African Journal of Reproductive Health Women's Health and Action Research Centre ISSN: 1118-4841 Vol. 7, Num. 1, 2003, pp. 17-26

African Journal of Reproductive Health, Vol. 7, No. 1, April, 2003 pp. 17-26

Gate-Keeping and Women's Health Seeking Behaviour in Navrongo, Northern Ghana 

Pierre Ngom¹, Cornelius Debpuur², Patricia Akweongo², Philip Adongo² and Fred N Binka³

Senior Research Scientist, African Population and Health Research Centre, Population Council, P.O. Box 17643, Nairobi, Kenya. ²Senior Research Scientist, Navrongo Health Research Centre, P.O. Box 114, Navrongo, UER, Ghana. ³Executive Director, INDEPTH Network, 9 Adenkum Loop, P.O. Box GP 14550, Accra, Ghana.
Correspondence: Pierre Ngom, Senior Research Scientist, African Population and Health Research Centre, P.O. Box 10787, 00100 GPO, Nairobi, Kenya. Tel: (254)2.2713.480/1/2/3 Fax: (254)2.2713.479 E-mail: pngom@aphrc.org

ABSTRACT

Among the Kassena-Nankana of northern Ghana, compound heads and husbands impede women's prompt access to modern health care. This paper shows that such gate-keeping systems have a negative effect on child survival. To investigate the social construction of compound-based gate-keeping systems, the authors relied on a series of qualitative interviews conducted in the Kassena-Nankana district. These data reveal that whilst compound heads are gate-keepers for spiritual reasons, husbands play such role for economic reasons. But more important, this article presents health interventions that are on trial in Navrongo (northern Ghana) and how they undermine such gate-keeping systems. (Afr J Reprod Health 2003; 7[1]: 17–26)

RÉSUMÉ

Le contrôle de passage et le comportement des femmes à la recherche de la bonne santé à Navronge, au Ghana du nord. Chez les Kassena-Nankana du Ghana du nord, les chefs de concessions et les maris entravent l'accès rapide des femmes aux services médicaux modernes. Cet article démontre que de tels systèmes de contrôle de passage ont un effet négatif sur la survie de l'enfant.  Les auteurs se sont servis d'une série d'interviews qualitatifs recueillis dans de district de Kassena-Nankasa pour étudier la construction sociale des systèmes de contrôle de passage fondés sur les concessions. Ces données révèlent que les chefs de concessions jouent les rôles de gardiens de concession pour des raisons spirituelles et que les maris jouent les mêmes rôles pour des raisons économiques.  Plus important encore, l'article présente les interventions médicales qui sont à l'essai actuellement à Navrongo (Ghana du nord) et comment elles ébranlent de tels systèmes de contrôle de passage. (Rev Afr Santé Reprod 2003; 7[1]: 17–26)

KEY WORDS: Gate-keeping, women's health, Ghana, access.

INTRODUCTION

The marvels of contemporary health technology have moved the frontiers of human longevity and well being beyond expectation, but the role of socioeconomic, cultural and behavioural factors in this transition process is still unfolding in various areas of the developing world.^1-4 In certain parts of rural sub-Saharan Africa, a vexing dilemma remains the low utilisation of modern health facilities even where these are accessible and affordable. In search for pathways of explanation, recent contributions of the social sciences research point to low women's education and autonomy, and backward traditional beliefs and taboos.^5-10 We have been served, however, with more of macro explanations and less of micro studies aimed at shedding light on these issues. This article elucidates compound level processes that impede the prompt seeking of modern health treatment by women in rural northern Ghana.

Among the Kassena-Nankana people of northern Ghana, constrained women's mobility and resulting delays in the modern treatment of illnesses have been identified as barriers to the transition towards better health especially for women and children.^11-16 These studies suggest strongly that delays are caused by the fact that populations try a series of traditional therapeutic options for their patients, and modern treatment is often a last resort. These patterns of health seeking behaviour compare with those found in other developing countries of Asia, Africa and Central America.¹⁷ Needless to say, the most affected are women and children because of their low bargaining power within their compounds, and the poignant question has always been what to do in terms of public policy. Solutions proposed so far are well summarised by Caldwell and Caldwell¹⁸: In much of the Third World the most rapid gains against mortality can be made by giving women, especially wives and mothers, greater confidence, more decision-making power and greater access to resources for care and treatment. Although this work subscribes to the view that education is the major route to women's empowerment, it presents alternatives that are on trial in Navrongo.^a

