Exploring gender, health, and intersectionality in informal settlements in Freetown

ABSTRACT This paper applies an intersectional lens to health in informal urban settlements in Freetown, Sierra Leone. We explored how intersecting social characteristics including gender, age, wealth, occupation, and tenant status influence health and well-being outcomes. We found that hazardous environmental conditions, poor waste disposal, and waste burning contribute to health problems at a neighbourhood level. Health-care access was also generally poor in informal settlements. However, beyond this, there were differences in people's experiences of coping with health burdens and accessing care. Against a backdrop of limited state support, coping and access strategies were found to be heavily mediated by people's social positions and status, especially their ability to draw on support from social networks. There are particular challenges around the management of prolonged health problems. For population groups such as the old and the chronically ill, this creates further vulnerabilities including social isolation, stigma, and cycles of poverty. Although intersecting power dynamics apply to men too, women are particularly disadvantaged by coalescing social inequalities: they are both expected to perform caring roles but are less likely to be cared for. Young and old women were especially vulnerable and reliant on external support or self-sacrifice. This paper contributes to knowledge gaps in intersectional dynamics in urban settings and provides evidence that suggests policy shifts are needed to address the multiple social and health inequalities faced by women in informal settlements in Freetown, Sierra Leone.


Introduction
The world is now mostly urban, and nearly 1 billion people live in informal urban settlements (UNSTATS n.d.). In sub-Saharan Africa, 56 per cent of the urban population lives in 'slums' or informal settlements (UN-Habitat 2016). Problems of poor sanitation and infrastructure are plain to see, and evidence is mounting of poor health and well-being outcomes for the urban poor in low-and middle-income countries (Ezeh et al. 2017;Satterthwaite et al. 2019). However, few studies have unpicked the complex social and political determinants of health in informal settlementsas distinct from urban settings more generallyor how they differ across social groups and identities, and especially by gender. Failing to understand and address the way recognised determinants of health such as access to health care, water, and sanitation are mediated by social difference, will mean we are unable to achieve key Sustainable Development Goals (SDGs) such as good health and well-being, gender equality, reduced inequalities, and sustainable cities and communities. This paper applies an intersectional lens to explore this terrain, asking how social and structural inequalities linked to gender, spatial injustice, economic and social exclusion, and even faith, interact to shape health and well-being in two of Freetown's informal settlements.
Intersectionality as a concept was proposed by a feminist activist, Kimberlé Crenshaw, in the United States to draw attention to oppression of black women, specifically gender and racial exploitation (Crenshaw 1991). The concept is used in public health to describe how different social characteristics, including gender, ethnicity, class, or wealth, result in advantages or disadvantages, and shape inequalities in health and well-being outcomes. Its application to health systems research and policy in low-and middle-income countries has been limited (Larson et al. 2016;McCollum et al. 2019). However, it has been gaining traction recently in global health discourse. Recognising intersectionality allows for a multi-dimensional analysis (Hankivsky et al. 2012), including how social processes interrelate to shape health and social inequalities (Larson et al. 2016) for people with different social identities. A challenge is that it is not clear what the intersecting axes of oppression are in urban Sierra Leone, which is what this study aimed to explore. We begin by mapping out some of the social, spatial, political, and gendered terrain below.
Informal urban settlements: service provision and spatial inequality Rapid urban growth has outpaced the capacity of cities in Africa to provide adequate infrastructure for water, sanitation, and health-care services (UN-HABITAT 2016). Informal settlements are notable for their lack of recognition, meaning governments may not recognise their right to exist, and tenure is often insecure or opaque. As a result, the threat of eviction looms large and is a disincentive for comprehensive infrastructure development. Furthermore, some settlements are built on precarious and unsuitable land which makes providing services difficult and exposes residents to environmental hazards. As a consequence, large numbers of residents do not have reliable access to basic services, especially government-provided ones. Informal settlements tend to be where infrastructure and service access challenges are most acute, a result of both spatial and political drivers. Where government supply is lacking, private and informal providersof water, health care, sanitationfrequently fill gaps but on a fee-paying basis which has implications for access. It is also common for the limited services there are to be monopolised or captured by rent-seeking or exclusionary groups.
