Level and determinants of pharmacovigilance programme awareness in Nigeria : A multilevel analysis

Previous studies have reported poor awareness of the Pharmacovigilance Programme (PP) among health professionals in Nigeria but similar information on the general population is unavailable. This study was designed to investigate the individual and contextual factors associated with awareness of the PP among the general population. The study utilised data collected in the 2012 National HIV and AIDS and Reproductive Health and Serological Survey which were collected through a multi-stage cluster survey among women aged 15-49 years and men aged 15-64 years across all the states in Nigeria. Data on regulatory activities about food and drugs as well as household and individual characteristics were extracted and analyzed using descriptive statistics and multilevel logistic regression. Mean age of the respondents was 31.45±11.8 years. Females constituted 50.1% while 68.7% resided in rural areas. Only 26.0% of the respondents were aware of the PP and this was significantly higher among those with higher education (57.7%) and those who have seen/heard any campaign on Adverse Drug Reactions (ADRs) (79.7%). Participants who had seen/heard any campaign on ADRs were more likely to be aware of the PP (Odd Ratio [OR]: 32.85, 95% Confidence Interval [CI]: 29.13–36.57). Tertiary education (OR: 4.29, 95% CI: 3.51–5.07), and secondary education (OR= 2.35, 95%C.I= 2.0–2.70) significantly increased PP awareness. Participants who were employed and those who resided in urban communities were more likely to be aware of the PP in Nigeria. Generally, awareness of the PP in Nigeria is low. Awareness campaigns should be re-packaged to reach rural dwellers and those with lower education.


INTRODUCTION
According to the World Health Organization (WHO, 2006), pharmacovigilance is defined as "the science and activities related to the detection, assessment, understanding and prevention of adverse effects or any other possible drug-related problems.The scope of pharmacovigilance in Nigeria has been widened to include herbal products, medical devices and vaccines.Good pharmacovigilance involves identification of the risk factors for adverse reactions in the shortest possible time in order to minimize any harm that can be caused.
Adequate pharmacovigilance reporting provides information that allows for evidence-based use of medicine and prevention of many adverse reactions (World Health Organization, 2006).
A major component of pharmacovigilance is the documentation of adverse drug reactions.According to WHO, adverse drug reactions (ADRs) are significant causes of sickness and deaths globally (World Health Organization, 2006).ADRs can also lead to disease resistance and relapse of diseases.The concept of pharmacovigilance is not well known among health professionals and the general population (World Health Organization, 2006).A study among community pharmacists in the state of Lagos (south-western part of Nigeria) reported that 55% of the participants had never heard of the word "pharmacovigilance" (Oreagba et al., 2011).Also, studies have shown that there is a low awareness of the pharmacovigilance programme among healthcare workers in the state of Sokoto (north-western part of Nigeria).It was reported that 95.1% of the physicians in the state were not aware of Adverse Drug Reactions (ADR) reporting systems (Bello and Umar, 2011).Likewise, it was revealed that 78.1% of the resident doctors in tertiary hospitals in Edo and Lagos States had inadequate knowledge of pharmacovigilance and that 71.2% were unaware of the ADR reporting scheme (Ohaju-Obodo and Iribhogbe, 2010).
In several developed countries awareness of ADRs reporting was also low.An Australian study found that only 10.4% of the consumers in Australia were aware of the available ADR reporting scheme (Robertson and Newby, 2013).Similarly, the awareness of an ADR reporting scheme among the general population was reported to be low in the UK: only 8.5% of the general population in UK are aware of the UK system for collecting information on suspected ADRs (Yellow Card Scheme) (Fortnum et al., 2012).Also, a study among physicians at the Malaysian Medical Centre showed that about 40% of the participants were not aware of the existence of the national reporting system in Malaysia (Aziz et al., 2007).
All the previous studies on the awareness of the Nigerian Pharmacovigilance Programme (NPP) were conducted among health care professionals (health care workers).Information on the awareness of the NPP in the general population is scarce in the literature.Furthermore, factors associated with the low awareness of the NPP have not been examined in any literature.Hence, the present study was initiated to assess the awareness of the NPP in the general population and examine the individual and community-level factors associated with awareness of the NPP.

Study settings
Nigeria is a federal republic with 36 states including the Federal Capital Territory, Abuja.Based on the 2006 National Population Census figure, Nigeria"s population was 140,431,790 (Federal Republic of Nigeria, 2007).The National Pharmacovigilance Centre (NPC) in Nigeria was opened in 2004 and affiliated to the WHO Collaborating Centre for International Drug Monitoring.The NPC raises awareness on the magnitude of drug safety problems, and encourages health professionals in becoming vigilant in the detection and reporting of ADRs.There are pharmacovigilance centres in all the states in Nigeria (National Pharmacovigilance Centre (NPC), NAFDAC Nigeria, 2004).

