Introduction
Globally, pneumonia remains a leading cause of morbidity and mortality, accounting for an estimated 502,000 deaths among children under five years of age in 20221–4. These deaths are largely preventable through readily available interventions5. However, healthcare-seeking behavior is often suboptimal for childhood respiratory illnesses, attributed to caregivers’ perceptions and beliefs6. Studies in other contexts have shown that knowledge of pneumonia symptoms is poor7–9. Symptoms are frequently misattributed owing to their overlap with those of other childhood illnesses and are often perceived as mild, leading caregivers to rely on home remedies10,11. Caregivers' beliefs about their children’s health, perceived causes of illnesses, and how to resolve them can greatly impact health-seeking behaviors. Beliefs, such as those related to causative agents and perceived severity, are often acquired through primary socialization, which plays a key role in shaping caregivers’ perspectives and health-seeking practices12,13.
The use of traditional medicines is based on core belief systems about illness causation, such as supernatural, social, or biomedical agents/natural causes14–17. The supernatural causes of illness include witchcraft, sorcery, breaching taboos, and disease-causing spirits. Belief in God/Allah was prevalent in many studies in Africa and was held responsible for causing illness17,18. Regardless of the basis of the individual’s belief system, it can inhibit preventive efforts, delay or complicate medical care, or result in the use of traditional medicine that can be beneficial or toxic18. In the context of divergent health belief systems, caregivers may face decision-making dilemmas regarding the best course of action for sick children.
Previous studies have revealed beliefs about witchcraft, which are attributed to the use of traditional remedies14,18. Symptoms such as convulsions and fever are often attributed to bewitchment, especially when the improvement is slow or the disease becomes severe. In such cases, the disease is treated using traditional remedies14,15,19,20. In contrast to findings from malaria and Ebola-related studies, there is limited literature linking the use of traditional remedies for pneumonia to bewitchments. For instance, a study conducted among fathers in the Philippines found that they utilized the services of traditional healers for pneumonia because it was widely accepted and easy to prepare; no other reasons were established for this study21. Similarly, a study in Sierra Leone examining healthcare-seeking behavior for pneumonia reported the use of traditional remedies because they were cheaper than treatments at government hospitals22. Additional motivating factors include social expectations and obligations, such as respect for elders, in-laws, and herbalists22–24, as well as the perception that traditional remedies are suitable for diseases that do not require biomedical treatment20. These findings highlight that traditional remedies are often deeply embedded in individuals’ dispositions, cultural norms, and social structures.
According to Hartog et al. and Workneh et al., healthcare is more than illness and pain; it reflects the basic worldview and values within the culture that, in turn, influence health and illness behavior25,26. Therefore, communities comprehend health problems and solutions within their cultural frames of reference17,27. For example, in the rural Bangladeshi community, it was believed that pneumonia could cause asthma and heart disease and require quick medical attention28. Although some communities did not appreciate the severity of pneumonia29, it was perceived as a serious illness requiring immediate attention in Nigeria, rural Ghana, and rural Bangladesh9,21,27,28,30–32. Opwora et al. further stated that caregivers’ failure to appreciate the severity of pneumonia may hinder the prompt treatment of the illness31. Although pneumonia is often acknowledged as a severe illness, the criteria for what constitutes "severe" can vary culturally. For example, some symptoms may not be perceived as alarming within a particular cultural frame of reference, leading to delayed response. This variability suggests the need for further exploration of the cultural definitions of severity and how they impact health-seeking behaviors.
Previous literature has reported that pneumonia is caused by a common cold, headache, cough, asthma, hot body, poor hygiene, measles, exposure to cold, dust, and hot summer months10,27,28,33. In addition, studies have shown a consensus that pneumonia is caused by environmental variables and existing illnesses; these studies show the importance of causation perceptions in treatment choices10,27,28,33,34. Owing to differing perceptions, home remedies for pneumonia vary from community to community. The treatment involved applying a methylated ointment, using hot water, and keeping the child warm27,28. It is logical for caregivers to use hot therapy to prevent and treat pneumonia because of the belief that pneumonia is caused by the cold27,28,34.