In Freetown, service provision within informal settlements is underpinned in part by historical inequalities in city planning and service provision (Lynch et al. 2020) which cause differences in service provision between formal and informal areas. Historical patterns of settlement have also created a hybrid and diverse tenure system, with early 'original' settlers having higher status and more secure tenure as landowners or landlords. Other residents, with access to government officials and an understanding of the legal system and the required resources, may also have managed to secure official ownership of land. People who own a house are respected and considered to be a privileged class, and are included in community decision-making, while tenants who tend to be newcomers, rural migrants, or the socially and economically disadvantaged are frequently excluded from such processes.
Our previous research into health in informal settlements in Freetown confirmed extremely limited health and basic services, and a high reliance on informal providers. There are significant access problems, due to cost and distance, which expose people to risks, especially women. For example, pregnant women reported being unable to reach delivery facilities and instead relied on traditional birth attendants (Macarthy et al. 2018). Toilets are mostly makeshift and access to the few public ones is based on payment of fees, resulting in widespread sanitation problems. People who cannot afford the cost often use nearby streams for open defecation, and women particularly face discomfort using these open spaces (Macarthy and Conteh 2018). Waste disposal infrastructurese.g. collection or designated dumpsitesare inadequate, which leads to indiscriminate and improper disposal, including burning or blocking drainage. Residents link these environmental conditions to high health burdens including malaria, typhoid, diarrhoea, and colds (Macarthy and Conteh 2018;Macarthy et al. 2018). While these studies looked at the quality of services, such as water and sanitation within informal settings, they did not explore the ways these affect people in different social positions. There is a need to explore how the conditions and constraints highlighted above are experienced along intersectional lines and relate to the interlocking challenges of urban informality and poor access to services.

Gender in Sierra Leone
Gender relationships in Sierra Leone are complex and changing. Much of the literature has focused on rural settings where traditional institutions of patriarchy lie at the heart of gender relations and are enforced by social and cultural norms. Culturally, gender norms have been regulated by initiation into village-based male and female sodalities, or 'secret societies', which govern social behaviours, confer power, and instil hierarchies not only between men and women, but also among them (Bledsoe 1984;Ferme 1994). As Sierra Leone urbanises, it is not clear the extent to which such institutions remain influential, but they are active in urban Freetown (Fanthorpe 2007). However, people who migrate from the rural areas are still sent home to villages for initiation and connection to their cultural roots through secret societies. It is not clear what effect the government's recent ban on secret societies will have following the death of a ten-year old girl during initiation rites in December 2018 (Batha and Payton 2019).
In rural settings, customary laws about ownership reinforce male dominance and power over women (Physicians for Human Rights 2002). Despite these power dynamics, a recent study in rural Sierra Leone by Cornish et al. (2019) suggests that women in profitable economic engagements are able to achieve a better balance of power with men. However, gender parity is far from being attained. For example, men maintain control over their wives in major household decisions, including health-care preferences and time of seeking care. This imbalance is exacerbated by the costly out-of-pocket payment which women incur when seeking care for themselves and their children (Cornish et al. 2019), despite the government's free health-care initiative for pregnant and lactating mothers and children below five years.
In urban Freetown, violence and discrimination against women and girls in schools and communities persist. A recent study, set in the congested eastern suburbs of Freetown, has highlighted how teenagers from poor families and not living with their own familycommon in urban Sierra Leonewere often vulnerable to transactional sex, including for food and clothes, for school fees, for grades (sex with teachers), and to lessen time spent on duties such as collecting water and petty trading. Combined with poor access to contraception, this results in increased chances of unwanted pregnancy (November and Sandall 2018). Furthermore, the study identified risks of maternal death among these girls as a result of delayed health care-seeking, stigma associated with their pregnancy, and abandonment by the non-parental adults with whom they live (November and Sandall 2018). This powerfully illustrates the relationship between intersecting vulnerabilities: living away from birth families, poverty, inadequate service provision, lack of protection, and gendered power dynamics; and how these combine to create poor health outcomes for young girls. Similar analysis is urgently needed to understand patterns of health vulnerability and inequality in informal urban settlements.

Research methods
Research took place in two informal settlements in Freetown between 2018 and 2019, using three research methods: life history interviews, governance diaries, and participant observation. The two settlements were Moyiba and CKG, 1 described below. The methods were chosen to enhance our understanding of the lived experiences of women and men. We used purposive sampling to identify male and female interviewees and groups with different social identities and health circumstances. Starting from these externally applied social categories (described for each below) our data analysis explored people's experiences to identify how additional categories and intersecting relational dynamics influence health. All names of participants were changed, to protect their identity. The study was approved by the Sierra Leone Ethics and Scientific Review Committee.