^aThe Navrongo area (or Kassena-Nankana district) has been designated by the Ghana Ministry of Health to field health and family planning research conducted by the Navrongo Health Research Centre (NHRC). NHRC was established in 1992 after the Ghana Vitamin A Trials (VAST) ended. Major research projects of NHRC include the Bednet Trial, the Navrongo Demographic Surveillance system (NDSS), the Geographic Information System (GIS), the Community Health and Family Planning Project (CHFP), the Panel Survey System (PSS), the Unmet Need Project, the Diffusion Initiative, the Malaria Attack Rate Survey (MARS), and the Adolescent Health Project.

This area fields the largest factorial experiment aimed at testing innovative ways of delivering health and family planning in a rural African setting. Door to door delivery of health and family planning services is being compared to other delivery schemes such as government fixed structures and mobilisation of village volunteers.¹⁵ This article demonstrates that doorstep delivery of health services is a promising way of breaking barriers that impede women's prompt seeking of modern health treatment.

The Navrongo Study Setting: Health Services and Compound Structure

The study was conducted in the villages of Yua, Chiana and Naga, which are located in the Kassena-Nankana District of the Upper East Region of Ghana. According to the Navrongo Demographic Surveillance System (NDSS)^b, the district population was 137,577 on July 1, 1996. Forty seven per cent of the district population is of the Kassena ethnic group, 51% are Nankana and about 2% are Buli. Unlike in other parts of Ghana where Christianity and Islam are massively practiced, most Kassena-Nankana people are affiliated to traditional religions; 85% of the population are of traditional religion.²¹

^bThe Navrongo Demographic Surveillance System (NDSS) is the NHRC's longitudinal system that follows up the population dynamics of the Kassena-Nankana district and routinely collects demographic events (births, deaths, and internal and external migrations). Since July 1, 1993, the NDSS has monitored, on a 90-day work cycle, the demographic dynamics in the Navrongo area. The NDSS has been adapted from the Matlab (Bangladesh) surveillance system using FoxPro language. Recent analyses of the NDSS data provide insights into the levels of fertility and mortality in the area under surveillance. For July 1, 1994, to June 30, 1995, the infant mortality rate (IMR) and probability of dying by age one (1q0) were 117.4 per 1000 and 117.6 per 1000 respectively. Life expectancy at birth was 52 years, 53 for females and 50 for males. The total fertility rate (TFR) was about 6.20. The NDSS is funded by the Rockefeller Foundation with technical support provided by the Population Council.

The district is primarily rural except the small town of Navrongo, the district administrative centre. During the rainy season, families cultivate mainly millet, rice, guinea corn and maize. Due to the existence of the Tono irrigation scheme and the numerous small scale dug-out dams, people engage in irrigation farming in the dry season. Women also collect and sell shea nuts. During the dry season, some men migrate to the southern parts of the country to work on plantations or seek other jobs.^c

^cThese movements outside the district create a deficit of men between the ages of 20 and 50, which is clearly reflected in the population pyramid of the Kassena-Nankana district¹²

The district health services include a district hospital, three health centres and 27 outreach stations run by the government, two private clinics operated by religious groups, several private chemists' shops, a number of traditional healers and traditional birth attendants, and several itinerant drug vendors. The only sources of health care in Yua are ambulatory drug vendors and traditional healers. In addition to these ambulatory drug vendors, Naga benefits from the services of a community health nurse who is resident in the village. Chiana is the most privileged of the three study areas, as one of the three health centres in the district is situated there.

The Kassena-Nankana district population settlement pattern is characterised by extended families living in dispersed compounds surrounded by farmland. Descent among Kassena-Nankana is patrilineal and families are organised based on extended relations. Usually, two or more nuclear families come together to form a compound that is headed by a male who by unanimous reckoning is said to be the most senior. The compound head is responsible for the social, religious, economic and political well being of all the people in the compounds. Women's autonomy is very low in the area covered by this research. When these women or their children are sick, authorisation is required from compound members before attending a modern health facility. The consequences on compounds' production of good health are substantial.

