Perception of community-based health insurance scheme in Ogun State, Nigeria

Nigeria’s health performance has been one of the poorest worldwide over the last 2 decades [1]. Studies have attributed this to severe limited access to health care [2] due to factors that include inability to pay and inequitable health care provision especially among the poor rural dwellers who account for 70% of Nigerian population and are prone to further impoverishment [3-7]. Out-of-pocket health expenditure in Nigeria is over 60%, making it one of the highest globally [8,9]. The predominance of out-of-pocket health fi nancing makes it diffi cult for the poor to access quality health care services at the point of utilization [4,7].

private partnership policy that gives room for public and private sector participation in medical care provision [15,16].
While Nigeria accounts for 2.8% of the world's population, it bears, by contrast an alarming 10% of the global burden of under fi ve year old children and maternal mortality [7]. Some of government's key responses to this are the Health Systems Reforms. Part of the reforms of government in the health sector aimed at improving effi ciency in both public and private sectors and covering the marginalized poor is introduction of the National Health Insurance Scheme (NHIS) to help spread the risks and minimize costs of health care [5].
Evidence abounds on the impact of schemes to achieve universal health coverage [17]. A number of countries have successfully implemented National Health Insurance Scheme (NHIS) to provide effective and effi cient health care for all or majority of their citizens in order to achieve universal access to medical care [18,19]. In Nigeria, having conceived the idea of the NHIS in 1962, it was promulgated 37 years later in 1999 (Decree 35 1999 now Act 35 1999) and launched 43 years later in 2005 [14,20]. The NHIS was designed to provide comprehensive health care delivery at affordable costs, covering employees of the formal sector, self-employed, rural communities and the vulnerable groups [21]. Presently, the NHIS covers less than 7% of the population that mostly constitute civil servants and those in the formal sector through community-based health insurance scheme (CBHIS) and other health insurance scheme(s). For the realization of the presidential directive of ensuring universal access to quality health care in Nigeria by 2015, all States of the Federation need to have a buy-in since the provision of healthcare is a concurrent responsibility of the three tiers of government in the country [9,22,23].
The NHIS Agency is presently intensifying efforts at reaching the informal sector which constitutes those in greatest need, focusing especially on community based and other health insurance schemes [22,24].
Although some studies on various aspects of health insurance schemes in Nigeria have been carried out [25][26][27], these have been limited in scope, focus, and study population groups. Furthermore, these studies have also not examined community perception on critical issues that could enhance or militate against the success of the health insurance schemes from the perspective of all key stakeholders of the schemes within one study, as designed in this present study which also focuses especially on targeted clients. This study is part of a major study that addressed the role of health insurance schemes in improving equity and access to quality healthcare in the country; the success and challenges to effective implementation, given the plan to roll out the schemes across the nation starting from 2014. The States' level of buy-in into the health insurance scheme and their challenges were also assessed as well as other issues relating to sustainability of the schemes such as enrollees' satisfaction, uptake, acceptability and, non-enrollees perception of the scheme, were also investigated. This paper reports on the sub-study that assessed the community perception on critical issues for the success of the schemes, such as access, utilization, acceptability, willingness to pay, and issues on sustainability and effective operations of the scheme from the viewpoint of the target population, preparatory to the rolling out of the CBHIS in the study State. This is based on the premise that community participation is key to the success of the scheme. It also provides useful policy lessons with insights for improvement in the design of interventions aimed at improving the scheme for a guaranteed and affordable basic health care provision.

Study areas
The study was carried out in May 2014 in two selected urban local government areas (LGAs), Abeokuta North and Ijebu Ode in Ogun State, South West Nigeria where CBHIS was about to be rolled out for implementation. These are two of the twenty LGAs in the State for the state-wide phased roll-out of the scheme. LGAs are located about 100km north of Lagos and the Atlantic Ocean ( Figure 1).