Governance diaries
This method involved repeated visits to the same household over a designated period. Despite being called a 'diary', data are collected in an interview. We selected eight households in each settlement, 16 in total across the two communities, and interviewed one member of the household every month for a period of six months. This household member was therefore engaged with six visits in total. The governance diaries method was developed to understand how people interact with governance actors, from the state and beyond (Loureiro et al. 2020), and especially to explore everyday governance from below' in places where the state is limited or where there is institutional complexity. As such, it was thought fitting to understand governance arrangements in informal settlements. Households and household representatives were selected by age and gender but also by other social categories, including female-headed households (associated with economic hardship), tenure status (landlord or tenant), ethnic minority (not coming from the dominant Temne group 2 ), and those suffering health problems. The interviews sought to elicit ways in which residents go about their daily life and what health and social challenges they face and how they dealt with them, including who they trusted and relied on for help. Interviews were recorded and researchers took detailed field notes.

Narrative interviews/life histories
We conducted one-off life history interviews with 30 participants in CKG and Moyiba (15 for each community). The interviews probed important health events experienced by participants and their household members, and the impact these events had on their lives. Again, participants were selected by gender and within that also by high-and low-income groups, age, marital circumstances, tenure status, and included some participants suffering from long-term or chronic health problems.

Participant observation
During visits to the communities, researchers carried out participant observation and took detailed notes. The notes recorded general observations about community infrastructures, sanitation, access to services, and exposure to environmental risks, and noted characteristics and interactions in and between households and communities. The observations provided an opportunity to observe everyday life, establish rapport with residents, and to contextualise the other forms of data collection above.

Study communities
Although the settlements of CKG and Moyiba are both informal and face challenges around infrastructure, they are very different socially and topographically. CKG is a coastal community in the westi.e. the central partof Freetown. It lies close to Kingtom, which accommodates one of the two largest dumpsites in Freetown. Waste is deposited at the Kingtom dumpsite from the rest of the city, and frequent burning emits smoke, causing air pollution. Due to its proximity to the city centre and major services, the area has become densely populated over time. Livelihood activities of residents include fishing, small-scale businesses, and scavenging from the Kingtom dumpsite. Like many other informal settlements in Freetown, the settlement's residents experience challenges with access to water and sanitation services and have limited public provision of water (YMCA et al. 2016). The community faces acute shortages of water in the dry season. The main source of drinking water is a makeshift tap connected to an under-bridge pipe, run by youths. Residents collecting water from the tap pay small fees to gang members who control the service. Other sources of water include unprotected private wells where residents also pay fees to their owners to access. There is only one governmentrun health centre serving the three adjacent CKG communities. Access to toilets is also a challenge for most residents. Open defecation in nearby streams and the dumpsite is common, particularly for people who cannot afford fees to access the only public toilet within the community. For people who have access to toilets within their homes, hanging toilets (makeshift toilets) which empty human waste into the nearby environment, 3 are the most common type in use.
Moyiba is a hillside community in the far east of Freetown and is less densely populated than CKG. Its main livelihood activities include stone mining, petty trading, and commercial motorbike riding known locally as Okada. 4 Income levels are generally low, with the majority of residents deriving their livelihoods from stone quarrying. The settlement is large and extends far up into surrounding hills. Lower parts of the settlements have some services, such as a health clinic and water, and are more densely populated, whereas further up the hills the settlement is characterised by a lack of health facilities, lack of water sources and sanitation services, sparser population and housing, poor road networks, and extreme poverty. Large rocks and poor drainage systems cause erosion and flooding during heavy rain, contaminating the downstream rivers which are used by residents for bathing, laundry, and cooking (Macarthy et al. 2018). Like CKG, Moyiba faces severe water shortages in the dry season. The community has many wells, but most are unprotected. Taps were installed in a few areas but this water is not regularly treated. The community has no designated areas for waste disposal so many residents dispose of their household and market wastes in the streams, drainages, and in the streets, which causes flies and mosquitoes to hover around the decomposing rubbish.

