Understanding self‐construction of health among the slum dwellers of India: a culture‐centred approach

Abstract

Disembarking from a traditional approach of narrow hazardous environmental and structural conditions in understanding urban slums’ health problems and moving towards a new notion of what constitutes health for slum dwellers will open a new avenue to recognise whether and how health is being prioritised in disadvantaged settings. Drawing on in‐depth semi‐structured interviews with a total of 67 men and 68 women from Kolkata slums and 62 men and 48 women from Bangalore slums, this study explored how knowledge, social realities, material and symbolic drivers of a place interweave in shaping slum‐dwellers’ patterned way of understanding health, and the ways health and illnesses are managed. The current study adds to the growing evidence that ordinary members of the urban slums can articulate critical linkages between their everyday sociocultural realities and health conditions, which can support the design and delivery of interventions to promote wellbeing. The concept of health is not confined to an abstract idea but manifested in slum‐dwellers’ sporadic practices of preventive and curative care as well as everyday living arrangements, where a complex arrangement of physical, psychological, financial, sociocultural and environmental dimensions condition their body and wellbeing.

Introduction

Although health conditions and illness play a pivotal role in poor health outcomes of urban slums, knowledge is still very scant about the subjective dimension of health, which is inextricably linked to poor health occurrence (Garcia 2006, Marshall and McKeon1996). Health is shaped not only by biological risk factors but is also conditioned by the human‐built external environment such as social learning, cultural norms, ethnic traditions and interactions in which it takes place (Blaxter 1990, Bury 1997). The emphasis, therefore, should be on phenomena that build up through dealings in a social milieu rather than on meanings of phenomena that do not essentially exist in the phenomena themselves (Airhihenbuwa 1995, Berger and Luckmann 1991). For instance, Dutta‐Bergman (2004), in their study, showed that Santal tribes of West Bengal located healthy life not in illness but primarily in the realm of their access to food. Sen (2002) observes that spite of diverse medical facilities and favourable life expectancy rates, people in the US tend to complain more about their health‐related problems, while health does not necessarily constitute a major concern for people in the Indian context in general. Again, studies by Anderson (1976) and Shanas et al. (1968) show that the elderly usually have a propensity to perceive their health as good, even in the presence of clear pathological symptoms, and attribute their discomforts to the normal ageing process. It is therefore evident that there are conceptual differences in the way different groups think about health, given the multiplicity of the sociocultural landscape involved (Bott 1971, Kelman 1975, Twaddle 1969, Zola 1966).

Because of the wide ethnic and cultural diversity that exists in India, the health practitioners are constantly called onto interpret how the cultural context influences health meanings, how they are constructed and practised (Airhihenbuwa 1995, Dutta 2008, Lupton 1994).

More specifically, slum settlements as unique areas entail ‘intimately shared physical and social environments’ for their inhabitants (Lilford et al. 2017). Living in slum areas is characterised by insufficient access to water and sanitation facilities, inadequate housing space, job and housing insecurity (UN‐Habitat 2003). In addition to that, slum settlements have a spatial dimension that entails strong neighbourhood effects (Oakes 2004), where high social and spatial proximity to others affect the residents’ health outcomes (Ezeh et al. 2017). For example, overcrowding leads to higher competition for scarce resources and hence to crime, but also epidemics due to collection of garbage and excrements in concentrated areas, or catastrophic effects of extreme weather conditions (Ezeh et al. 2017, Meijer et al. 2012). Post‐migration further leads to problems of acculturation, cultural shock and bereavement, the discrepancy between expectation and achievement, loose social support, lack of group solidarity, lack of access to appropriate resources, low perceptions of safety, financial precariousness and the like. Altogether, it creates significant biological, social and psychological dysfunction and disorganisation among the residents and their immediate environment.

This neighbourhood‐built environment coalesces to constitute the central axis to public health risks, shapes health behaviours, beliefs and even disease (Dalgard and Tambs 1997, Ellaway et al. 2001, Wen et al. 2006, Wight et al.2013). Because of these peculiar circumstances, slum health needs warrant scholarly attention, as interventions that can improve health conditions in non‐slum areas might not be transferable to slum settings (Ezeh et al. 2017).

As a coping mechanism to the new form of social desirability, the slum migrants accept the social structure (rules and resources) encircling them, while constantly modifying some of this urban context with their practices (Giddens 1984). Hence, the co‐existence of traditional collectivism with emerging individualism in the same neighbourhood (Pedersen and Rytter 2018) with different health effects (Conde et al.2018) can be witnessed. This calls for research aimed to gain a proper understanding of the health perceptions harboured by the slum residents which in turn is heavily dependent upon their new sociocultural backing, demographic and economic changes that form the crux of the self‐perceived health (Castañeda et al. 2015, Ghosh and Shah2004, Neto et al. 2017).

We respond to the call towards the understudied topic of slum health by introducing a culture‐centred approach. This is because, to effectively explore the utilisation gap in the healthcare system, unlike the predominant top‐down approach; the culture‐centred approach treats communication as a gateway to participatory avenues. It emphasises on cultural members’ articulation about their shared meaning of health experiences derived from socially constructed identities, relationships and norms (Airhihenbuwa 1995, Dutta 2008, Dutta‐Bergman 2005). This approach is constructed upon culture, structure and agency as a means to explain social behaviour. Structure refers to the arrangements of institutional and organisational interactions that affect the availability of resources (Dutta and Basu 2007). Agency, on the other hand, constitutes the thoughts and actions of the subjects to either reproduce certain culture‐derived behaviour or challenge the status quo to address the emerging demand of real‐time situations (Emirbayer and Mische 1998). Culture‐centred approach expostulates that both structure and agency are deeply rooted within culture, which itself is a dynamic set of values influencing the attitudes, perception and behaviour of the subjects thus enabling or constraining their social action. While the culture‐centred approach places agency at its theoretical core, Dutta and Basu (2007) recognise that culture ‘emerges as the strongest determinant of the context of life that shapes knowledge creation, sharing of meanings, and behaviour changes’ (p. 561). Situating agency within a cultural context adds personal and social history to the structure–agency framework, where different ‘contexts support particular agentic orientations, which in turn constitute different structuring relationships of actors towards their environments’ (Emirbayer and Mische 1998). Agency, then, creates the possibility for actors to transform their relationship to structure and offers the possibility for social change through time.