Data collection procedures
The data collection exercise was carried out within a period of two weeks (a week per LGA). Multi-stage sampling technique was adopted in sample selection for the study. The sampling frame includes all the 6 LGAs namely: Abeokuta North; Abeokuta South (Ogun Central); Ado-Odo/Ota; Yewa North (Ogun West) Ijebu Ode; and Sagamu; (Ogun East)where CBHIS was planned to become operational as a pilot scheme in the State. Of these, two LGAs (Abeokuta North and Ijebu Ode) located in two (Ogun Central and Ogun East respectively) of the three senatorial districts of the State were selected by random sampling. First, two of the three senatorial districts (Ogun West, Ogun Central and Ogun East) in the State were selected by simple random sampling. Second, adopting the balloting approach, the names of the two LGAs in selected senatorial districts were written on pieces of paper, placed in separate containers, shuffl ed and one LGA was subsequently picked at random from each container without replacement. In each selected LGA, the only 5 wards comprising communities where CBHIS was planned to be fl agged off as a pilot scheme were purposively [29] selected for the study. In the selected wards, communities served by the Primary Health Care (PHC) facilities designated as health care providers for the CBHIS were selected for the study.
Purposive sampling technique was used to select all the interviewees and FGD participants for the study in the communities. Efforts were made to achieve geographical dispersal of these groups throughout the communities of the selected LGAs. The FGD participants were recruited from places such as households, market associations, mosques and churches through the community, and market and opinion leaders in the study communities, and it was ensured that the FGD participants did not know of one another's selection prior to the start of the discussions.
The Focus Group Discussions (FGDs) were conducted among community members that included adult males and females aged 25 years and above and adolescent males and females aged 24 years and below who are potential enrollees in the scheme in the study LGAs.
The data collection tools and procedures were pretested to test the adequacy and consistency of the research design and tools prior to the main study. The main sections of the FGD and in-depth interview guides from which the focus of this paper was derived included those that probed the background characteristics of the participants/interviewees such as age, religion, level of education, marital status and occupation.
Others probed into major health problems, people's health seeking behaviour, out-of-pocket expenditure on health, knowledge of insurance, household health insurance issues, acceptability, willingness to enroll and pay for health insurance, form of community engagement/involvement in the scheme, expectations of the scheme and perceived ways of sustaining the scheme.  A total of 6 FGD sessions were conducted in the study LGAs.
The focus groups formed were homogenous with respect to sex, age and social class in order to minimize inhibitions in the fl ow of discussion. They had common characteristics relating to discussion topic. In each LGA, FGD sessions were held among adults aged 25 years and above and adolescents aged 18-24 years. Each FGD session was held in a comfortable neutral setting, and consisted of a moderator, a note taker and 7 to 10 participants of the same sex with similar social background. A total of 61 males and females participated in the discussions.
Sample size for this study was decided based on saturation commonly used to determine sample sizes in qualitative research [30], The FGD sessions were recorded on tape and

Data analysis
The qualitative data from the FGDs and IDIs were analysed using the textual analysis programme, Textbase Beta, developed by Bo Summerlund and distributed by Qualitative Research Management of Desert Hot Springs, California, Textbase Beta software [31,32]. First, the tape recorded discussions in local language were transcribed and back-translated into English language. Second, the transcripts were subsequently typed, summarised, categorised, coded and sorted into text segments according to similarities and differences in individual opinions and views based on themes arising from the discussion guides.

Ethical issues
Ethical approval for the research protocol for the larger study with assigned number IRB/13/237 was obtained from the Institutional Review Board of the Nigerian Institute of Medical Research. The informed consent of the participants was sought and obtained in writing for those literate while the non-literates (to whom the contents of the document had been read out in local dialect) were requested to thumb print in the presence of a witness to signify their willingness to participate in the study. Privacy of each participant in the in-depth interviews was ensured through individual interview to avoid interference and infl uence by other persons. The participants were assured of the confi dentiality of the information provided for the purpose of the study and no personal identifi ers were used in the collection, analysis or reporting of the data.

Socio-demographic characteristics of participants
The selected participants for the focus group discussion (FGD) sessions were residents of the study communities. In general, the participants' ages ranged from 19 to 81 years with an average age of 46.8 years [43.9 years Abeokuta North vs. 52.4 years Ijebu Ode]. There was gender difference in the age distribution of the participants. The ages ranged from 20 to 40 years for men and 24 to 57 years for women and an average age of 49.8 and 42.5 years for men and women respectively. Most of them were married with a few who were widowed and never married. High literacy level was reported among the participants as most had a minimum of secondary education with a few having primary education while some had no formal education. Their occupations ranged from being artisans, drivers, clergymen, traders, farmers, and formally employed in both public and private sectors as civil servants and professionals to unemployed, pensioners, students. Some of the female FGD participants reported to be housewives.