Gate-Keeping and Compounds' Production of Health

For a total of 2,856 women who were interviewed in the 1994 Kassena Nankana District Birth History Survey^d and who had at least one birth recorded by the Navrongo Demographic Surveillance System by the end of 1995, only 14.5% said they do not require authorisation from any man in their compound before attending a hospital, while 38.2% and 38.3% need authorisation from their husbands and compound heads respectively. Debpuur and Adongo¹³ reported that mortality varies across children born to these groups of women. The proportion of deceased children is higher for mothers who must face the veto of their husband or compound head as compared to those who require no authorisation or require permission from their mother-in-law. The proportion of children who died, among those who were born in the three-year period preceding the survey date, rises from 59 per 1000 for mothers who do not need authorisation from any man in their compound to 78 per 1000 and 123 per 1000 for those who require authorisation from their husband and compound head respectively.

^dThe 1994 Birth History Survey is part of the NHRC's Panel Survey System (PSS). Under the PSS, the NHRC maintains a panel of 1800 NDSS compounds in which a survey is taken on a yearly basis to monitor changes on a certain number of factors that are relevant to research projects run by the centre, especially to its Community Health and Family Planning Project. The PSS has been underway since 1993 when the baseline survey on fertility and family planning was conducted.

In this article, the term gate-keeper refers to individuals whose authorisation is required before women can attend a modern health facility. Compound gate-keeping systems thus characterise the nature of these constraints on women's prompt seeking of modern health treatment for themselves and their children.

Debpuur and Adongo¹³ have ascertained the above mortality differentials using multivariate statistical techniques. The present article does not intend to replicate such analysis. The focus here is rather on using qualitative data (focus groups and in-depth interviews) to understand the social construction of compounds' gate-keeping systems and how their negative effects on women's health seeking behaviour may be alleviated by appropriately designed health interventions.

DATA

Data were obtained from a series of focus group discussion sessions and in-depth interviews conducted in the Kassena-Nankana district (Upper East region, Ghana) between June and August 1995. The study sites have been described in the previous section. The Navrongo Demographic Surveillance System (NDSS) database, which follows up the dynamics of the area's population, was instrumental to the selection of subjects interviewed during the fieldwork. First, the three villages described earlier were chosen as study sites. Secondly, clusters, and subsequently individuals, were randomly selected in each village for the qualitative interview. The related selection procedures combine both random and convenient sampling techniques.

The type of health delivery system and variations in distance to modern health facilities were the criteria used for selecting the study sites. These are good indicators of health services accessibility, which is usually considered as an important determinant of health status,^22,23 although there is grounded scepticism about such effect.^24,25 In Naga and Chiana where health services are accessible, we hypothesised that prompt seeking of modern treatment may not be greatly affected by gate-keeping mechanisms within the compound. However, when health services are accessible, the way they are delivered may have an impact on the relationship between gate-keeping and health seeking behaviour. Chiana and Naga are appropriate study sites for testing such hypothesis because in Chiana, health services are provided in government fixed structures while in Naga there is a door to door delivery of health by community health nurses. In the third study site, Yua, where access to modern health services is poorest, gate-keeping may obstruct women's prompt seeking of modern health treatment for themselves and their children. Apart from accessibility to health infrastructures, we could not find any other meaningful community variables that might explain adding more villages to our study sites. All villages are located in rural areas and have close similarities with respect to culture, economic characteristics and financial costs of health services.

In each of the three villages, an NDSS cluster^e was randomly selected and 15 compounds randomly picked from the cluster. Then a random sample of subjects was done for focus group discussion sessions or in-depth interviews. Although the selection of subjects was done at random, the composition of the different groups is theory driven. We have proposed earlier that with respect to women's health seeking behaviour within compound gate-keeping mechanisms are a function of the balance of power between (a) married women and their husbands; (b) married women and their mothers-in-law; and (c) married women and their compound heads. The composition of the focus groups and the individual interviews reflects these different actors involved in making decisions on how, when and where to treat compound members.

^e NDSS clusters comprise an average of 65 compounds. The relatively low spatial dispersion of compounds within the same cluster allowed for a random selection of compounds to constitute our focus groups. This would not have been realistic if the selection were to be done without any prior stratification.

In total, 12 focus groups and 30 in-depth interviews were conducted during the fieldwork. In each village, focused discussions were held with four separate groups of compound heads, mothers-in-law, married men, and married women. In addition, in each of the three study sites, five in-depth interviews were conducted with married women 15–49 years old who were sick at the time of their compound visit, and five other married women were interviewed in-depth on one of their children's sickness. These group discussions and in-depth interviews investigated, among other issues, the cultural and economic rationale for compound gate-keeping systems and how such systems may be affected by the prevailing mode of health services delivery.