Findings
We begin by describing the common health problems in each settlement, as experienced and shared by our participants. We then describe health-care access and the significant challenges around affordability. In providing these descriptions we characterise some settlement-wide patterns which result from the lack of reliable formal health and basic service provision, and then we examine the differential implications for men and women and for different social groups. In particular, we try to understand how people according to their gender and other characteristicsare more or less able to draw on different sources of support to sustain their health, while others continue to experience heavier health burdens.
Overview of health hazards and burdens: 'Everything here is not decent' Health conditions in the two informal settlements are, by most standards, dire. People struggle with health problems related to hazardous environments, poor sanitation, and occupational risks. Among all population groups, malaria, fever, and 'pain' are frequent complaints. Malaria is referred to as a 'normal' or 'simple' sickness. Residents linked their health conditions to environmental factors such as waste accumulation which causes mosquito breeding and provides favourable habitat for other disease vectors such as rodents. Waste accumulation in drainage systems also causes frequent flooding. An old widowed womanliving not far from a particularly run-down strip of corrugated iron houses right next to the dumpsite known locally as 'Gaza'explains how the community has experienced annual flooding, a breeding ground for mosquitoes and malaria: I am seriously concerned that this year's flooding will get worse than the previous years if something is not done about the main gutter that has been blocked with huge piles of garbage. The gutter often gets blocked because we do not have a designated place to throw our waste, hence the reason why we dispose of our waste into the sea and streams which breeds a lot of mosquitoes and malaria for the community.  Irritation and respiratory problems as a result of air pollution, especially the emission of smoke from regular and uncontrolled waste burning, is common. Occupational health hazards are also common, e.g. related to the dust from stone mining in Moyiba. One man commented casually that 'In Freetown, the sickness is just in the air', revealing both awareness and resignation about the pervasive nature of health hazards facing residents. A young woman summed up living in CKG as 'This area is really not easy to live because everything here is not decent' (Gov-CKG-001-2).
In addition to the commonly reported communicable diseases, and the effects of air pollution, many residents also reported suffering from chronic conditions such as stress, ulcers, and hypertension (known simply as 'pressure'), and a few had suffered 'stroke', leaving them with lasting disability and ill health. Many of these conditions were in older populations but not always, e.g. one woman described how her son was 'sick with stroke' and had been taken back to their natal village for care. A feature of many of these long-term conditions is that the diagnosis was elusive or vague. People had tried numerous health providers and were still unable to get good answers, or they had given up on paying for consultations and tests. In CKG, for example, a 34-yearold man told us he had been suffering from suspected Parkinson's disease for ten years, but he also reported this was suspected to be the result of exposure to chemicals while working in a photo studio without being provided with protective gear. Like for many others, there was little clarity about what his condition was or what caused it: Doctors were unable to make any clear diagnosis when my friends took me to Mabesseneh hospital in Lunsar in the north of the country. I was referred to another doctor in Freetown where I was examined by seven doctors and I was told I had tremor. I was later told by the same doctor that I had Parkinson's disease. He related the disease to the chemical I was exposed to in a photo lab without using protective gears.  Our samples are too small and data are not intended to make assertions about the relative burdens of health between men and women, or sub-groups within them, but they do paint a picture of general low standards of health and of widespread hazards which affect community members quite universally, e.g. poor sanitation, waste, and air pollution ).