This study drew on Bourdieu’s theory of practice to understand caregivers’ perceptions of pneumonia13,35. Bourdieu’s theoretical concepts of field (social environment) and habitus (conceptualized as caregivers’ perceptions of pneumonia) were employed to examine how their ingrained dispositions, along with their cultural and social capital, shaped their understanding of pneumonia. The investigation focused on an in-depth understanding of pneumonia perceptions as shaped by caregivers’ social and cultural capital and the environment in which they exist.
Childhood pneumonia remains a leading cause of under-five mortality in sub-Saharan Africa despite the availability of preventive and curative interventions1–5. In Western Kenya, timely health-seeking remains suboptimal36. While many pneumonia studies focus on access and structural barriers, few explore caregivers’ pneumonia perceptions, particularly as shaped by their social and cultural capital and social environments. Applying the theory of practice allows for the analysis of the interplay between social and cultural capital within a specific social environment.
Methodology
Study design
This study adopted an ethnographic design with a qualitative approach, utilizing the philosophical paradigm of Martin Heidegger’s hermeneutic phenomenology. The ethnographic design enabled prolonged engagement with participants at their homes from April 2024 to October 2024, facilitating a nuanced understanding of the socio-cultural, economic, and structural contexts that shape caregivers' perceptions of childhood pneumonia and their healthcare-seeking behaviors. Through this study design, the use of Martin Heidegger’s hermeneutic phenomenology philosophy, and Bourdieu’s theory of practice, the study sought to move beyond surface-level descriptions by interpreting how caregivers understand illness and practices within their everyday lives.
Study area
The study was conducted in Opanga village, Karemo division, a predominantly Luo community, within the Health Demographic Surveillance System (HDSS) area in Siaya County, Western Kenya. As of 2022, unpublished data reported that the Karemo division had 11,464 children under five years of age, with 1,674 (14.6%) child deaths, 234 (14%) of which were due to pneumonia. The main health facility, the Siaya County Referral Hospital (SCRH), has 200 inpatient beds. In addition, there were 23 health facilities in the Karemo Division. Opanga village, near Siaya Township and close to SCRH and private hospitals, has an estimated population of 5,122.
Study participants
The study participants were caregivers, who included 22 mothers, 18 fathers, and 2 grandmothers of children less than 5 years old in Opanga village. The study team initially screened participants at their homes and asked the following questions: (1) did the child have a cough in the last two weeks, 2) had an illness with a cough, did he/she breathe faster than usual with short, rapid breaths or have difficulty breathing, and 3) was the fast or difficult breathing due to a problem in the chest or to a blocked or runny nose? (Problem in chest only/Blocked or runny nose only/Both/other. 4) Did you seek advice or treatment from any source? (Yes/No/Do not know)
To be eligible, the respondent had to answer “yes” to questions 1, 2 and 4, and ‘a problem in the chest’ to question 3.
The study was conducted over seven-month period. One hundred and seventy-three children under the age of five years were screened using the procedures described above, of whom 23 met the eligibility criteria. Of the eligible 23, 12 caregivers were purposively selected for in-depth interviews, and 11 of them also participated in long-term participant observation. In the sixth month, two additional in-depth interviews were conducted with the two male caregivers. Additional selection criteria required participants to have primary caregiving roles for children under the age of five, reside in the study area, and provide informed consent. Finally, three focus group discussions were conducted within the community. The selection criteria were as follows: primary caregiving roles for children under the age of five, residing in the study area, and providing informed consent, one male-only group, one female-only group, and one mixed-gender group. Detailed characteristics of the participants are provided in Table 1 and Table 2.
Table 1. In-depth and participant observation characteristics.