Cultural and Economic Rationale for Compound Gate-Keeping Systems

According to the qualitative field work carried out in Chiana, Naga and Yua, gate-keeping mechanisms that constrain women's prompt seeking of modern health treatment are a reflection of the spiritual role of the compound head and the prevailing economic power of husbands relative to their wives.

In the Kassena-Nankana society where traditional religion is massively practiced, the compound head is the sole mediator between the ancestors and compound members. The compound head is therefore the spiritual link between the dead and the living and is thus the pillar of the compound hierarchy. Any major decision should be discussed with him first; he will in turn consult the ancestors through the Baga^f before any further step is taken. It is this rigid mechanism that makes the compound head a legitimised gate-keeper regarding major decisions to be taken in the compound. As one compound head in Chiana explains:

“In our custom, the leadership of the compound has been given by the gods and he must always be the one to give permission before anything can be done.” (Compound head, Chiana)

^fThe Baga (also referred to by some researchers as 'soothsayer') is a traditional priest who provides guidance to compound members on their relationship with the ancestors. He is also a very trusted fortune teller. Baga is Nakam language while Vora is its equivalent in Kassem language.

The compound head therefore plays the role of a gate-keeper in a wide array of decisions especially those relating to health and reproduction, areas in which the ancestors' bearing is paramount. The compound head performs marriage rituals to seek the blessing of the ancestors. When a woman is pregnant, he consults the Baga to hint him on the personality of the expected new born. When a woman delivers, he also pours libation and provides millet to be used for the newborn's ritual bath.

When it comes to health related decisions, the role of the compound head as a gate-keeper is even more pronounced.²¹ This is mainly because of people's belief about the causes and treatment of illnesses. The Kassena-Nankana society believes that individuals fall sick mostly because of supernatural forces or because of disobeying the ancestors. Any attempt to treat the sick person must be preceded with a cautious consultation with the ancestors through the Baga, who will then inform on the illness type and whether or not such illness should be treated in a modern health facility or with a traditional healer, as well as the sacrifices required by the ancestors to help ease the healing process. In the FGD of married men in Naga, one participant explained:

“As you know, we are in a complex world. Despite the fact that we do not go to church, when someone is sick, we have to go to the Baga and come back and offer requested sacrifices before the person can be sent to the hospital if necessary.” (Married man, Naga)

There was a general consensus across all focus groups and individual in-depth interviews that the role of the compound head as a gate-keeper on all health related decisions should neither be vilified nor ignored for whatever reasons. Because of possible negative consequences, most participants indicated that it is not in anybody's interest to shake up the status quo. The best procedure when a compound member is sick is to inform the compound head first. The idea of seeking health treatment without asking for authorisation from the compound head is foreign to most interviewees. Doing so will lead to a host of severe sanctions from the gods, the compound members and the society at large. For women, the most cited consequences of seeking treatment for themselves and their children without the compound head's prior consultation with the ancestors are presented in the following focus group excerpts:

“Treatment will not work.” (Married man, Naga)

“If she does that (i.e., goes for treatment without asking for permission), it means that the man does not own the child. It shows that that woman has no respect for anybody and must be dismissed.” (Compound head, Yua)

“When a child is sick and the mother sends him/her to the hospital without permission, humm....... If the child survives, fine, but if s/he dies, they will say that the woman is a witch and had killed the child before sending the child to hospital.” (Married woman, Naga)

“If the child dies in the course of the treatment, where are you sending it to? You cannot send the child to the house.” (Married woman, Chiana)

“If she (the woman) goes for treatment without permission, should anything happen, that is left to her. If she dies, they will have to bury her there because we are not aware of her going there. So the corpse of such a woman will not be brought into this house.” (Married man, Yua)

“There are so many types of illnesses, some of which may come from the compound or which may be caused by the ancestors and thus cannot be treated in the hospital. These should be treated by the compound head by just pouring libations or offering sacrifices. If you bypass the compound head and later realise this, the compound head may refuse to perform his duties because you did not respect him.” (Married man, Chiana)

Participants at the various focus groups and individual in-depth interviews evoked other reasons for complying with compound gate-keeping systems, but those were not mentioned as unanimously as the ones above. From the previous excerpts, it is thus possible to identify the predominant reasons that legitimise the role of the compound head as a gate-keeper. These are mainly related to fears of certain sanctions from compound members such as divorce, accusation of witchcraft, and refusal to perform rituals for the woman. 