Subaltern studies theory (Guha 1988, Spivak 1988) proposes that the dominant paradigm regarding historical, ideological and economic forces often stifle marginalised communities by imposing a standard set of values that are not tailored to individual needs, thereby excluding endogenous community values. Subaltern subjects are marginalised through different institutional mechanisms that are sometimes invisibly enacted through their diverse cultural ideologies on the healthcare system. This results in the lack of information on the subaltern sectors from the dominant spaces of knowledge. Through verbal communications and dialogues about locally constituted stories, culture‐centred interrogations seek to understand how community members at the margins of mainstream healthcare systems negotiate meanings of health and constitute their actions located within dominant structures of meaning. Many studies guided by culture‐centric approach have already been applied in various context‐specific areas to explore their health‐related perceptions (Basnyat and Dutta 2012, Jamil and Dutta 2012, Kandula and Baker 2012, Yehya and Dutta 2010). In India, however, this approach has been limited to studying the health perceptions of tribal populations (Dutta and Basu 2007, Dutta‐Bergman 2004). Drawing on the concepts of the culture‐centred approach, the present study attempts to get a more nuanced understanding through the narratives of urban slum dwellers the diverse components of knowledge, neighbourhood, material culture and symbolic drivers of a place that interweave in shaping their patterned way of valuing health and managing in practice.

Methods

Study setting

This study is part of larger qualitative research investigating the diverse health beliefs and practices of people living in urban slum settings. Motijheel Basti and Sahid‐Smriti Colony from Kolkata, and Nakkale‐Bande and Ullalu‐Upanagar from Bangalore as the research settings were purposively selected, because of their distinct and diversified history of slum emergence, living conditions, cultural landscape and plethora of health services (Das et al. 2018, Dyson and Moore 1983, Karve 1965, Miller 1981, Nagendra et al. 2012, Sopher 1980). A transcendental phenomenological approach was adopted in this study, in consideration of the fact that generating the essence of the lived experiences and meaning‐making on health concept shared by the participants is the core objective of the study. This design allows researchers to hear and illuminate how the participants apply meaning to the phenomenon without prejudgement (Dowling 2007).

Study participants

The participants in this study included 67 men and 68 women from Kolkata slums and 62 men and 48 women from Bangalore slums. The respondents were selected with a stratified purposeful sampling technique, to ensure a theoretically representative sample in terms of gender, age, religion, place of origin and linguistic/cultural variation. In each subgroup, a saturation approach was adopted to ensure appropriate sample size, and respondents were recruited until information redundancy occurred (Angeli et al.2018). Mostly, after the 8th interview, no new shared themes were generated from the interviews. Therefore, based on the data saturation model (Sandelowski 2008, Saunders et al. 2017), it was accounted that the data collection had reached a saturation point. Snowball technique was deemed to be appropriate for recruitment because participants found it confounding to openly discuss about their multiple situations (e.g., pertaining to sexual and reproductive health or illnesses that are associated with social stigma) with the first researcher with whom they were not familiar with and perceived a risk associated with self‐disclosure (Sydor 2013). The first author's ‘native’ status offered the opportunity to recruit initial informants by using existing networks and contacts. Participants were selected based on two essential eligibility criteria. First, the participants must include a composition of the older migrants (both in terms of age and stay that is <15 years), new migrants (residing in the slum for a minimum of 5 years) and native migrants (born and bred in the slum) and these terminologies will be maintained throughout the paper. Second, the participants must belong to the predominant religious and cultural group of the slums. Regarding the cultural and religious background of the respondents, we noticed that cultural minorities (Buddhist, converted from an originally Hindu caste) were present only in the Bangalore slums. When interviewed, we noticed that other than the rites‐de‐passage (that is ceremonies related to birth, marriage and death) daily based rituals, in particular, health‐related activities, were followed in line with the Hindu customs. Based on these observations we inferred that their cultural practices were not strong enough to induce cultural changes in the neighbourhood or bring new forms of social desirability in residents’ perception of health. In fact, the minorities tried to assimilate their rituals with those of the majority culture to avoid cultural isolation and instead secure community trust as a resource for social and material support. At the same time, they kept the very major life events intact to preserve their cultural identity as Buddhist. The demographic characteristics of the participants are presented in Table 1.

Table 1.

Demographic characteristics of participants

	Kolkata				Bangalore
	Motijheel (n = 69)		SahidSmriti (n = 66)		NakkaleBande (n = 48)		UllaluUpanagar (n = 62)
	Men (n = 33)	Women (n = 36)	Men (n = 34)	Women (n = 32)	Men (n = 29)	Women (n = 19)	Men (n = 33)	Women (n = 29)
Age
Below twenties	4	3	3	3	2	3	3	4
Twenties	6	14	6	13	8	6	8	10
Thirties	10	9	12	7	12	5	12	8
Forties	7	6	10	6	4	4	6	4
Fifties	4	3	2	2	2	1	2	1
Sixties	2	1	1	1	1	0	2	2
Religion
Hindu	12	12	12	12	12	12	12	12
Muslim	12	12	12	12	12	4	12	11
Christian	9	12	10	8	5	3	9	6
Place of origin
Rural	1	10	24	27	9	9	9	8
Within the city	15	17	5	2	5	3	4	10
Displaced	6	0	0	0	4	0	16	9
outside state	8	9	0	0	11	7	4	2
Outside country	3	0	5	3	0	0	0	0
Linguistic/cultural groups
Hindi (Bihar, and UP)	15	17	4	7	8	10	12	12
Bengali	18	19	30	25	0	0	0	0
Kannada	0	0	0	0	12	3	10	8
Tamil	0	0	0	0	5	3	6	7
Telugu	0	0	0	0	4	3	5	2
Years of stay
Older migrants	8	12	13	18	11	4	6	5
New migrants	6	9	16	12	9	5	7	3
Native migrants	19	15	5	2	9	10	20	21

Source: Based on data collected in the earlier phase of the field study

Data collection

The interviews were primarily conducted by the first author, in colloquial languages of the respective cities, namely Bengali and Hindi in Kolkata and Kannada and Hindi in Bangalore. However, under the supervision of the first author interviews in Kannada were conducted with the help of a Kannada‐speaking female translator who was a native, having a linguistic background (M.A in Kannada, PG Diploma in BashaVignana [Linguistics]) and was working in a health and social care organisations. A semi‐structured interview guide was used for the face‐to‐face in‐depth interviews that lasted between 45 and 60 minutes, depending on the participants’ interest. Interviews were conducted at locations chosen by the interviewees (homes and workplace) and the guiding questions covered participant's knowledge about health, the process of sense‐making about health within their sociocultural settings, and method of preserving and managing health and illnesses. Interviews were tape‐recorded with the participants’ consent.