| Pseudonym | Age | Child’s Age | Ethnicity | Educational
Level | Marital Status | Occupation | Spouse
Occupation | Monthly
income (Ksh) | No. of
Family
members | No. of
Children | Children <5 yrs |
|---|
| Atiela | 26 | 2 | Luo | O level | Single | Business | N/A | 9,000 | 5 | 2 | 1 |
| Okumu | 38 | 4 | Luo | O level | Married | Business | Housewife | 20,000 | 4 | 2 | 1 |
| Pascalia | 39 | 3 | Luo | Primary | Single | Business | N/A | 3,000 | 3 | 2 | 1 |
| Florence* | 33 | 4 | Luo | College | Separated | HRO | N/A | 10,000 | 5 | 4 | 2 |
| Mercy | 24 | 4, 1 | Luo | O level | Married | Online job | Teacher | 8,000 | 4 | 2 | 2 |
| Siri | 22 | 0.5, 3, 5 | Luo | Class 8 | Married | Housewife | Business | Not known | 5 | 3 | 3 |
| Achieng | 29 | 2 | Luo | Bachelor’s
Degree | Married | Teacher | Engineer | 50,000 | 32 | 1 | 1 |
| Atieno | 34 | 4 | Luo | Form Three | Married | Business | Driver | 7,000 | 4 | 2 | 1 |
| Patricia | 26 | 4, 5 | Luo | O level | Married | Business | Business | 20,000 | 4 | 2 | 2 |
| Benta | 24 | 2 | Gisu | Class 6 | Married | Business | Engineer | 9,000 | 3 | 1 | 1 |
| Prisca | 32 | 0.5**, 4** | Luo | Bachelor’s
Degree | Married | Teacher | HRO | 50,000 | 6 | 4 | 4 |
| Akumu | 22 | 2 | Luo | Form 3 | Married | Business | County
employee | 5,000 | 3 | 1 | 1 |
| PatrickP | 25 | 3 | Luo | Form 4 | Married | Student | Housewife | 20,000 | 3 | 1 | 1 |
| OchiengP | 28 | 4 | Luo | College | Married | Business | Business | 10,000 | 4 | 1 | 1 |
Table 2. FGD participant socio-demographic characteristics.
| Characteristics | N28 |
|---|
| Sex | |
| Male | 15 |
| Female | 13 |
| Ages (Years) | |
| 18–24 | 8 |
| 25–34 | 13 |
| 35–44 | 7 |
| Education | |
| Primary incomplete | 3 |
| Primary complete | 6 |
| Secondary incomplete | 5 |
| Secondary complete | 8 |
| College | 6 |
| Ethnicity | |
| Mganda | 2 |
| Luhya | 4 |
| Luo | 21 |
| Teso | 1 |
| Marital Status | |
| Married | 27 |
| Single | 1 |
| Occupation | |
| Informal employment | 6 |
| Formal employment | 1 |
| Self employed | 19 |
| Housewife | 2 |
Data collection
Data collection was conducted between April and October 2024, with in-depth interviews conducted by the lead author at the participants' homes. Most of the interviews were conducted in Swahili. The interviews took approximately 30–50 minutes and were audio-recorded after informed consent was obtained, except for one caregiver who opted out of recording; for this participant, longhand notes were taken instead. Participants were informed of the study objectives before the in-depth interviews and requested to participate in participant observations.
After the in-depth interviews, 11 participants agreed to participate in the participant observation. They were informed of the purpose of the study and provided consent. The caregivers who consented were informed that their entire household would participate in the study.
Participant observation for this research involved spending time with research participants at their homes at least twice a month for approximately six hours at each visit. The activities involved observation and participation in day-to-day childcare activities, including accompanying participants for care-seeking. An observation tool was used to observe and record critical elements, such as the appearance of the home environment and participants, verbal behavior and interaction, physical behavior, care-seeking, and care for the child. Informal interviews were conducted with caregivers during these visits. Research assistants with experience in qualitative methods and social science background were trained on how to note observations and informal conversations.
Finally, during the sixth month of the study, Focus Group Discussions (FGDs) were conducted with community members. This method was used as a data collection technique, in which selected caregivers engaged in in-depth discussions about their health-seeking behavior for pneumonia. This technique was used to clarify the data collected through individual in-depth interviews and informal interviews. The method was useful in clarifying the shared community knowledge about probable pneumonia with regard to the objectives of the study. The FGD interview guide questions were based on thematic areas previously identified in informal and in-depth interviews. The interviews took approximately one hour to one and half hours, and were audio recorded.
The research team consisted of two research assistants: one with a Bachelor of Arts in Sociology and the other with a Bachelor of Arts in Anthropology. The lead researcher is a PhD candidate in anthropology. The research team was trained in the research procedures to ensure consistency and rigor in data collection and analysis.
Ethical considerations
Ethical approval was obtained from the Maseno University Scientific and Ethics Review Committee (MUSERC) ref number MSU/MUSERC/01330/24 and the National Commission of Science, Technology and Innovation (NACOSTI) ref number NACOSTI/P/24/34015. Written informed consent was obtained from all participants enrolled in this study. We used codes to name the audio recordings and stored them on an offline device. All study staff involved in data collection and transcription were trained to handle confidential information. Raw and refined data were stored in a password-protected device. The storage device was kept in a lockable cabinet and was only accessed by authorized staff. No personal identifiers have been used for reporting or publication.