Perceived common health problems of people in the study LGAs
The focus group discussions among the men and women in the communities showed the consensus that measles, dysentery (jedi-jedi), stomach pain, rheumatism, typhoid, diabetes, cough, pneumonia, anaemia, eye problem, convulsion, hernia, cholera, hypertension, with its consequence stroke, cancer and malaria are the common health problems that affect people of all ages and gender in communities of the study LGAs. It was disclosed that malaria affects a lot of children under fi ve years and pregnant women in their localities while pneumonia, cough and measles were reported to be more among children only. Some health problems such as hernia that are gender-related were mentioned mostly by male participants who are more at risk of the health challenge than their female counterparts in both LGAs.
Malaria was described by many of the participants as a very common public health challenge of people in communities of the two LGAs. A female participant explained that, "…malaria is a general problem, there's no one not aff ected…it is very common here because there are many mosquitoes around us...it aff ects everyone regardless of age, it even aff ects pregnant women." A male FGD participant in Totoro further emphasized that, "malaria is the most prominent illness in our community and it has serious health complications in children." A female FGD participant from Oke Ago Owu community in Abeokuta North LGA revealed that, "Malaria aff ects both the young and old people. Also, everybody coughs…it aff ects children as well as adults…I observed recently that cough is so rampant now among people in the community." A male participant in Ijebu ode expressed his concern thus: "Malaria is the commonest sickness in our community despite eff orts to control it through measures that include protection of oneself from mosquitoes and clearing of our environment…malaria is still rampant." Another male participant in Totoro was fatalistic in his resort to fate on the persistence of malaria in the community saying: "… even with these measures, malaria is still rampant because of our environment…we pray that God would help us."

Most FGD participants in both
LGAs reported to live in close proximity to health facilities located within their communities.
Citation: Adeneye  When asked what they do when they or their children are ill, a large number of the FGD participants reported going to the hospital for treatment. Other actions mentioned were: self-treatment at home, which varied from using left over drugs; going to the chemist or pharmacy; cooking and drinking herbal remedies prepared from leaves, roots and barks of some specifi ed trees (agbo) freely obtained from the home environment or bought from herbalists in the market; visiting a traditional healer to praying. A female FGD participant in Abeokuta North LGA stated that, "Our main means of taking care of the sick around us is by going to the hospital for treatment." Also, a male FGD participant in Ijebu Ode LGA pointed out that, "I rarely go to the hospital when feeling sick. Once I drink agbo, I usually get well but in case the sickness persists I'd just visit any nearby chemist shop to buy some drugs particularly paracetamol…that's all.
It is only if I don't get well after visiting these two places that I can then resort to going to the hospital." Regardless of their gender and location in the two LGAs, it was a consensus among those participants who preferred taking herbal remedies to going to the hospital because the herbal preparations are affordable, more accessible and perceived to be more effi cacious against all kind of illnesses than the orthodox medicines given in the hospital. On the other hand, the reason for high preference for hospitals by most participants particularly the women is because of the experienced and trained medical personnel and diagnostic equipment available for provision of quality health care. This is contrary to the perceived trial and error style of health management by traditional medical practitioners and patent medicine sellers.
It is worthy of note that most FGD participants demonstrated good health seeking behaviour with good knowledge of what do and how to take action as fi rst aid for example when a child is convulsing to alleviate the illness at home before seeking more appropriate treatment outside at the hospital. There was gender difference in the health seeking behaviour of the participants, as more women knew the action to take in providing health care within the home than the men. Similarly, more women than men prefer going to the hospital when they or their children are ill.

Knowledge and perception of CBHIS
There were mixed reactions on awareness of CBHIS from FGD participants in the two LGAs. Those in Abeokuta North LGA particularly the males had more awareness than those in Ijebu ode LGA. Among the few who knew about the scheme in Abeokuta North LGA, radio jingles and informal means such as family, friends, neighbours and the Community Development Associations (CDAs) were the main sources of information about the scheme.