Study design and data extraction
The present study was a secondary analysis of data collected in the 2012 National HIV & AIDS and Reproductive Health and Serological Survey (NARHS Plus II) in Nigeria.The survey covered all the 36 states of Nigeria and evaluated female (aged 15-49 years) and male (15-64 years) participants selected through a multistage probability sampling method.In the parent study (NARHS), data were collected using a structured and semi-structured questionnaire in face-to-face interviews.The questionnaire covered characteristics of the household and survey populations, sexual behaviors, opinions and attitudes about HIV and AIDS knowledge, regulatory activities about food and drugs, etc.Data on characteristics of household and survey populations and regulatory activities about food/drugs were extracted for all respondents (31235 respondents) and used for the present analyses.

Variable identification and data management
The main outcome variable in the present analysis was awareness of Nigerian Pharmacovigilance Programme measured by the question: "Are you aware of any government programme asking people to report adverse reactions to drug/food products in Nigeria" (Responses: "Yes" was coded as 1 while "No" was coded as 0).The independent variables were classified into two levels: person or individual level (including sex, age group, marital status, religion, occupation and educational level) and community-level characteristics (place of residence (location) and geo-political zone).Other individual level characteristics investigated included whether the individual had seen/heard any advertisement on the National for Food, Drug Administration and Control (NAFDAC) programmes on what people should do when they experience adverse reactions to drug/food products, how often individuals listen to the radio and how often individuals watch television.
Furthermore, missing data were excluded from analysis.Participants" occupations were recoded into the following categories; not working, professionals/civil servant, semiskilled/self-employed, student and unskilled/agricultural worker.Marital status was also recoded as: never married, currently married / living with sexual partner, separated / divorced *Corresponding author.E-mail: oalo@cihpng.org.Tel: +2347066107467.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License and widowed.

Statistical model and analysis
Descriptive statistics were used to summarize the variables in the study while association between individual-level factors and the outcome variables were initially investigated using the Chi-square test.In order to further examine the individual-level and communitylevel factors associated with the outcome variables, the hierarchical structure of the data cannot be overlooked.Hence a multi-level binary logistic regression was used to assess the role of measured individual and community (cluster) factors on the outcome variables.
Specifically, a three-level random intercept model was used to assess the predictive values of measured individual and community-level factors using the "gllamm' command in Stata (Rabe-Hesketh and Skondral, 2008).Two models were estimated: a null model that contained no covariates and a full random intercept model that included fixed effects (individual variables) and community and state-variables as random effects.The null model was used to verify if the magnitude of random effects at the community level and state level justified assessing the random effects at that level.The Intra-Class Correlation (ICC) was used to measure the amount of dependency that was observable due to the clustering of the data at the community level.All analyses were carried out using STATA version 12.

Individual-level characteristics of participants
Females constituted 50.1% of the total sample with 68.7% residing in rural areas.The mean age of the respondents was 31.45 years (S.D=11.8).One-fourth (26.0%) of the respondents were aware of the pharmacovigilance programme in Nigeria.The proportion of males (28.8%) reporting awareness of the pharmacovigilance programme was significantly higher than the female (23.1%) participants (p<0.001).Also, the proportion of respondents who reported awareness of PP was higher among those with a tertiary education (57.7%) than any other level of education (p<0.001) in Nigeria.
Similarly, the proportion reporting awareness of the PP (79.7%) among respondents who had heard/seen any NAFDAC campaigns on ADRs was significantly higher than those who had not heard of any NAFDAC campaigns (9.4%) in Nigeria (Table 1).Furthermore, the proportion of the participants reporting awareness of the PP in Nigeria was significantly higher (p<0.001)among older respondents (27.1%), professional/civil servants (53.5%), those who had never been married (31.2%) and those who watched TV (27.8%) or listened to radio at least once in a week (28.1%)(Table 1).
The bivariate relationships between selected characteristics and awareness of the PP in Nigeria as reported in Table 1 could be due to interrelationships among the various measured characteristics as well as the unmeasured characteristics at the individual, community and state levels.
The intra-class correlation (ICC) in the intercept-only model indicated that 68.6 and 14.7% of the total variance in awareness of the PP in Nigeria was attributable to the dependency of observations within the communities and states respectively.This implies that, the awareness of the Nigerian Pharmacovigilance Programme correlated significantly within community and state (Table 2).
In Table 3, respondents who had heard/seen any NAFDAC campaign on ADRs were more likely to be aware of the PP (OR: 32.85, 95%CI: 29.13 -36.57) compared with those who had never heard/seen of any NAFDAC campaigns in Nigeria.Also, a positive association between educational attainment and awareness of the PP was evident.Participants with a tertiary education (OR: 4.29, 95%CI: 3.51 -5.07), a secondary education (OR= 2.35, 95%C.I= 2.00 -2.70) and a primary education (OR: 1.51, 95%CI: 1.26 -1.75) were more likely to be aware of the PP compared to those without a formal education in Nigeria.Similarly, participants who resided in the urban communities were more likely to be aware of the PP (OR: 1.42, 95%CI: 1.28 -1.52) compared with those who resided in the rural communities in Nigeria.
However, after controlling for observed factors, the residual intra-class correlation for the community and state-levels were 17.03 and 7.25% respectively.