Data management and analysis
Interviews were recorded and transcribed verbatim. Transcripts in the local languages were translated into English. Quality checks involved comparing portions of the transcripts against audio recordings. Comprehensive field notes were written after every participant observation, guided by an observation template that included observation of the appearance of the house and its occupants, verbal behavior and interactions, physical behavior and gestures, and care-seeking and caregiving. In addition, a narrative guide informed by the tenets of the theory of practice was used to facilitate conversation. Field notes and transcripts were uploaded to Dedoose software for data management. Data analysis utilized a phenomenological hermeneutic approach, where transcripts were read multiple times and significant statements were highlighted and coded. These codes were then grouped into themes that captured the essence of participants' experiences. The Hermeneutic Circle was employed, with the researcher engaging in a continuous process of interpreting the data, moving back and forth between the parts (individual statements) and the whole (overall understanding), to deepen the interpretation to get the ‘essence’ of the lived experiences of the participants. Illustrative quotes were used to provide textual descriptions of the qualitative data.
Credibility involved prolonged engagement with informants through participant observation for a period of six months. The multi-method approach facilitated the data triangulation needed to validate the information collected across the two different data sources. To minimize bias, conformability was achieved by reflecting upon interviews and narrative data obtained from FGDs, participant observations, and IDIs before analysis. Two researchers independently analyzed the data and then discussed and reached an agreement. Finally, a homogeneous sample with maximum diversity in demographic characteristics, marital status, education level, and caregiver relationship with the child was ensured to maximize the generalizability of the findings.
Results
Participants’ characteristics
The study enrolled 14 caregivers (11 females and 3 males. One female caregiver was the maternal grandmother of the child for in-depth interviews. Refer to Table 1.
More males (n15) than Females (n13) participated in FGDs. Education levels varied considerably, although most of the participants were from the Luo ethnic group. The majority of participants were either self-employed, such as motorcycle taxi riders (Boda Boda), or informally employed, which included working for other people in the informal sector (jua kali sector). Refer to Table 2.
Caregivers' sociodemographic information was utilized to contextualize and interpret the results, providing deeper insights into the factors influencing their health-seeking behavior.
Thematic findings
Local beliefs and perceptions of pneumonia causation by caregivers
Participants across all categories of data collection methods described a range of local beliefs about what causes pneumonia, including exposure to cold, hereditary factors, and pathogens. The most common cause of pneumonia was exposure to cold, while causation by hereditary factors and pathogens was rarely mentioned across all categories. Caregivers also use a variety of terms and phrases to describe pneumonia and its symptoms.
Exposure to cold
All caregivers in this study had the prevailing notion that pneumonia is caused by exposure to cold, cold water, and cold weather. Caregivers often believed that cold could ‘enter’ the bodies of the children and cause pneumonia if they were not wearing socks, were exposed to cold through their feet, were not dressed warmly enough, or rode on motorbikes and were exposed to cold.
Achieng, a university-educated young mother of a 3-year-old, shared her experience with her son's recent episode of pneumonia. Reflecting on the situation, she attributed her child's illness to his daily use of a motorbike to attend school. In the excerpt below, Achieng explains that she took precautions by dressing her child in warm clothing, believing that this would protect him from the chills of a moving motorbike, especially since her child preferred to sit at the front. However, the motorbike rider strongly objected, insisting that the child should not sit in front of the motorbike because cold air enters through the mouth and nose, potentially causing pneumonia. When the child became sick from pneumonia, Achieng immediately linked the illness to motorbike rides.
I just thought maybe he was playing and maybe injured his back or chest but at the time he [the doctor] said the chest congested] I thought of pneumonia since I take him to school every day on a motorbike and the motorbike rider always said he should not sit in front because he will get pneumonia and I tell him he is putting on a jacket and a sweater with gloves, but he tells me that it is not enough and cold can get through the mouth or nose and most of the time the child insisted on sitting in front and because I do not want morning arguments, I always let him sit in front. When I was told the chestcongestion], what came to my mind was maybe pneumonia. Achieng’, IDI
Atiela, a 26-year-old single mother with four levels of education, believes that a child can develop pneumonia when exposed to cold, especially during the cold season. She believed that the cold could enter through the feet of the child and cause pneumonia. During one of the informal interviews, she shared that she had learned about the cause of pneumonia from her grandmother and other people within the community. She explains her belief in the excerpt below.