The authors’ host institution has followed the study throughout and guaranteed for the ethical suitability of the adopted methods in the local Indian context. Oral informed consent to participate was obtained from all the participants. The study has thoroughly followed the ethical guidelines for social science research in health framed by National Committee For Ethics in Social Science Research in Health (NCESSRH) (Jesani and Barai 2000, Shah et al. 2000). Ethical approval for the study was obtained by the Institute of Socio‐Economic Change in Bangalore, India.

Data preparation

All the audio‐recorded in‐depth interviews were transcribed verbatim in their respective colloquial language. Translation occurred concurrently with data collection, to enable the research findings to be shared with research participants and the local research assistants who could check and confirm the interpretations accurately reflecting their perceptions and experiences. For this cross‐cultural study, Brislin's (1970) model of translation was followed where two bilingual persons were assigned to reach concordance of meanings between two different languages. One bilingual person who was proficient with both English and Kannada was assigned for forwards translation of the field transcripts from the source language (non‐English) to the target language (English). Another bilingual person, who also acted as a translator in the field, back‐translated the documents from the target language to their source language to ensure the accuracy of the translation. The translated versions were compared to achieve equivalence between original and targeted language. Any discrepancies that have occurred during the process were resolved through discussions with the two bilingual translators.

Data analysis

Transcripts were analysed with an inductive approach or ‘bottom‐up’ way (Frith and Gleeson 2004), preliminary coding and subsequent themes were identified from the data after reading and re‐reading and was not informed by pre‐existing theory or hypothesis. Interviews’ transcripts were coded using the guidelines of Braun and Clarke's (2006) thematic analysis for identifying themes, build patterns and develop complex themes. In particular, Gioia et al. (2013) three‐step approach was applied, that involves coding process of open coding and axial coding, in line with described by Strauss and Corbin (1990). In the first step interviewees responses were deconstructed into common codes based on shared ideas. Codes were assigned to words, phrases, sentences, or paragraphs from the transcripts. By this first process of open coding, 455 segments were coded along with 78 first‐order codes. The second step followed axial coding that consisted of identifying relationships among the open codes and collate them to develop abstract themes from an empirical to a more theoretical level. This process allowed clustering of 16 general themes into second‐order themes. In the last step, the second ordered themes were analysed for emerging common threads, resulting in five abstract aggregate dimensions.

During each iterative cycle, transcripts, codes, aggregate concepts and global themes were compared in all directions; transcripts were re‐coded and re‐categorised as themes emerged from the data. Consultation with the second author helped refine aggregation of the last themes and overcome biases, as the second author did not participate in data collection and was hence having the role of an independent observer. While the open coding was fully conducted by the first author, the processes of axial coding and generation of abstract constructs followed iterative consultation rounds between the first, second and third author. Because the second and third author was removed from the field and did not participate to data collection, they acted as impartial and independent observers, therefore reducing the potential bias that the first author – being fully knowledgeable and immersed in the field – could have carried through in the analysis of the material. These consultation rounds, leading to a process of collective coding, was also important to reduce the bearing of any of the authors’ cultural background and identities on the understanding of the collected evidence. The first author is of Bengali heritage and ethnic group, which could have led to a more in‐depth understanding of the Kolkata slum, a setting that is closer to the first author's cultural and linguistic background. However, the second and third author each hold a different cultural and disciplinary make‐up, which allowed for a multicultural, interdisciplinary view on the interview material, and on the coding results. Finally, an important aspect of the research design of this study is its comparative approach, which further reduced the potential bias related to the first author's cultural identity. In fact, cultural meanings and theme aggregations have been gauged through the inclusion in the sample of a second site, Bangalore, where most respondents did not share any linguistic, cultural or ethnic commonalities with the main researcher.

Measures to ensure the validity and reliability of data analysis and interpretation

To ensure methodical clarity, structured and provable approaches throughout the analysis; various techniques were applied. The validity of our findings was further enhanced by utilising the methodological and investigator triangulation methods. Methodological triangulation allowed comparing and contrasting the data collected during individual interviews while investigator triangulation ensured participation in data analysis by multiple members of the research team both in and outside the field.

The former triangulation method included the availability of topic guides as mnemonic devices, digital audio‐recording, independent preparation of the verbatim transcripts by the research team, thorough rechecking of the translations against the original audio‐file, standardised coding and analysis of the data along with the creation of an analysis audit trail to document analytical decisions. To address deviant cases or possible conflicting interpretations, discussions with the research team were conducted to assess the robustness of our findings and ensure a wide range of participant views. Our analysis was further aided through investigator triangulation by regular discussions about the findings with the multidisciplinary, international researchers and co‐authoring team, to mitigate potential biases of the first author that might relate to her disciplinary background and/or her embeddedness into the study context. Data generation continued until the point of saturation, where no new major ideas or perspectives were emerging.

Results

A dichotomous concept on health meaning emerged from the interviews with the participants where health is internalised either as wellbeing (positive aspect) or ill‐being (negative aspect). However, this concept is not confined to the abstract idea but furthermore manifested in their cultural sporadic practices in the form of preventive and curative care as well as in their everyday living arrangements. Such abstract ideas, every day or occasional activities are conditioned by a complex arrangement of physical, psychological, financial, sociocultural and environmental dimensions that primarily shape the self‐construction of health. The results are presented in five broad themes further explained in this paper, with their relevant subthemes.

Wellbeing/Ill‐being understood in the realm of physical elements

The majority of the participants considered the physical dimension of health as the initial and core of self‐assessed health. The concept of wellbeing for them is embedded in the absence of any symptoms of an abnormality in the body. As a participant expressed, My health is excellent. I have not lost weight; my body is toned even at this age and my face is still graceful like those of young girls.