Health-care access: 'Money is the help I am looking for' The limitations of health carein terms of cost and availabilityare also stark across the two settlements where most people are poor and there are no state safety nets. In our study settings, health-care services are accessed from multiple private and informal providers. Informal health-care services are widely utilised in part because of wide availability as they are often based in settlements or travel to them, whereas there may be no formal health-care clinics located in or nearby settlements. Cost and method of payment is also a factor, as informal providers often allow deferred payment, or payment in instalments. Faced with the need to seek care for multiple and complex health problems, some people establish customer relationships with drug peddlers who credit them medicines daily. This is the experience of an old woman at Moyiba who experiences multiple health problems, including chronic cold, ulcer, and a past history of tuberculosis: I buy drugs from a man whom I call 'doctor'. The man helps me with medicines at reasonable cost, so I wait for him whenever I have money. I have established so much trust with the doctor that he gives me medicines on credit, and I do payments usually in two instalments. (Gov-MYB-007) Health-care affordability is a challenge that residents grapple with daily, and which contributes to the diverse health-seeking strategies, including use of informal care services and rationing of treatments. In CKG, an elderly man described how he employs the strategy of buying half doses of drugs for malaria treatment from a nearby pharmacy, to selfadminister: I buy drugs sometimes at Le 10,000 5 for half of a dose; I receive three tablets if I can afford to pay Le 5,000 at the time of illness. The full malaria dose containing ten tablets costs between Le 15,000 and 20,000 whereas the half dose costing Le 10,000 only gives five tablets. I am sometimes given discounts by the pharmacy attendants if my condition is severe in order to help me administer the full dose, but if my condition is not severe, I take the first half dose and complete the second half when I get money. I discontinue the second half dose once my health improves. (Gov-CKG-006) Across the board it was clear that the cost of health carebe it from public or private sourceswas a major challenge, leaving people reliant on help or, when it was not forthcoming, facing terrible predicaments. One widow explained how when her husband was dying, she was unable to afford care for him and had turned to his family for support, but they had decided that it was too costly: I thought they were going to take him to the hospital because that would have been the best place to take him, and I believe the situation would have been better, but they only took him to their house and lay him [out] without taking him to the hospital and they were saying that the health worker had said that the pressure had gone up his brain and that everything was mixed up within his system so there was no need to go and waste money in the hospital. (ESRC-MYB-012) Another woman whose husband had been injured in a traffic accident which had left him unable to walk, put it succinctly: 'Money is the help I am looking for.' She elaborated: There is no money and when you don't have money it's difficult to do something. You see? If we had money to see a good doctor, he should be able to walk by now. He should have been able to walk long ago, but it's due to the lack of money. (ESRC-CKG-006) Women's experiences of health, health care, and coping Although money was a dominant feature of most people's health (and non-health) experiences, the other recurring theme was whether people had friends, family, or others; 'helpers' who could assist them when they were in need, e.g. by paying for hospital fees or sending small sums of money to keep their families fed and housed. In a context where very little is free, access to support and additional money could be critical to whether people could achieve basic standards of health and well-being. Exploring men and women's experiences with coping with heath burdens and accessing support is where important differences could be discerned. These differences also show that economic and marital status, household composition, and social ties intersect with each other and other characteristics in complex and changeable ways to affect health.
Across our interviews, women were more likely than men to report having no one to help them, in addition to being more likely to be caring for others, e.g. for parents, the children of relatives, as well as their own husbands (especially if sick or unemployed) and children. This finding was revealing of two things: first, that there are strong societal expectations for women to take on caregiving roles; and second, that beyond their own husbands, who are expected to act as heads of family and provide for women, there is little guarantee of support from elsewhere. The following sections describe the toll this takes on women's own health and well-being, and then we show how in coping with these pressures, women have different degrees of security according to social positions and resources.
Although in Sierra Leone it is conventional for men to be head of the household, many women were looking after their husbands who were unemployed or sick. For example, Mariama, a 28-year-old mother of six children whose husband had lost his job at a phone company, became the sole provider for the family (Gov-MYB-004). She was struggling to survive by selling onions and condiments but her own health had recently declined, reportedly from low blood pressure and typhoid, which left her less able to work. A brother had sent her some money, but he had stopped when he lost his job. She was rationing her own health care, waiting to buy cheap drugs from informal providers who would visit her home, while seeking health care from charities for her children. While school was on a break her eldest daughter was assisting a friend, who had a stall selling cooked food or 'cookery', and would bring the family leftovers to eat. Mariama worried that when the schools re-opened, they would not have enough food to eat.
