communication to curtail the incidence and prevalence of HIV/AIDS.
Health communication on HIV/AIDS and SRH could be improved by first targeting the parents and guardians in a massive campaign against HIV/AIDS. Such massive campaigns will increase the knowledge base of parents on HIV/ AIDS, which they can in turn pass on to their children. The capacity of parents and guardians to discuss this important health topic should be developed. Similarly, the misconception of some parents that sexual issues should not be discussed with children should be corrected as the knowledge gap created by absence of this communication could be detrimental in the long run. Parents should be made to understand that communication on HIV/AIDS and SRH between them and their wards is crucial. They should not allow their children to be misinformed by peers and other people outside the family.

Study Limitation and Strength
Secondary data from a survey that requires event recall was used for this study. It is not unlikely, therefore, that recall bias might have been introduced. However, the large size of the data might have eroded a likely effect of such bias on our findings. Besides, studies on the effect and influence of parent-child communication as a strategy to prevent the spread of HIV/AIDS in Africa is relatively scanty, especially in recent times. The present study, therefore, addressed one of the multisectoral approaches to reduction of the scourge of a contemporary public and clinical health problem ravaging developing countries.