However, perceived susceptibility differs between older migrants and native or new migrant participants. Older migrants exhibited better self‐rated health compared to the latter. For instance, when asked to all the three groups of participants how they contemplate the ‘symptoms of abnormality’, older migrants referred to a functional definition of health in terms of physical aspects. Reported examples are pain and tiredness, feeling of being overwhelmed by their daily activities or when tasks take longer than usual to complete they consider it as a symptom or clue from the body that they are not well. In this context, one participant expressed, when I experience shortness of breath, I can do a little activity at a time and then had to rest as I don't feel well. The severity and importance of health problems were controlled more by the capacity to deal with problems than by the problems themselves. This ability was associated with retaining traditional rural‐urban dual social structure such as strong family support, traditional family‐based caring, well‐built social and kin network ties with the native kins attenuating the perception of difficulty in obtaining care‐giving. Such social position, network size and normative practices are outcomes of past life course events. As one older migrant expressed, When we came here seventeen years back, there were no worries about eating, no worries about drinking. You see … there were no worries about money. Even today for us continuing care for the family, eating right and healthy, observing healthy habits, maintaining strong bonds with the roots constitute to fight with depressive symptoms, stress, chronic diseases, and functional limitations … this generation everyone is running after money. Yet no one is happy. Money cannot secure everything. They (indicating native and new migrants) are lonely; don't eat right and simple food, no space for respect, security and belonging. Inter‐generation relations have undergone a drastic change that simultaneously slowed down their chance of well‐being.

At the new and native migrants’ level, for an illness that is not serious in its initial manifestation, a common practice is to not consider it as ill‐health in its early stages and to hope it disappears in a few days. However, when physical functional ability and stamina severely decline, the participants acknowledge that their bodies are becoming hard to control. As said by a participant, I kept on pushing myself in doing work thinking that everything will be fine … because feeling sick is so emotionally demanding. But 1 day when I suddenly collapsed I tried to retrospect that my energy was actually getting worn out faster and faster and then suddenly stopped.

Further inquiry with them revealed that health by habit is not an everyday concern. For new migrants fear of denunciation, lack of knowledge as to how to navigate the health system and access to costly treatment negatively impact their health‐promoting habits and entails negative health consequences. The moment when diseases become unbearable is where they acknowledge their concerns for health and that defines their construction of health. To this, a participant responded, I don't know whom to approach here, there are so many doctors, different types like you know different doctors for different diseases, I feel like they are all money‐suckers!!! Just sitting there for you to come and give every penny that you have with you. Back at home, we used to have one doctor, whom we could approach for each and any kind of health issues, yet he is friendly and nice. Here, going for the treatment itself is very confusing and doubtful … so we try to avoid those places until emergency calls.

For native participants in spite of having better social integration and local support (due to home communities), stressful economic(difficulty to enter the labour market and securing the position) and social experiences (low self‐esteem, acceptance, struggle for quality and sustaining resources)related to living in a slum environment demand everyday solution and health; therefore does not constitute one of those factors; impeding adequate use of health services. For instance, many participants reported an irregular water supply, forcing them to buy water drum for cooking. To this, a participant said, We use pond water for washing and bathing. So the water is not for drinking and the drum that we buy cost us Rs.500 monthly … living in a city where everything is so expensive this amount of money also matters and is quite difficult to earn … but every month you have to earn this money to ensure the provision for drinking water, which is more important than healthcare.

Wellbeing/Ill‐being understood in the realm of psychological elements

In this context, health and illness are professed as two different entities and not as contrariety. This means that disease does not inevitably lead to ill‐health and that absence of disease is also not an assurance for good health. It is a matter of how one is doing (feeling) in the actual situation. As one participant replied, during the day I might be happy but by night I may feel sad from inside and may not feel like eating. This, however, does not mean that I am actually ill but is not healthy either … maybe for some reason I am too stressed out. Therefore, the holistic senses of feeling healthy are subjective; it is about feeling good, being content, even if a person is ill.

While the subjective notion of health for some participants is purely measured by psychological factors, for a few participants the nature of illness controls the mind‐body expression of subjective wellbeing. For instance, bodily experiences such as cold, cough and fever are considered a small illness that does not match an individual's emotional response with more serious or chronic illness. The mundane illnesses are so commonplace in their life that they hardly consider it as perceived risk. As a participant said, every night I get this headache and pain in my body which disappear during the day. Yet I say I am healthy because I can endure this pain, feel contented and can do my work without feeling unpleasant. Now, I cannot say I will feel the same if I had cancer; I will be more worried and stress.

Chronic illness, on the other hand, once detected leads to emotional dysfunction among participants, like anxieties, uncertainties, fears and losses. As a participant said, I am trying to become hopeful..but often I get devastated; I know I cannot afford the money for treatment and I am dying too … I am lost and don't feel like to live life anymore. Coupled with that are strained social relations and isolated life due to the inability to work. Participants said that social relations are of particular help for improved wellbeing, as they provide social support in the form of emotional and practical assistance. A participant recalled her experience, I was bedridden for 3 months and felt like totally cut off from the outside world … I used to miss badly my friends where I work because, other than work, we used to chitchat and give emotional support to one another for dealing with practical problems of life.

An important toll on the psychological wellbeing of slum dwellers is taken by the financial insecurities related to living in a slum settlement. Participants said that loss of work changes their financial situation, which directly affects their social life and consumption patterns. This results in feelings of isolation, loss of self‐esteem and feelings of hopelessness that affect their mental wellbeing. Feelings of hopelessness and longer duration of unemployment also, in turn, affects their physical wellbeing (e.g. in terms of sleeping problems and triggering damaging health behaviours such as smoking and drinking problems). As a participant said, like this will sound really stupid if someone says that he lost his job and yet he is very happy … you understand it right!! losing a job means no food, no peace of mind, quarrels at home … you feel like sinking … sick from both in and out.

Some participants mentioned that they have to do various types of work completely unrelated to their former work experiences; often they do not have the appropriate skills, cannot endure the physical demands of the work, or find that the wage is too low to cover ongoing living costs. They face myriad of health‐related challenges while getting accustomed to the job. As a participant said, to pursue wellbeing is really difficult here … I, for example, pull a rickshaw, have to work entire day to earn a decent amount and at the end of the day I feel sores all over my body … no matter how we feel health.. in a positive or negative way, wet pay no heed till we can endure.

Wellbeing/Ill‐being understood in the realm of spiritual elements

Majority of the older migrants and women participants indicated spiritual health purely in terms of restoring and preserving better health and quality of life. Spirituality has been found to be important in making meaning of life, combating physical and mental illness and is a predictor for the use of traditional practices. The participants believe that one who does not tend to shape a relationship with God or spiritual power cannot combat illness. The ascription of misfortunes to an individual pervaded by spirits was observed to be common among the participants. Sicknesses, contagious or chronic diseases and even accidental or painful death were explained in terms of divine punishment by enraged or malevolent spirits. As a participant said, offering namaz (Islam's five daily ritual prayers) is a way of mind‐body therapy. Namaz provides not only physical but also spiritual and intellectual growth.