I know that it is brought about by the cold. If the child is not kept warm during the cold season, pneumonia can get through even the legs if the child has not worn socks or something like that. Atiela, IDI
Although Okumu, a 39-year-old father of two children, shares the same belief that cold causes pneumonia, he believes that after exposure to cold, the child becomes infected with, or attacked by, pneumonia.
Yeah, cold. When the weather changes or when the weather is too cold, they say that ''Ooh, dress the child well, it will be infected with pneumonia or athung'a [congestion or asthma], so you dress the child well to prevent it from being attacked by pneumonia. Therefore, it is always associated with cold weather and is believed to cause pneumonia. Okumu, IDI
Caregivers shared an understanding that exposure to cold causes pneumonia. The perspective of the mechanism of cold entry to cause pneumonia is conceptualized as something that physically ‘enters’ the body through exposed parts, and settles in the chest to cause pneumonia. Additionally, some caregivers such as Okumu, viewed cold exposure as triggering an "attack" or infection, aligning their cultural understanding with a mix of lay beliefs and biomedical explanations.
Hereditary causes
Only a few caregivers mentioned hereditary causes of pneumonia, and this aspect did not emerge in focus group discussions.
Atieno, a 34-year-old caregiver who did not complete secondary school, explained that during the cold season, her son was ill due to a condition that affected his breathing, but she believed that dust had a worse effect on him. Pneumonia recurs frequently in her child, especially under cold and dusty conditions. She had made several visits to the hospital for breathing problems, which doctors told her of as pneumonia, and each time her child was nebulized. Atieno explains that she and her other son have the same condition.
During the IDIs, Atieno explained that the pneumonia ran on her side of the family. She reiterates that what they have is inherent pneumonia, which recurs occasionally. In the excerpt below, Atieno explains that what her child had was strong pneumonia, meaning that he had severe pneumonia.
I: What has the doctor explained to you is happening to him?
R: They just told me that he has that pneumonia...strong
I: Every time you go...
R: Very much
I: They tell you that he has pneumonia?
R: Exactly that. Because it is something I have, my oldest son has it, and he also has it
Atieno, IDI
During an IDI, Florence, a 33-year-old college-educated mother of four, stated that pneumonia was caused by cold. She thus explains,
‘….and then again it usually comes during the rainy seasons, the cold...so they told me that when evening reaches the child should wear a sweater, that I should not expose him to the cold. Florence, IDI
On probing further during the same interview, Florence explained that she believed that her son’s pneumonia was caused by asthma, which she believed runs in the family. She believed that her child’s pneumonia was directly attributable to his father’s asthmatic condition. She further explained that when she took her child to the hospital with breathing difficulties, the doctor made a diagnosis of pneumonia/asthma. She averred that she did not know if pneumonia and asthma were the same thing. Florence explains her beliefs as follows:
R: But, now, for me, what I was thinking is that maybe it is something that someone inherits from their parents.
I: Ooh in the family?
R: Yes, in the family
I: Have you ever heard a story like that or were you just...
R: No, I heard it from someone. As such, his father had asthma. So, I just knew that it was that... like maybe, it was something that he inherited from his father. So, I do not know if pneumonia and asthma are the same thing. I do not know. Florence, IDI
Pathogens
During FGDs, caregivers have reported that bacteria can cause pneumonia. Most of them were not sure of this fact and requested further information. Okumu, a 39-year-old father of two and a form four leaver, was the only caregiver who participated in participant observation and attempted biomedical explanation of pneumonia. Okumu explained that pneumonia can be caused by a virus:
Whatever I know...pneumonia is a what...I think it is a virus...it is an infection. I think this is a viral infection. When it attacks somebody, I know is always dealing with breathing and it is always brought about by cold weather. Okumu, IDI
He explained that viruses hibernate in cold water, and under the same conditions, the environment allows for the growth of viruses. So, when his child got sick with pneumonia, he immediately knew that his child had taken cold water.