DISCUSSION
This study was designed to assess factors affecting the awareness of the Nigerian Pharmacovigilance Programme based on the data from the National HIV and AIDS and Reproductive Health and Serological Survey of 2012.We found that awareness of the Nigerian Pharmacovigilance Programme was generally low among the general population.Though it may be difficult to explain the exact reason for this low awareness, it however revealed that not much emphasis had been placed on the pharmacovigilance campaign in Nigeria.Also, except for some occasional advert placement in the media, comprehensive education about adverse drug reactions (that focused on the population) is practically non-existent in Nigeria.Findings from this study are not too different from what have been found in other countries, both in developing and developed countries.For instance, awareness of pharmacovigilance programme was also reported to be low in some developed countries like the UK (Fortnum et al., 2012) and Australia (Robertson and Newby, 2013).
Education has had a positive influence on the awareness of the pharmacovigilance programme in Nigeria; those with a higher level of education are more likely to be aware of the pharmacovigilance programme.It is really not because information about pharmacovigilance are built into the curriculum of higher education but maybe because most pharmacovigilance programmes and advertisements are packaged for thos who can read and for those who are educated.In China,  a study to investigate the awareness of pharmacovigilance among healthcare professionals also revealed that the educational level was highly related to the degree of pharmacovigilance awareness (Xu et al., 2009).
Furthermore, in the Nigerian general population, awareness of the PP is largely dependent on whether an individual had heard of or has seen any NAFDAC advertisements/campaigns on what people should do when they experience adverse reactions.In fact, in the current study, about 80% of those who claimed to be aware of the NPP were those who had heard or seen NAFDAC advertisements on ADRs in the media at one point or the other.This may be because the NAFDAC advertisements/campaigns are very informative on the pharmacovigilance programme in Nigeria.
Also, the type of place of residence influences the awareness of the pharmacovigilance programme in Nigeria.This may be as a result of NAFDAC activities having been more visible in urban communities than in rural communities.For instance, NAFDAC and Your Health Programme being aired on some Nigerian television stations and radio stations may be urban biased.This is coupled with the erratic nature of power supply in most localities in Nigeria.More importantly in rural areas people may not be able to afford an ensuring power supply through their personally-owned generating sets.
However, how often an individual listens to a radio or watches the television was not associated with the awareness of the Nigerian Pharmacovigilance Programme.This may be because very few jingles or programmes related to pharmacovigilance are available on Nigerian television and radio stations.It could also be linked to the fact that most NAFDAC jingles are in English rather than the local languages.
Finally, even after controlling for individual and community-level characteristics, there was still a considerable inter-community and inter-state heterogeneity in the awareness of the Nigerian Pharmacovigilance Programme.The intra-class correlation at the final model was still considerably high.
This may be due to the effects of some unobserved community-level and state-level factors such as: media concentration of the community, etc.

Awareness
of the Nigerian Pharmacovigilance Programme in the general population was low.Also, the NAFDAC campaign on what people should do when they experience adverse reactions has a significant impact on the awareness of the Nigerian pharmacovigilance programme in the general population.Awareness of the Nigerian Pharmacovigilance Programme is further a condition of the level of education, and the place where people live.Awareness of the PP could be enhanced through a more robust NAFDAC campaign on ADRs.Findings from this study will provide NAFDAC"s management team with evidence-based regulatory decision-making and the opportunity to ensure the need to develop low-literate campaigns, jingles, billboards and hand bills.Further, NAFDAC should ensure and adopt an approach that will intensify its presence in rural areas in terms of its sensitization of the populace.

Table 1 .
Awareness of Nigerian Pharmacovigilance Programme according to selected background characteristics.

Table 2 .
Random intercept only model for awareness of Nigerian Pharmacovigilance Programme.

Table 3 .
Three-level random intercept model for factors associated with awareness of NPP.
*Significant at p<0.05 a Ever heard/seen any advertisement on NAFDAC programme on what people should do when they experience ADRs.ref refers to reference category.