Many participants believe God to have human personalities; that is, they are benevolent and kind if they are worshipped, while they can get angry or irritable when they are hungry or get jealous if more attention is given to other deities. Spirits and Gods demand tributes in return for protecting physical and emotional health and punish for failure to pay proper homage. Not practising religious beliefs such as praying, visiting temples, mosque or church or organising religious ceremony attracts malevolent spirit and a projection of sufferings and death. Spiritual rituals and practices are mean to exorcise spirits for a participant, We must show respect to God through proper rituals such as praying, observing the local customs and values, etc., in this way you can ward off the bad spirit and keep your family safe from harm and bad health.

Sorcery and witchcraft are also considered as a causative agent of ill‐being. Participants consider them as a human creature who can metaphorically suck the soul out of another human being until the victim gets very sick and die. It is commonly believed among the participants that witches operate through fowl‐wind and fetishism. As a participant said, Customarily after marriage for 1 year we don't allow the new bride to cut her hair, wash her bridal sari or pull off a head veil … witches can easily use them to create a fetish object, cast spell on them resulting into sudden death of the husband or an unhappy marriage.

Again the construct of health and spirituality for some older migrants is typically determined by diminishing family support. Religious activities and faith‐based organisations, therefore, compose the most objective dimension as social and instrumental assistance that is available from one's social network (e.g., providing transportation, care when sick). Being involved in religious activities as group members’ further enable these older participants to experience not only support but help anchor identity by connecting with people from their community and derive inner strength. One participant, for instance, expressed, Our children remain preoccupied with work and fulfilling family demands, so they cannot offer much time to providing companionship to their ageing kins. Therefore to avoid isolation, bad thoughts that make us sad, or to gain inner strength by thinking that we are not alone; we engage ourselves in religious practices and domestic rituals such as prayers, attendance, services or celebrating religious functions with our fellow brothers … we get the feeling of the homeland. Social support mediates resources and opportunities for healthcare services such as hospital visits and medical examinations, understanding clinical care and translating its languages, and transportation to see physicians. Therefore, greater social support simultaneously functions in greater healthcare services. As added by a participant, ‘My son does not stay here to take care and she (indicating daughter‐ in‐ law) is already juggling with the tasks of taking over her husbands’ roles and caring for the children. I do not want to overburden her with making care arrangements for me. The Sangha (faith‐based organisation) with whom I am attached give adequate care assistance whenever required … they have good connections with big doctors. Therefore, Sangha members can easily and usually organize for our treatments and help us by paying the associated cost … once they even arranged my transportation to the hospital along with the medicines, they also have their own brand of common pain relief medicines which they provide free of cost when asked for. I made my daughter‐ in‐ law join the Sangha so that she too can avail the facilities. They helped with some loan in starting up her home‐based businesses.

New and native migrants (primarily male participants) mainly relied largely on their faith to cope with physical hazards, misery and isolation. They put down their destiny in the hands of Supreme because of their inability to control circumstances. Religious support also manifested itself through local agents (religious institutions, faith‐based organisations, traditional healers, priests, church members, etc.) trusted as intermediaries between the slum residents and the divinity. These associations also provide much‐needed assistance with money, basic needs for survival and legal assistance in adapting and securing their place within the slum and, at the same time, keep up their cultural legacy and warrant their ethnic identities. Some participants said that they believe in visiting the church every day as the form of healing retreats and do not prefer any other form. This is because the Church helps them provide conventional health care through community‐based interventions such as frequent medical camps, wellness clinics or wellness group sessions. A participant commented that wellness group sessions as give us knowledge on how to engage in healthy behaviours and help us practice them by providing us with food and medicine. They deliver various messages as for how to remain healthy, make godly choices and conduct programs on healthy habits.

Despite witnessing escalating transcultural habitat because of rapid migration processes, participants’ particularly older males and females (both young and old) demonstrates to still heavily draw on their cultural framework to conceptualise health. Participants said that, throughout the year, they perform manifold rites and rituals associated with village folk deities who are the pivot of disease controllers. These activities seem to be quite constant in the study slums of both the cities. Although, religious and communal variations are witnessed in names and forms as well as in obligatory methods; but all of them have the most codified and monitored set of health practices, have explicit prescriptions and proscriptions about health behaviours, food habits, teach their members to respect and take care of their bodies by showing gratitude towards God through various ritual manifestations (prayers, animal sacrifices, calendrical worships etc.). For instance, in Kolkata slum, Shosthi is worshipped in various forms with various significances. She is worshipped in the form of Shosthipujo (for maternal and infant mortality), jamaishosthi (for the wellbeing of sons‐in‐law), sheetalshosthi (sexual wellness), neelshosthi (child's safety and prosperity). One participant said, She is considered as the guardian angel in every household and these rituals are primarily performed by women only, some by young unmarried girls and some by both. Hindu participants of Bangalore slums worship Mariamman who is believed to take care of several diseases like a ‘general physician’, and is quite popular among the slum dwellers. As an older male participant said, We worship her by offering boiled rice, fruit, flowers, kumkum, burn incense and camphor and she protects us from all demonic attack Muslim participants approach spiritual or folk healers, locally known as fakir or Pirbaba, who observes various healing rituals. They use various kinds of herbal made ointments or give talisman or amulets made of religious scripts or dohas (lucky chants in written form) to ward away any malignant forces. One woman participant said, When I was pregnant I used to get terrible dreams at night and shivering as well. Baba gave me a tabiz (talisman) inscribed with Doha and said this will protect me from an evil spirit who was trying to enter my womb. For treating diseases, spiritual healers recite verses from Quran or employ touch, breathe over the forehead, apply coolant herbs or scented massage, or use physical objects such as talisman, amulets and mascots to touch the skin.

Wellbeing/ill‐being understood in the realm of cultural practices

Post‐migration challenges such as loss of social structures, self‐identity and cultural values produce and re‐produce (mobilising, enacting, validating) cultural capital in a new way and health construction is of no exception. Health conceptualisation in slums is compounded of congruity of the culture of origin and culture of the destination.