Virtually all the female FGD participants in Abeokuta North
LGA knew about health insurance from their past experience either as students of higher institutions where they were mandated to enroll as students or as mothers paying a health insurance plan premium of N500.00 ($3.13) to N600.00 ($3.75) per term for their children in public primary and secondary schools. All the women however had negative stories to tell about poor implementation of the school health insurance plan.
A woman in a focus group in Abeokuta North LGA recalled that,  Their argument for this stance was their lack of perceived need for insuring themselves and or dependants against health risks. They claimed to be healthy, rarely fall sick and visit the hospital, hence no merit for pre-payment.
In a bidding process, the minimum amount the participants were willing to pay as annual premium ranged from N1,000.00 ($6.25) to N2,000.00 ($12.50) while the maximum amount ranged from N1,000.00 ($6.25) to N4,000.00 ($25.00) per individual. At the minimum and maximum bidding amount,  -Maternal and child health including antenatal booking, antenatal care, normal delivery, postnatal care a, family planning, emergency obstetric care, ectopic pregnancy, routine immunization, growth monitoring, nutrition services, treatment of diarrhoeal diseases in children, integrated management of childhood illnesses and prevention of mother-to-child transmission (PMTCT).
-Diagnostic tests that include malaria parasite, widals, PCV, pregnancy test, urinalysis, blood sugar, ultrasound scan for pregnant women, full blood count and AFB for tuberculosis diagnosis. virtually all the participants were willing to pay for all dependants in their respective households. There was no gender and geographical difference in the expressed willingness to pay the health insurance premium among the participants whenever the scheme becomes operational in their respective communities.
However, there was gender difference in the amount participants were willing to pay as more males were willing to pay higher amount than the females. Similarly, the trend of discussions showed more older FGD participants particularly those with more stable source of income expressed willingness to pay higher amount as premium that he younger ones and those with irregular source of income. The aspiration of some elderly ones aged over seventy years and above in some focus groups on provision of free health care services for them by government was very high. On the mode of payment, a very few expressed readiness to pay the premium once/ annum while most preferred instalmental payment, ranging from 2 to 4 times a year.
Despite their willingness to pay the premium, many FGD participants were still of the opinion that the scheme should be fully subsidised by government. Some participants particularly the female focus groups in Ijebu Ode LGA suggested "starting There was little or no sense of community ownership of the scheme among the FGD participants and those interviewed as a few of them also opined that "government represented by the LGA Chairman should be in control of the scheme."

Strategies of mobilizing and sensitizing the communities for involvement
On the best means of mobilizing and sensitising the community, the common responses were the use of microphones and loudspeakers for public announcements using mobile town criers, directed by community leaders, distribution of handbills, and airing of radio jingles.
For the purpose of the scheme, communities were mobilized and sensitized using town criers who were sent round the communities to beat the gong 1 to 2 days before any meeting The suggested ways of sustaining the scheme and ensuring continuous patronage are summarized in Table 3.