Beliefs about causes and treatment of illness create the necessity to ask for permission from the compound head before trying any medication. The Kassena-Nankana people, as confirmed by one FGD participant, believe that some illnesses should be treated in hospitals and others by offering sacrifices to the ancestors. The picture here is very similar to what Janzen²⁶ refers to as `disease of God' and `disease of man' in his award winning study of medical pluralism among the Bakongo of lower Zaire. With such beliefs about the causes of illness, seeking treatment without authorisation from the compound head is almost adventurous because the treatment given may not be appropriate and thus may not work. Concerns about treatment efficiency have been raised by a sizable number of interviewees.

Seeking treatment without authorisation from the compound head may be seen by compound members as a sign of disrespect to them and may lead to divorce. The hierarchy within the compound puts men on top of every decision, and women must not defy this situation. Women are socialised to abide by the rules governing the functioning of the compound and the society in general. Most interviewees explained that the control over married women's movement is justifiable because they live in compounds which own them and their children, because of the investments made during the marriage process.

As found in other African societies,²⁷ beliefs in witchcraft are widespread among the Kassena-Nankana people. In some instances, compound members may accuse one of their members of being the witch who has caused the death of another member. The evidence for such accusations may be based on simple suspicion. For a married woman, such accusation may lead to radical consequences such as sending her back to her father's compound, beating her, humiliating her in public, etc. Participants in the various qualitative discussions argued that when a woman sends her child to a hospital without asking for authorisation, she will be accused of witchcraft if the child dies there. The resulting punishments are severe enough to deter women from bypassing the compound head's authority.

In addition to the role of compound heads as gate-keepers, women in need of modern health care for themselves and their children have to face delays caused by their husbands. Unlike compound heads, who are gate-keepers for moral and spiritual reasons, husbands are gate-keepers for economic reasons. This has its roots in the meaning of marriage among the Kassena-Nankana. Marriage among the Kassena-Nankana is an affair between families and to some extent lineages, although would-be partners court each other for sometime.²¹ In the Kassena-Nankana society, marriage is viewed as an institution that confers on men the domestic and sexual rights over the women. Marriage is incomplete until the groom's family pays dowry to the bride's family. Because of the dowry system, women's ability to initiate decisions is restricted. Married women depend on their husbands for every basic necessity especially when it comes to money to buy food and clothing for themselves and their children and money to pay medical bills. The consequences on women's autonomy are clearly pointed out by a female interviewee in Chiana:

“There are so many things I cannot do without my husband's permission. For instance, like today, I wanted to go to the market, but my husband said that unless I finish harvesting the millet, he is not going to allow me to go. Also, there are times when I want to go to my farm, he will say no because he already has some work for me, so until I finish with his work, I cannot do my own. That apart, even if I want to cook food, it is the food he wants that I would cook, not what I want.” (Married woman, Chiana)

It is almost common sense that obliges women to ask for their husband's permission before attending a modern health facility. The treatment costs, including transportation and medical fees, are supported by the husband. Even in the case of traditional treatment, non-cash payments (usually a fowl) are provided by the husband. As shown in the following excerpts, both male and female participants in the various focus groups legitimised the role of husbands as gate-keepers.

“Of course, because the woman is married to me, I own her and have to see to her welfare; in that case, when she is sick, for instance, she does not go to her father's house to be treated. Everything will be my headache, so how can she go for treatment without my permission?” (Married man, Naga)

“It is a must because it is the man who came to your father's house to say that he wanted you and he brought you into his compound. So if your child is sick in the house and you do not ask for permission before sending the child for treatment, it means that you are responsible for the man or that you own the man.” (Married woman, Yua)

“As a woman, when you are sick you will not have the power to take any treatment or go anywhere for treatment unless you are permitted by the man. It is the man who will either give you money to go for treatment or find any other treatment for you.” (Married woman, Naga)

Health related decisions cannot be taken by women without consulting either their husband or their compound head. Because economic and spiritual resources are controlled by men, women do not have enough bargaining power when it comes to decisions that are economically or spiritually costly. It is always left with husbands and compound heads to decide whether or not the sickness of the woman or that of her child is serious enough to necessitate mobilising some resources to treat it. The main consequences of such gate-keeping systems are substantial delays in seeking modern treatment. There are, however, as shown in the next section, emerging health seeking behaviour patterns that point to the possibility that the negative effect of gate-keeping on women's and children's health may be levelled down by implementing adequate health delivery systems.