The older migrants, who migrated to the city 15–20 years ago, have seen a rapid change throughout the period and gave retrospective accounts of the change in lifestyle that they have witnessed. They made comparisons between native villages or then newly built slums and the current living conditions affecting their wellbeing. Regardless of circumstances, they moved in the past mainly to able to enjoy the quiet and the good weather, going at a slower pace with less bustle and aspirations of a better life, especially for their children. They share health meaning in general as an individualistic search for ‘the good life’. As one participant remarked, When we arrived here we came not with the notion of the rat race … we wanted a decent life, good education, good food, large information‐based network, we had an optimistic attitude towards life and future, which is consistent with good health. We could do that because we had more native‐based counterparts on whom we could rely unlike this younger generation … we migrated at a very early age, were, therefore, ahead in time and settings (e.g., work, neighbourhood, recreation centres etc.) to develop and maintain social ties. These days kids miss those opportunities, … these places nowadays are of no prospects or good life as everyone started to migrate here from every corner … faces are unfamiliar and hard to trust … people here often had to even leave the family to get better work, they feel isolated and without friends which is very depressing.

Many participants mentioned that earlier when they moved in they use to live in open spaces and relatively bigger houses with good ventilation, but gradually influx of migration and crunching space made them shift in a one‐room house, which is congested and overcrowded, making them unable to move around freely. The expectations and practices of the older generation that they acquired over time through historical position are difficult to maintain in the new environment and they do not possess the right capital to claim desirable field positions. As a participant said, When we first arrived in Nakkale‐bande, the slum was comparatively new, it was not that crowded … everybody knows each other … we all came from nearby villages, friendly atmosphere it was and everything was so pure and fresh … air, water and vegetables … I felt like I was still in my village … it never came to our mind we have to stay fit and healthy … I mean what is health … I don't remember talking anything related to health or even knowing its existence … in due course, everything has changed … the city has become crowded and polluted, more people started to come in to settle, we don't know many of them or what they did in the past … we now ask people in the form of greetings … are you in good health … is everything ok.

A considerable proportion of widows shared their views about foods that activate sexuality are avoided. These foods comprise of non‐vegetarian (meat, fish, eggs, onions and garlic) and spicy food and certain pulses such as red lentils. They are substituted by cold food such as muri (puffed rice), chera (beaten rice), green lentils, flatbread or dietary products such as curd, milk or cottage cheese. Also, on the eleventh day of each lunar month (Ekadasi) fast is observed. When asked how fasting is connected to their understanding about health one young widow participant said, Fasting is observed not to attain salvation. It is simply to weaken the body from sexual desires. Elderly married women as well as male participants also observe the food restrictive code (though not necessary for them to follow if they do not wish to) out of ‘habit’ or to maintain a celibate status. To this one woman participant said, For me, food avoidance is not associated with sexuality because anyway I am asexual after menopause but I prefer to eat cold food to make mind and body self‐contained. Another old male participant also similarly remarked that We are not imposed with these diet restrictions but still with growing years we get inclined towards cold food as our system itself cannot digest hot food. Food taboos for young female participants primarily revolve around the concept of ‘hot and cold’ foods and divergent opinions were noted on its avoidance or restrictions … For instance, pregnant and lactating women believe that following food restrictions helps to avoid health risks and mortality of both mother and baby. As a pregnant woman participant said, we avoid eating ripe pineapple, papaya or any food that makes your body hot. This directly affects the foetus and leads to abortion. Although eating well to remain healthy is a known knowledge yet in many instances adolescents and young participants believe that these food restrictions are not feasible to follow in urban settings where everything is added to the expense. To this, a male participant said, These do's and don'ts fit in villages where everything is very cheap or where you grow vegetables in your backyard … but here you have to get everything with money and therefore it is not wise to become choosy … you have to eat what gets into your heart … Someday it may be fish, meat or someday very plain food like salt and rice. Almost every Hindu participant abstain themselves from taking beef while some of them never eat or handle meat, fish or eggs. For them consuming vegetables and plants in the form of food is considered less morally wrong than killing animals. For a participant, food that requires slaughter and sacrifices causes pain, distress, and disease. Muslim participants abstain themselves from taking pork or dead fish and the reason described by them as, Any dead animals and flesh of a pig are not permissible for consumption as they are considered to be unclean and is the root cause of many diseases. However, no food taboos are observed by the Christian participants as they believe in consuming any edible that are readily available in the market and are fresh.

Adherence to traditional cultural practices for older males was related to the length of stay in the slum. For instance, older generations participants who are also matured residents do not have to face the stresses of fresh migrants who constantly fight adapting the strange and unpredictable environment of the host culture. Being older residents they learned social competencies such as (community and leadership skills, cultural capital and coping skills), and established social and organisational ties (social capital, local network closure and extra‐community skills). As a cultural majority, being able to share social meaning and cultural values, older residents could maintain their original religious beliefs as a source of comfort to attenuate both ageing‐related and new age challenges. In the case of women, it was changing authority within the household, traditional family structure, gender roles and expectations. For instance in female‐dominated households; a household is a symbolically fixed resource that serves as an anchor for the members to preserve their cultural identity. These home‐based rituals provide space to worship in familiar method and also create avenues to connect with familiar ethnic ways of life, community information in the migrants’ native language, psychological and instrumental support for newcomers who need health care, housing and jobs (able to organise themselves and their communities in a way that reconstitutes their native systems). As one participant remarked, this is not just an occasion or a ritual … we now stay in cities, faces many disputes and hostility, family problems, job insecurity … this festivals give us strength to survive, get a chance to rejoice amidst sorrow, to connect with people of my community and makes us feel one family in an alien land, feel my culture through language, food or festivals. Even, religious rituals carried out in the home helps to maintain my cultural identity. I feel still connected to my roots otherwise here everyday things changes … food, dress, lifestyle even your “self”.

Economically dependent female participants expressed that their culture undervalues, especially women's freedom and decision‐making autonomy. This dependency promotes the stereotype among them that they are the ‘keepers’ of religious norms and practices. It is their prime duty as a caregiver. Traditional values further impart that women are responsible for a performative act such as care‐giving to preserve the life and happiness of their family even to the extent of sacrificing personal aspirations. As one woman participants said, I believe I must protect and promote the well being of our family … I perform and fulfil my role by following those every ritual taught by my elders that can bring the good prospect to the household including health. Women are helped in their caring mission by access to resources or stakeholders that are close at hands such as following traditional customs and rituals or approaching faith healers or traditional medicine men. These arrangements are more often seen as offering complementary rather than adversarial services involving easy access to free or inexpensive health services. As a participant remarked, I feel little confidence in my abilities to make health‐related decisions, go out and seek a doctor, because I don't have money of my own. Hence kind of care that is traditionally acceptable is following these rituals. I don't know whether these works or not but; at least gives me a sense of safety and helps in meaning‐making to my actions as a caregiver.