Discussion
This study assessed the community perception of CBHIS preparatory to its rolling out in Abeokuta North and Ijebu Ode LGAs of Ogun State, Nigeria. This is in a view to understand critical issues the community perceived are essential for the development, maintenance, institutionalization and sustainability of the scheme, and perceived way of effectively operating the scheme for the success of the scheme. The manifestation of high knowledge of health insurance and the impressive perceptions about the scheme is encouraging. This encouraging scenario was more evident in Abeokuta North LGA than in Ijebu Ode LGA and could be used as a platform to launch more community education programmes on the scheme. Hence, for the success of ensuring that more people in the two LGAs become aware of the scheme, community mobilisation and sensitisation about the scheme and its benefi ts as well as challenges need to be intensifi ed in the LGAs targeting the men who are mostly the household heads with fi nancial responsibilities for all household members. This can be achieved using the best applicable local channels of mobilisation and sensitisation identifi ed in the study. The level of awareness can also be increased through the production and distribution of handbills with simple illustrated messages and, by intensifying airing of information on the scheme, through both radio jingles and television, in different local languages particularly at prime time. Given that radio or television can bring household members information and new ideas [33], the use of radio and television will have more far-reaching impact on the population of LGAs where the scheme is presently being piloted and other LGAs of the State where the scheme would be expanded to after the pilot phase.
Even though the premium for the scheme was determined by the State government through acturial studies, a large number of the study participants who are potential enrolees of the scheme were artisans, traders and farmers who have been reported to earn low income [34]. Given that the generally accepted core of universal health coverage is that the health system should be fi nanced in accordance with the ability to pay as a desire to enhancing protection against health care costs and improve equity of access to health care [35], the ability to pay the insurance premium among the potential enrolees should be of concern to the planners of the scheme in the State.
It is important to highlight the fact that while the expressed willingness to pay the insurance premium in the study LGAs found similar to fi ndings on willingness to pay for Community Based Health Insurance Scheme in Lagos South-Western Nigeria [36] and Katsina, North-Western Nigeria [37] seems encouraging, there is need for caution because the focus groups and those interviewed could possibly have responded to a service not currently available, their actual willingness to pay could wane by time the reality of the amount of premium to be paid for themselves and dependants actually dawns on them taking cognisance of their earnings, household size, and ability to pay.
Entirely free programmes are unsustainable due to lack of government funds and limited time of donor support in sustaining programmes. Hence, avenues of mobilizing the communities to fi nancially embrace the scheme as a selfhelp programme as emphasized by Onwujekwe, et al. [38] and Adeneye, et al. [39,40] need to be pursued.
The managers of the scheme need to be mindful of some hospitals outside the scheme providing similar services at lower cost as found in the study which could jeopardize the  [47], community engagement with emphasis on involvement and participation is an essential element of health programmes. Positive change is reported to more likely occur when communities become integral part of a programme's development and implementation [33]. This is demonstrated in successfully implemented health programmes on Chagas disease, leishmaniasis, leprosy, onchocerciasis, African trypanosomiasis [48], schistosomiasis [39,49,50] and malaria [40].
The goal of the concept of community involvement and participation is to transform communities from passive recipients into active participants of health care programmes [34]. Local communities are to be actively involved in both programmes and decisions that affect their health, using their own supportive and developmental capacities to address their needs [47,51]. This result fi nds connection with the fi ndings of Shittu, et al. [52] that emphasised that community participation is essential to the continuity or collapse of the scheme given that the propensity of membership participation and the volume of peoples' contribution in CBHIS are crucial to determining the fi nancial freedom people enjoy under the scheme when seeking healthcare service. State for active participation in the scheme as emphasised in the CBHIS blueprint [22]. The State and local government authorities need to help mobilise and organise communities to implement the scheme successfully by providing technical assistance to build local management and technical skills to operate the scheme. Most importantly for the people to be actively involved and appreciate the signifi cance of taking control and ownership of the scheme, education programmes on the goal of the scheme which is to allow communities organise a pooling arrangement to fi nance their health care needs as a strategy of expanding affordable and quality health care to more people, are suggested to be designed and implemented targeting the communities. Futuristic plans also have to be designed targeting adolescents in this respect. This is because adolescence marks a period of developmental transition from childhood to full-fl edged adulthood; the experiences and behavioural patterns formed on CBHIS in these adolescent years can have lasting effects on their adulthood which will eventually impact on the sustainability of the scheme.
It is believed that involvement of the existing community structures identifi ed in the study in the implementation of the scheme particularly in the planning process of the scheme as a practical step in the blueprint for the implementation of CBHIS in the country [22] will provide a platform for easy take-off of the scheme particularly in other LGAs in the State where the scheme is planned to become operational after the pilot scheme.
Taking cognisance of the expectations of potential enrolees of the scheme as expressed in the study, and the need to guarantee consumer satisfaction and encourage patronage of the scheme, it becomes imperative that the minimum standards for primary health care in the country are met in the PHC facilities designated for the scheme in the LGAs with focus on systematic infrastructure upgrade, management, adequate number and proportion of the various cadres of health workers and support staff, service provision and essential drugs as emphasised by the National Primary Health Care Development Agency [53].
The suggestion on empowerment of the PHC health staff on how to regularly conduct research into the health needs of priority of their immediate population of coverage attests to the concern and expectations of the people about the scheme being responsive to their health needs. It is recommended that the scheme managers therefore address this concern as evident in the study because it is in conformity with the role of managers of CBHIS as stated in the blueprint for the implementation of CBHIS [22].

Limitations of the study
As a result of insuffi cient funding, the study was limited to only 2 of the 6 LGAs where CBHIS was planned to become operational as a pilot scheme in the State. This exploratory qualitative study was on a small size of community members who participated in the FGDs and IDIs. The data therefore may not be regarded as fully representative of and generalisable for all categories of people in the State. This, however, does not discount the validity of the fi ndings. Rather, further studies are needed to be carried out using a larger and more inclusive sample.