Gate-Keeping and Health Delivery Systems

Two important criteria that guided the selection of the study sites were their distance to the closest health facility and the type of prevailing health delivery system in the area. There is no evidence from the qualitative data collected in the three study sites that distance to health facilities affects the impact of compound gate-keeping systems on women's health seeking behaviour. Compound gate-keeping systems are as strong in Chiana where populations live at short distances to the health centre as in Yua which is very far from any health facility. Obstruction by compound heads and husbands to women's prompt seeking of modern treatment is also common in Naga. However, the interviews conducted in these study sites suggest important emerging transitions in the communities' social management of ill health.

As indicated earlier, Naga was selected because the health delivery system prevailing there departs notably from the usual bureaucratic machinery that provides health services in fixed structures.¹⁵ The Naga area has been assigned to a nurse who lives within the communities and visits all compounds on a three monthly basis. The nurse, also known as community health officer (CHO), is retrained frequently and supervised by the Navrongo Health Research Centre's Community Health and Family Planning Project (NHRC/CHFP) in close collaboration with the District Health Management Team (DHMT). She is provided with a motorbike and has the task of providing door to door health and family planning services. The CHO visits on average seven compounds a day, gives health and family planning talks, educates compound members on environmental sanitation and basic hygienic practices, treats minor ailments and refers serious cases to the Navrongo Hospital.^g

^gNaga thus resuscitates well known post alma ata primary health care (PHC) strategies aimed at providing accessible, affordable and sustainable health services to developing countries populations. The late 1990s seem to have witnessed a revival of field testing of such PHC strategies. The Navrongo Community Health and Family Planning Project has now matured from the pilot phase in Naga and two other villages to a full scale-up of the experiment over the Kassena-Nankana district using a four-cell factorial design. Another example is the Tanzania Essental Health Interventions Project,²⁸ which is the field testing of the conclusions of the famous World Bank's²⁹ report 'Investing in Health'.

As suggested by the qualitative data collected in Naga, this innovative health delivery system has induced major ideational changes especially when it comes to women's health seeking behaviour. When asked the first step to take when they or their children are sick, women in Chiana and Yua said unanimously that they would inform their compound heads and/or husbands who would then instruct on the actions to be taken. During the focus group discussion with married women in Naga, the presence of the CHO seems to have shaken up such gate-keeping system, as one focus group participant suggests:

“You know we have a health worker in our community, so when someone is sick, we go and call her and she will come and look at the person and she will tell what is wrong with him/her, especially if the illness is sudden and severe.” (Married woman, Naga)

The CHO has supplanted compound heads and husbands as the first resort to whom women turn when sickness occurs. The Naga communities trust the CHO and defer to her the authority to diagnose any sick compound member. This is so because the nurse is considered by local populations as one of their own. The relocation of the CHO from fixed health structures to Naga was carried out after a series of contacts with village elders, chiefs and local populations.¹⁴ Durbars were organised and the new health delivery systems presented and discussed in-depth. Focus group discussions were organised with villagers during which they made important contributions as to what can make the system work. The task of building a dwelling unit or community health compound (CHC) for the nurse was left to the Naga communities. The CHC is not a clinic but an ordinary compound where all villagers are welcome. When necessary, they may call on the CHO at any time, and as suggested by the previous FGD excerpts, the nurse always responds to such calls.

The presence of the nurse has vilified to some extent the impact of the Baga on communities' health seeking behaviour. In all three study villages, delays in the prompt seeking of modern treatment are mainly caused by preliminary consultations with Bagas. Bagas advise compound members on the type of illness and whether or not it is appropriate to attend a modern health facility. There are, however, persistent signs that these health seeking behaviour patterns are being reversed in Naga. The following excerpt from the focus group discussion with married women in this village is supportive of such dramatic changes that are still unthinkable in the two other study sites:

“When you are sick, you will not be able to tell what illness it is until you go to the nurse and she will tell you what it is. If she is not also able to tell what kind of illness you have, you will have to come back home so that they (compound members) go to the Baga to know the type of illness and treatment.” (Married woman, Naga FGD)

The emerging new patterns of health seeking behaviour in Naga suggest that the nurse is consulted first when a compound member falls sick. It is only when the nurse is not able to handle the case that consulting a Baga is considered as a reasonable alternative. There is no doubt that increasing reliance of the Naga community on their nurse will cut enormously on delays that previously affected the prompt seeking of modern treatment. This will obviously hasten the transition towards better health in the Naga community.    