Wellbeing/Ill‐being understood in the social realm of the slum environment

Many participants mentioned that living in a slum can itself become a reason for unhappiness, as life is harder than before, food more expensive and jobs difficult to fetch or earns them little money. Consequently, crime exists, social cohesion not easily available, familial conflict and stress and lack of support give rise to psychological distress that contributes to the construction ill‐being of health among the participants. Many women participants revealed about intimate partner violence, which profoundly affects their overall health. As a participant said to this, we understand health as happy from inside but when I see many women here being beaten by their drunk husband I feel though they show happiness they are very sad from inside … and when you are distressed from inside you are not healthy.

Some participants highlighted that unsettling is also caused due to lack of social solidarity, lack of supportive spouses and lack of supportive environment in the slum. According to many women participants, their husbands do not contribute money to meet the household expenses as a result of which they have to go out for domestic work without any formal leave. Most children are left alone at home during those hours with no one to look after even when they are unwell. To this, a participant remarked, there is no way in relieving your stress … you have to go leaving your small kids back at home with your main door locked and no one to look after … many women like me goes out so you cannot say anyone one to take care. While we are away we constantly fear about the safety of our child. Again some women participant expressed that, because of some antisocial activities that take in the public toilets at night, these usually remain closed at that hour and they had to control their nature's call. To this a woman participant said, We don't go to any faraway public toilets at night because it is not safe for a woman to go out alone at night and the one which is very close to our slum remains closed at nights … this is very uncomfortable and we are not happy with such arrangements.

Almost every participant expressed that they do recognise health to be a vital element of existence and taking care of health is of equal importance for subsistence. However, the everyday problems that they experience like the dirty environment, lacking toilets facilities and insecurities without any viable alternatives do force people to exhibit hopelessness towards maintaining good health and dejection that things would never change. Consequently, they choose to overlook health, regardless of whether they are in good or bad health. As a participant said, often it is also good to try and feel satisfied with what you can afford … facing problem is nothing abnormal so one should not constantly worry about health … in the course of life sometimes you will be ill and again you will revive. While another participant said, I can't say I am satisfied but we have to adopt certain survival strategies by overlooking things and accepting the problems of life the way it is.

Discussion

The current study addressed the subjective construction concerning health, as well as views about the key drivers shaping the concept of health among the urban slum dwellers of Kolkata and Bangalore city. Understanding the direct perspectives of slum dwellers around the implication for their health of being a member of urban slums has been rarely explored in the Indian slum context. The findings suggest that the study participants associate health with existential meaning (DeMarinis et al. 2011), understood in their cultural setting. Existential sense making in here refers to all sorts of expressions and to how health is interpreted, such as traditional and religious expressions as well as other circumstantial expressions (Lilja et al. 2016). Health is conceptualised as a feeling that can control every mechanism of the human body, including their physical, psychological, social, emotional and spiritual features, and this aligns with the model as defined by the World Health Organization (1946). The holistic understanding of health is consolidated into a binary concept of wellbeing and ill‐being and can be summarised in the statements of ‘feeling good’ and ‘feeling bad’.

As an explanation of their health construct, participants affirmed a general acceptance of bio‐psychosocial understanding. The idea of health implies healthy body syndrome, which is an accepted concept among the slum participants and is considered as an ideal. This view finds support from the findings of another study that explores the meaning of health and wellbeing (Saltonstall 1993). Participants recognised that when far‐reaching physical endurance gets exhausted without any further resilience, this is when they consider the health implications. The key predictor that emerged for producing health meaning is any abnormal somatic symptoms that disturb the equilibrium of their daily life. Slum participants understand health as being associated with not only physical wellbeing but also liveliness, contentment and serenity of the mind, which indicate that our results are similar to those of other studies (Alfinger and Causey1995, Corbin 2003). From the narratives of the participants, a double hermeneutics emerged while viewing health (even though ill) from psychological constructs: (i) feels good and can do and (ii) feelings of bodily vulnerability. The former phenomenon emanates in the context of ‘knowing one's body’ (Corbin 2003). Brief instances have been accounted in this context, where participants remarked that as an individual they may be ill and yet may feel healthy. From their perspectives, it can be deciphered that participants trust their mental body's language till they can adapt with physical peculiarities, usually unconsciously, such as knowing in what manner their body reacts to stress, how much they can conform with the routine before becoming fatigued and what is the perceived time to carry out chores till the body cease to function no longer as desired. This is consistent with Lidler's observation (1979) that conceptions about the body are the creation of one's interactions with the society where he lives. However, for the latter phenomenon, when participants experience that there is a change in sensation or a change in appearance that cannot be anchored meaning but requires a clinical interpretation of vital signs, that leads to the suffering that is more emotional than physical. This is because of the self, which is affected by what happens to the body (Ware1992, Williams 1984).

Health as spiritual and cultural knowledge shapes diverse perceived health risks and goals across different genders and age groups, as corroborated by previous studies (Moridi et al. 2016). Specific health behaviours differ among demographic groups due to their distinct health values and goals. This study shows that some demographic groups, for instance, women and older migrants, engage in a health‐enhancing behaviour to achieve a health goal that emphasises maintaining a balance of opposing forces to promote health and supplement the clinical care that implies costs. Its preservations and restorations are delineated through various long‐established traditional rituals, sociocultural practices, dietary habits and food taboos, including notions of hot and cold foods, versus body habits which still hold meaning and relevance (Nichter1987, Raman et al. 2014). All these personified images of spirituality, as reflected in the participants’ narratives, regulate the slum‐dwellers’ poor attitudinal and behavioural outcome, as influenced by the strained slum environment. In support of Troyer's (1988) findings to harness spirituality, participants furnish a set of practices to follow in their everyday life. For instance, diets, prayers and corresponding religious activities helped them transform tragedy and despair into the positive meaning and thereby preserve wellbeing (Walton et al. 2004). The findings build on previous studies, as religiosity and spirituality have been found to help immigrants in making sense of their life and circumstances (Wong and Tsang 2004); acceptance of illness, and coping and managing it (O'Mahony et al. 2013, Schreiber et al. 1998). Instead, the new and the native migrants re‐interpret religious devotion, and connect it with wellbeing not simply through calendrical and customary rituals but by involvement with faith‐based organisations. The quotidian practices ushered by religious beliefs, norms and values are rather observed to render instrumental and sociocultural support and to amplify their chances of leading a quality life.