The story of another woman, Kadie, illustrates the strength of women's caregiving obligations, and the way they extend far beyond the immediate household unit and urbanrural boundaries. Kadie had lost two husbands, leaving her with four children. She was also responsible for her brother's children after their mother died of Ebola and her brother died soon after: 'when this happened, I had no option but to take the kids with me'. When her remaining brother back in their village had fallen sick himself, her last bit of support dried up, and Kadie now sent food and assistance to her brother's wife as she was grateful to her for sticking by her brother and in recognition that she was also heavily burdened: Well I am just trying on my own to upkeep the family because I have no way to do and I have all the children with me here and I also sometimes send something for them in the village because I really appreciate the wife a lot because since my brother fell ill, she has been staying with him, and did not abandon him because of his health condition. She is the one taking care of him presently, and because of this she cannot do anything else.  We now turn to the ways women cope with these pressures and what sources of support they can rely on. With no formal safety net, family networks are critical in times of need. Those without family find themselves in precarious positions. Support was expected most from one's own birth family, but one woman described how she was lacking such connections: I don't have anyone to support/take care of me right now. Our dad has passed away; he was killed during the rebel war and he [had not] shown us to the other family when he was alive. We were seven in total, but only four of us are alive right now and we don't have the upper hand to support one another financially … and we don't even know our hometown.  Married women taking care of sick husbands repeatedly claimed that help from their husband's family was not forthcoming. Bintu, a woman in her mid-40s and living in CKG, explains her daily struggles to care for her ailing husband: I have nobody to help me; even his family members have turned their backs on us, and they are saying that they are waiting for the man to die so that they will come and take his house from us.  This was also the case for women whose husbands had died, leaving them widowed and with dependants. Aminata described how her husband had been sick for a long time and had finally died and 'left me with the children without any helper'. She elaborated that not only did she receive no support from his family, she felt it was unlikely she would find another husband to care for her: Two of us were living together here but he died and left me with my family and my children and none of his family members is helping me; I am the only one fighting up and down for what to eat with my children … and up to now no man has come my way to marry me, because these days when men know that you have many children without a father they will not come your way for fear of the responsibility. (ESRC-MYB-012) Women's stories revealed that marriage, although a source of security if their husbands are able to earn money, was by no means a guarantee of safety and well-being, and could lead to additional burdens. This is especially true for women who were married young to older men at the wishes of their parents, which still happens where marriage of daughters is used to make alliances. Returning to Bintu, she describes how she had little choice in her marriage and must now live with the consequences which enhance her vulnerability and clearly affected her well-being. In her case, the lack of education, her position as a younger child, and obscure choices on the part of her mother combined to match her with an old man whom she now had to care for, but, as outlined above, with no support or security from his family: I was the second to last to my mother so she did not send me to school and she just decided to make me marry to this man … maybe the man was kind to her, the extent to which I don't know, because I don't see any reason why she decided to let me marry to an old man.  Beyond parental control, marriage is pursued by women themselves as a means of escaping economic hardships and homelessness. Yet, marriage is not always an escape from economic hardships and for some it increases vulnerability in other ways, as in the case of one woman whose husband was imprisoned. She described how she went into sex work to support her family while her partner served his prison sentence, leading to her contracting HIV: I ran away from my village and came here in Freetown to stay on my own. I was in the street with no definite place to call home when a guy came my way. He took me to his place to live with him. He committed a crime and was jailed. I was now all by myself with the two kids I have with him until some people took me in and my children. For now, I don't do anything, so I have to go to the street for my children and I to survive … I am a hustler; a sex worker, hustling at night to sustain a living for myself and my two children.  For the young woman quoted above, HIV treatment is a challenge because she does not often have enough food to take her antiretroviral drugs, and she sees this as a choice between dying of the disease instead of dying of hunger and complications from the drugs.
As mentioned in an earlier section, secure housing is a major difference influencing women's experiences of health and coping with health burdens. As in Bintu's story, for many women homes are secured through their husbands, but this can be taken away from them when their husbands die. In other instances, women had secure housing through their husband's job, but when either their husband or the job was lost, they were thrown out to fend for themselves. There were some women who were able to secure land and livelihoods which left them in an advantageous position even when their husbands had passed away, but these examples were rare. At the other end of the spectrum are women who do not own land and are relative newcomers to their communities, hence they have little personal security, and limited say in community affairs. This was the case for Adama (Gov-MYB-002) who had come to live in Moiyba with her husband a few years earlier, but he had died leaving her on her own, making a living from stone breaking to look after their three children and her elderly father. Adama described how she was constantly being asked for contributions to community development from the 'big ones', 6 apparently for upgrading the settlements, e.g. water and drainage, or for acting on their behalf in negotiations with stone cartels. She had little trust in these leaders, mostly men, and noted that these community upgrades rarely materialised and that the youth and women like her had little say in the decisions. As a result, she was refusing to pay, which was bringing her into conflict with the 'big ones'. This illustrates both patriarchal and age-based leadership which excludes young women and especially those without family ties to the area.