CONCLUSION

The present research has unveiled important factors constraining women's health seeking behaviour among the Kassena-Nankana of northern Ghana. Focusing on the compound as a locus for health decision-making, we identified compound heads and husbands as gate-keepers who impede women's prompt access to modern health care. As a mediator between the ancestors and other compound members, compound heads control access to the supernatural world. This liaison role gives compound heads considerable gate-keeping power over the health seeking behaviour of other compound members, particularly women, because of the prevailing beliefs about the causes and treatment of diseases. In addition to their limited access to the supernatural world, women also face considerable economic constraints that further increase their dependence on their husbands and restrict their ability to make autonomous decisions.

Community participation in the project has been crucial to the success of the health and family planning services offered by the community health officer. Social mobilisation, also termed zurugelu in local parlance, was instrumental in organising durbars to introduce the nurse, and in building the community health compound. Consequently, the zurugelu dimension of the project strengthened the sense of community ownership of the project because of the active involvement of local populations in the re-organisation of the existing health delivery system. 

Evidence from this study suggests that the mode of health delivery may dramatically weaken the impact of gate-keeping on women's health seeking behaviour. In Naga, the community health officer has literally opened the door of the compound to reach out to women and deliver affordable health and family planning services. The qualitative interviews show that the CHO has begun to undermine compound heads' and husbands' control over women's prompt access to modern health care. The CHO is also supplanting the Baga as the first consultant in cases of ill health. More important, however, is the fact that these emerging behavioural changes demonstrate that appropriate health interventions may play an important role in improving health standards even in economically deprived areas.

ACKNOWLEDGMENTS

This research was funded by a grant from the Population Council's research division and carried out at the Navrongo Health Research Centre. We thank our dedicated field workers and transcribers. Special thanks go to James F. Phillips and Cynthia Lloyd for their support. An earlier version of this paper was presented at the University of Pennsylvania/Population Studies Centre's 1996 Colloquium Series and benefited from useful comments/suggestions made by students and faculty. Research assistance from Shasta Jones is gratefully acknowledged.