The above described spiritual‐cultural norms and practices embracing health indicate that all domestic and traditional behaviours have not completely changed, rather re‐positioned with the new cultural settings, depending on the respondents’ experience as migrants. The pathways by which culture and religion shape the health meaning can be traced through cultural capital and gendered ways, by which extant sociocultural practices acquired different meanings and validations. For instance, cultural distances between older migrants were rather small; as enabled due to longer period of stay in the urban slum that helped to retain the cultural identity and life satisfaction. It is further validated by past literature that established homogeneity plays a major role for the cultural clustering of immigrant groups (Gross and Schmitt 2003),and also a strong predictor of preserving identity in host society due to linguistic and physical proximity created by the closer cultural atmosphere (Bredtmann et al. 2017). Religious services in the intimate atmosphere mediate social support and life satisfaction which is an indicator of positive health outcome (Yoon and Lee 2007).

The various sociocultural practices illuminate how religion and culture are used as a means of social security. Although new migrants reach the destination in search of livelihood and financial security; in this context, the concept of security can be pushed forward and can be argued that it goes beyond just economic outcome. It is about sociocultural identity. New and native migrants consider their wellbeing and their security not as individuals but also as members within the social and cultural framework of the new society. Therefore, as cultural agents, their decisions get reflected in a larger social structure where they want an opportunity to survive and thrive and to practice their culture in a safe environment.

Living in a slum alone can cause illness due to social divisiveness, like broken and marooned families, weaker social ties, financial hardship, job and housing insecurity and widespread inequity that inhibits managing a healthy lifestyle (Vaughn et al. 2009). Influences of the slum environment on residents’ health perception are pervasive and span across the physical domain, where attention to symptoms is reduced because of hardships of daily life; the psychological domain, where job insecurity induces worries and distress that inhibit a sense of wellbeing. This finding aligns with the recent studies published on Lancet (Ezeh et al. 2017, Lilford et al. 2017) that argue that insecurity is one of the foundational and unique characteristics of slum settings. Another feature typical of slum settlements is; the presence of neighbourhood effect, namely the negative influence of close physical and social proximity on health outcomes (Ezeh et al. 2017, Meijer et al. 2012). Interestingly, our findings highlight neighbourhood effects that are mostly related to psychological wellbeing. While the slum‐defining ‘intimate sharing of physical and social space’ has been often associated to increased epidemics and disease exposure (Lilford et al. 2017, Oakes 2004), the respondents in our study have instead highlighted its mental health repercussions, pointing to lack of solidarity, crime, feeling of unsafe and lack of supportive environment, stress and domestic conflict as psychologically burdensome aspects of slum dwellings. Rather than favouring the development of strong and stable social bonds, the proximity of people in slum areas leads instead to heightened competition for scarce resources and therefore to an environment more prone to conflict and criminal activity, with related psychological distress.

Given the hardship of slum life, and despite the inherently multicultural environment due continuous migration flows to and from slum settings, respondents display strong attachment to their cultural practices, that they have retained mostly unaltered from their previous village life. Observing local rites and customs seems a way to avert the significant role that sociocultural, environmental and behavioural factors play in health in the form of poverty, social support and medical compliance with treatment schedule, flexibility and acculturation. Slum environment remains inescapable and implies casual influences on health and health disparities, adding new meaning to health constructs, and has a profound effect on overall wellbeing. Acceptance within these environments makes it a plausible coping mechanism, giving them a sense of control and enabling them to survive (Lapierre1986). Essentially, the structural circumstances of the slum participants we studied are difficult for them to change or alter. What they seem to have become accustomed to is the body that is conditioned by a complex arrangement of physical, psychological, financial, sociocultural and environmental dimensions in a manner that a person has some control over. Dennis‐Antwi et al. (2011) have noted that lay perspectives of health and wellbeing are not static, but are unceasingly constructed in the contexts of changes and continuities in the social, cultural, economic and political experiences of individuals.

Limitations and strengths of the study

Some methodological strength and shortcomings of the study should be mentioned. The strength of the study lies in its informative knowledge extracted by focusing on the target group's understanding of the concept of health through differential interactions within the local sociocultural context. Such a dialogue‐based understanding of health as applied in this study is functional in other similar settings or formulation of health policies (Horsburgh2003) aimed at the slum population. The first limitation of the study is that, since the response rate was restricted to religious groups, the possibilities to generalise the results to a diverse ethnic population residing in the slum areas are limited. Although every possible effort was made for inclusive recruitment to the entire slum study criteria, yet many important voices have likely been left out who could have contributed to this study. Second, perceptions in itself are a subjective matter, not are fixed states but can vary from day to day or situation to situation. Therefore, a thorough understanding of the phenomenon of slum life and its interaction with health needs further exploration for a more valid and generalised understanding of the mechanism of health and wellbeing in a marginalised context. Last, since in one of the study field local interpreters (even after receiving the required training) were engaged for conducting the interviews, the issue of communicative validity remains, as interpreting the non‐verbal communication and the fact that different languages have different concepts may have not been possible to translate directly.

Conclusion

In conclusion, our study illustrates that health is constituted and negotiated among slum dwellers within the exemplified multi‐factorial set of representations. Structural conditions of the area, adherence to traditional and modern city culture, fatalism, decaying quality of life and so forth contribute to the health construct among slum dwellers. Despite varied problems surrounding health and wellbeing, slum‐dwellers maintain hope to survive and learnt to live with their physical incapability. As a result, they take the body for granted and this is reflected in their self‐concepts and identities of what they can do instead of what they cannot do. It is not to say that participants do not make efforts to enhance their bodies’ efficiencies, as it can be witnessed through their preventive and curative methods. The current study adds to the growing evidence that ordinary members of the urban slums can articulate critical linkages between their everyday sociocultural realities and health conditions that can support the design and delivery of interventions to promote wellbeing.

Competing interests

The authors declared that they have no competing interests.

Funding

No funding was received for this study.

Data availability

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.