Finally, an elderly woman in CKG exemplifies several of the intersecting forms of powerlessness and insecurity mentioned above. Jeneba (Gov-CKG-004) had lived with her husband in central Freetown but when he died, five years previously, she had had to move in with her sister in CKG, where they shared a corrugated iron room in the rundown area of 'Gaza'. As well as being widowed, she had lost her only child, leaving her without anyone who would look after her reliably. Her health was poor and at one point a nephew was helping her with health-care costs, but he had stopped coming to visit and no longer answered her calls. Jeneba was depressed and declared, 'I am tired with life. I am tired of living.' The toilet in her compound had broken during the flooding the year before and had not been fixed; she had given up asking the landlords to fix it. She found that the smoke from the burning of waste near her home exacerbated her health problems but she had not reported the matter to anyone because she was resigned to the fact that nothing would be done: 'I have not reported the issue to anyone because as you know, a chicken has no power in the land of the cockroach.' As with Adama, she felt that the 'big ones' and chiefs did not act for the benefit of the community.
Men's experiences of health, health care, and coping Throughout this study, women and men suffered similar health problems but support systems were more likely to favour men than women. However, it is also clear that with employment hard to come by and some men having to live with chronic health conditions, women were the main source of support to them and other members of the household experiencing these health circumstances. This highlights that in Sierra Leone, women are culturally expected to take on a range of different caregiving roles, including for their children, husbands, their own parents and relatives, and sometimes the parents of their husbands. Amidst these difficulties, it was common for matrimonial conflicts to erupt between men and their wives who were providing social and material support for them. This section describes men's experience of health and care, and how this intersects with gendered power dynamics. In particular, it shows how conflicts occur between men and women as a result of shifts in power dynamics and how this affects the ability of men to cope with health problems.
In many instances, men were embarrassed by being looked after by their wives when they fell sick. Amadu, the young man in CKG suffering from a condition suspected to be Parkinson's disease, cannot walk without aid, and is unable to engage in any paid work. His wife has assumed most of the financial responsibilities, including buying him drugs, mostly over the counter from pharmacies, payment of house rent, school expenses for their children, purchase of water, and cost of transporting it home from where she sells 'cookery' at Murray town in the west end of the city. Amadu feels embarrassed that his wife is performing what should be his role as a man, and saddened that his siblings are failing to meet their promises of family support to him: I am embarrassed that my wife is the one taking all the financial burden in the home, including caring for the children, paying the house rent, and taking care of my medical bills. This is more of a concern to me because my own family doesn't care about me. Sometimes, my wife grumbles about providing certain things for me, so I don't ask her for certain things even when in dire need.  Loss of independence was construed by some men as loss of hope, which impacts on their capacity to cope with ill health. For Pa Sorie, an old man at CKG who has been incapacitated by a road accident, his daily living depends on the little earnings of his wife from iterant business, and he has not been able to seek proper medical care due to lack of money. While his wife tries very hard to look after him, for him being unable to look for his own money to meet his medical expenses brings alive feelings of hopelessness: Well, there is no hope for me, because if someone is able to go out, he can be able to secure food to eat. But since I don't go anywhere now, things will be very difficult for me if my wife doesn't go out to secure food to eat … Looking for your own money and living on charity are not the same.  However, having a secure home is a source of security for men, particularly those living with prolonged health problems. In the case of Pa Sorie, his current financial situation would have been worse if not for the compound he had built which he rents out to supplement his wife's income. He reflected on his relatively privileged position: 'If I hadn't built this compound, perhaps I will be in the village by now', meaning he had avoided the fate of being sent back to his natal village, a common occurrence when other care options are exhausted. Without a home, ill health and lack of money can become a downward trend for men.