REFERENCES

1. Caldwell JC, S Findley, P Caldwell, MG Santow, WH Cosford, J Braid and D Broers-Freeman (Ed.). What we know about the health transition. Proceedings of an international workshop, Health Transition Centre, The Australian National University, Canberra, 1990.
2. Chen LC, A Kleinman and NC Ware. Advancing Health in Developing Countries: The Role of Social Research. New York: Auburn House, 1992.
3. Gribble JN and SH Preston (Eds.). The Epidemiological Transition: Policy and Planning Implications for Developing Countries.  Washington DC: National Academic Press, 1993.
4. Mosley WH and LC Chen (Ed.). Child survival: strategies for research, population and development review. Supplement 10, The Population Council, New York, 1984.
5. Bledsoe C. Differential care of children of previous unions within Mende households in Sierra Leone. In: JC Caldwell, S Findley, P Caldwell, MG Santow, WH Cosford, J Braid and D Broers-Freeman. (Ed.). What we know about the health transition. Proceedings of an international workshop, Health Transition Centre, The Australian National University, Canberra, 1990, 561–583.
6. Cantrelle P and T Locoh. Cultural and social factors related to health in West Africa. In: JC Caldwell, S Findley, P Caldwell, MG Santow, WH Cosford, J Braid and D Broers-Freeman (Ed.).  What we know about the health transition. Proceedings of an international workshop. Health Transition Centre, The Australian National University, Canberra, 1990, 251–274.
7. Castle SE. Intra-household differentials in women's status: household function and focus as determinants of children's illness management and care in rural Mali. Health Tran Rev 1993; 3(2): 137–57.
8. Doan RM and L Bisharat. Female autonomy and child nutritional status: the extended family residential unit in Amman, Jordan. Soc Sci Med 1990; 31(7): 783–789.
9. Dyson T and M Moore. On kinship structure, female autonomy and demographic behavior in India. Pop Dev Rev 1983; 9(1): 35–60.
10. Mason KO. The impact of women's position on demographic change during the course of development. In: N Frederici, KO Mason and S Sogner (Eds.). Women's Position and Demographic Change. Oxford: Clarendon Press, 1993, 19-42.
11. Dollimore N, H Odoi-Agyarko, and O Owusu-Agyei. A community based study of risk factors in maternal mortality in the Kassena-Nankana district of Northern Ghana. A report prepared for the Safe Motherhood Initiative, Geneva, World Health Organization, 1993.
12. Binka FN, GH Maude, M Gyapong, DA Ross and PG Smith. Risk factors for child mortality in Northern Ghana: a case-control study. Inter J Epidem 1995; 24(1): 124–135.
13. Debpuur C and P Adongo. Family structure, women's autonomy, and child survival in rural Ghana. Paper presented to Population Association of America Meetings, San Francisco, 1995.
14. Antwi-Nsiah C, JF Phillips, P Tapsoba, P Adongo and K Asobayire. Community reaction to the Navrongo experiment. Community Health and Family Planning Project, Navrongo Health Research Centre, Navrongo, Unpublished, 1995.
15. Nazzar A, PB Adongo, FN Binka, JF Phillips and C Debpuur.  Developing a culturally appropriate family planning program for the Navrongo experiment. Stud Fam Plann 1995; 26(6): 307–324.
16. Ngom P, P Akweongo, P Adongo, AA Bawah and F Binka. Maternal mortality among the Kassena-Nankana of Northern Ghana. Stud Fam Plann 1999; 30(2): 142–147.
17. Ware NC, NA Christakis and A Kleinman. An anthropological approach to social science research on the health transition. In:  Chen LC, A Kleinman and NC Ware (Eds.). Advancing Health in Developing Countries: The Role of Social Research. New York: Auburn House, 1992, 23–38.
18. Caldwell JC and P Caldwell. Roles of women, families, and communities in preventing illness and providing health services in developing countries. In: Gribble JN and SH Preston (Eds.). The Epidemiological Transition: Policy and Planning Implications for Developing Countries. Washington: National Academic Press, 1993, 252–271.
19. Indome F, BB MacLeod, JF Phillips, K Adazu, P Ngom and FN Binka. Computational procedures for the Navrongo Demographic Surveillance System Rates Program. Documentation Note 26, Navrongo Health Research Centre, Navrongo. Unpublished, 1994.
20. Ngom P, AA Awasana, BA Agula and S Owusu-Agyei. The Navrongo Demographic Surveillance System: July 1993 to June 1996. Documentation Note 31, Community Health and Family Planning Project, Navrongo Health Research Centre. Unpublished, 1996.
21. Fayorsey CK, P Adongo and B Kajihara. The Social organization in Kassena-Nankana district: assessing the contexts for reproductive change in a traditional African society.  Documentation Note 10, Community Health and Family Planning Project, Navrongo Health Research Centre. Unpublished, 1994.
22. Orubuloye E and JC Caldwell. The impact of public health services on mortality. A study of mortality differentials in a rural area of Nigeria. Pop Stud 1975; 29(2): 259–272.
23. Frankenberg E. Infant and childhood mortality in Indonesia: the impact of access to health facilities and other community characteristics on mortality risks. PhD dissertation in demography, University of Pennsylvania, Philadelphia. Unpublished, 1992.
24. Rosenzweig M and T Schultz. Child mortality and fertility in Colombia: individual and community effects. Health Policy Edu 1982; 2: 305–348.
25. Katende C. The effect of accessibility to clinics on infant and child mortality: the cases of Liberia and Zimbabwe. Commissioned paper for the working group on the effects of child survival and general health programs on mortality. Panel on the Population Dynamics of sub-Saharan Africa, Committee on Population. National Research Council, Washington DC. Unpublished, 1992.
26. Janzen JM. The Quest for Therapy: Medical Pluralism in Lower Zaire. Berkeley: University of California Press, 1978.
27. Evans-Pritchard. Witchcraft, Oracles and Magic among the Azande. Oxford: Oxford University Press, 1937.
28. International Development Research Centre. The Essential Health Interventions Project (EHIP): background paper.  International Development Research Centre, Ottawa, 1995.
29. World Bank. World Development Report 1993: Investing in Health. London: Oxford University Press, 1993.

Copyright 2003 - Women's Health and Action Research Centre