Finally, while men with health conditions often had experiences of their wives taking care of them, a disabled man in his early 40s in CKG had dealt with it differently by deciding not to live with a woman. He feared that in addition to dealing with stress from his ill health and disability, he would have to contend with potential conflict from being unable to care for a woman financially. He takes care of his two children remotely, but with significant self-sacrifice: Against this backdrop of complex ailments and limited care, faith was an important source of support and hope for many people, and another meaningful social identity. The two major religions in Sierra Leone are Islam and Christianity, and many people reported turning to prayers and their religious community for help, especially when they found that their health problems were apparently beyond human and medical control. While in some contexts religious identity is fixed, or even divisive, one surprising finding was that in this part of Sierra Leone religious identity was often very fluid. A strategy employed by both men and women was to be flexible, or to be a 'Chrismus', the local term for people who practise a combination of Muslim and Christian religion. For example, an elderly woman at CKG explains how she has been switching between being a Muslim and a Christian to secure a solution to her health problem: In relation to my health, what I have held strongly is the church. I am a Muslim, I can't lie, but I have held on to Christianity very strongly for my health apart from taking medication, because prayers in church are helping me a lot. I have my anointing water and oil that I drink and bathe with, so I really thank God. Therefore, for now I put God first.  Another man (Gov-CKG-008) described how although he was a Muslim he had taken to going to a local church for deliverance, hoping his health problem will be healed by faith. For most of these people, reconciling being Muslim and Christian was easy and seemed to cause few problems. This was not always the case. One woman (Gov-CKG-005) whose husband had died, had found a new place to live through her mosque. She now lived in a compound owned by a local chief who was a Muslim. Having fallen on hard times, she had recently begun going to church too, which she credited for improving her health and helping with other misfortunes. However, her landlord had now asked her to leave which she suspected was because of her church-going. While religious flexibility can be a source of increased support, especially useful in navigating urban Freetown's numerous challenges, it can also have drawbacks, especially if one is in insecure housing or work.

Discussion and conclusions
In this paper, we have explored how gender and other intersecting factors and relationships shape health vulnerabilities for men and women in urban informal settlements. Urban areas are experiencing enormous changes, including high levels of inequality, underpinned by complex social, economic, and power relations (Grant 2010). This requires new ways of understanding social inequalities and how they impact on the health and well-being of the urban poor.
First, our findings describe that there are some common health problems related to localised and spatial inequalities, in particular the differences in facilities and service provision between formal and informal parts of the city. Residents of informal settlements face hazardous environmental conditions such as inadequate sanitation and waste disposal systems which lead to health problems, especially infectious and respiratory diseases. This is a direct cause of inequality in health burdens between richer planned areas and poorer unplanned areas within the city.
Second, the limitations of formal health-care provisions are evidently stark and, when compounded by high levels of poverty affecting affordability, cause different health-seeking preferences, with socially vulnerable groups often forced to pursue riskier options and strategies. Access to health care is influenced by a combination of personal and individual factors, such as gender, marital status, and household composition, and social factors, such as poverty and housing status. While most of those categoriesage, gender, and wealthwere purposefully used in the research (e.g. to select respondents), a key finding was that additional and more complex relational characteristics were evidently influential to health and well-being. These categories mediated access to care, support, and protection, and included tenancy duration, widowhood, and religious faith. Thus, aspects of identity intersected with more structural categories to have an impact on how people access services and sustain health and well-being. Notably, we found that belonging to, or ability to draw upon, support networks is pivotal in securing health in urban Sierra Leone.
Third, we found gender differences in how men and women were impacted by health challenges, especially around expectations of and availability of support. We found that conventions about caregiving increased women's burdens as women were expected to provide care for their husbands and family, often under difficult economic circumstances. Another common social expectation was that marriage was a route to security, but we found that in practice, in a patriarchal society such as Sierra Leone, this could also lead to additional burdens for women. Yet, being aloneand without supportwas also a major factor in hardship. When combined, these underlying social conditions can create acute stress and vulnerability. Some people face multiple intersecting disadvantages and negative health impacts, e.g. widowed or childless older women have little familial support and find themselves isolated, unable to deal with chronic health problems and sometimes in insecure housing. Men suffering from chronic conditions and cared for by their wives felt unease and shame at the unsettled gender norms and undermining of their masculinity. Resulting shifts in power dynamics allowing women to take major household responsibilities often lead to matrimonial conflicts.
We conclude that intersectional analysis of health is important for addressing complex health inequalities and understanding the holistic nature of health in marginalised urban settings. This study has contributed to filling knowledge gaps on how gender inequalities combine with other social, structural, and personal factors to produce health and social vulnerabilities. Policies aimed at addressing health inequalities must recognise and target these inter-relationships in order to achieve their desired outcomes.
Notes 1. CKG is the abbreviation used to describe adjacent coastal informal settlements known as Crabtown, Kollehtown, and Greybush. They are often referred to as one because of their proximity and similar social and environmental characteristics. 2. For a breakdown of the Sierra Leone population by ethnic group, see Minority Rights (n.d.) .