Like taking cold things, like after drinking sugary things you know...you know these cold things like water, any cold water they...it makes some virus...some viruses hibernate in cold things. So, I knew maybe this child has been taking maybe a lot of a cold thing or the...the drinks, you know some of the drinks during fermentation they give room for the viruses to grow because some...most of like drinks they are fermented and you know when something is fermented it is like there is poison inside it. Yeah, you know that fermentation brings about poisoning. So, I knew may be through this cold thing a virus...some viruses mutate...mutate through the cold. Okumu, IDI.
Okumu explained that his child had a persistent wet, rapid, and irritating cough, especially at night, which kept him up. He immediately knew that the cough was caused by the infection, and the infections were always treated with antibiotics. Okumu further narrates that he believes that cough is caused by a virus, which led him to conclude that his child had viral pneumonia. Acting on his own understanding, he gave his child leftover azithromycin, which he explained was meant for an adult, believing it was stronger and more effective.
Okumu has prior experience working in a research organization and maintaining friendships with medical professionals, which has contributed to his relatively advanced knowledge about pneumonia compared to other caregivers. However, Okumu’s knowledge of pneumonia causation is mostly inaccurate. Okumu accurately referred to pneumonia as a viral infection, although his knowledge about treatment was incorrect, whereby he treated pneumonia, which he believed was caused by a virus, with an antibiotic. Okumu’s actions highlight the idea that pneumonia remains a foreign and poorly understood concept.
Focus group discussions revealed that pneumonia can be caused by germs, although the majority of them did not have any knowledge about germs that cause pneumonia.
I know it is caused by cold, but there is a time I heard from the radio that it is caused by bacteria, but I am not so sure that you will give me a response to that. FGD, Female group
Terms and phrases used to describe pneumonia
Caregivers who participated in IDIs, informal interviews, and FGDs used diverse terminology and phrases to describe the signs and symptoms of pneumonia. The varying terminologies were ‘athung’a’ (meaning “congestion”), kore thung'/ridore (chest congestion), ang’ieu’ (a word used to refer to measles) or ‘kahera’ (tuberculosis), tuo koyo or midekre mag koyo (illness of the cold), and homa mar kor (illness of the chest).
Of the 14 caregivers who participated in the in-depth interviews, nine were unaware of any name for pneumonia other than the English term ‘pneumonia.’ Similarly, the majority of participants in the focus group discussions (FGDs) shared the same lack of knowledge.
Atiela, a young single mother, expressed unfamiliarity with any other term for pneumonia. The excerpt below expresses her unfamiliarity with pneumonia terminology:
I: In this community, what do you call pneumonia commonly?
R: I just know it as pneumonia
I: There is no other that you have heard people saying?
R: No
Okumu, on the other hand, is familiar with other terms describing pneumonia in Dholuo. He believes that different people have various ways of using terms that refer to their symptoms. For instance, he mentions the term athung’a which means congestion. He says that they call this measles or ang’ieu in Dholuo. Upon further probing, Okumu explains that the community uses the term ang’ieu to refer to both measles and pneumonia. Dual naming arises because measles presents with symptoms similar to those of athung’a, such as runny nose, chest congestion, or difficulty in breathing and coughing, which are similar to those of pneumonia.
Benta, a 24-year-old mother, shared that she did not know the local name of pneumonia. She explained that in her native community, pneumonia is referred to as Kwashiorkor. She reiterates that this is the term that her community uses to describe pneumonia. Similar to Okumu, she says that another terminology used for pneumonia is measles, using English to describe both terms.
In a mixed-gender FGD, one caregiver mentioned that pneumonia is referred to as athung’a- a term that is also used to describe asthma. The caregiver shares that she is not sure if athung’a refers specifically to asthma or if it includes pneumonia. She says, “I’ve heard some people calling it athung’a but I’m not sure if athunga is asthma.”
Siri, a 22-year-old mother of three, was not familiar with the term for pneumonia in her local language. Additionally, she is not sure of the illness itself, as she refers to pneumonia as the disease that causes blood to run low. She says that she is unfamiliar with the terminology because she is not conversant with the Luo language.
Terms and phrases used to describe difficulty in breathing
Difficulty in breathing was not mentioned spontaneously by the caregivers as a symptom of pneumonia. Instead, caregivers often conflated it with terms used to describe chest congestion, such as athung’a, kore thung’, or ridore. These local terms generally refer to congestion or tightness in the chest rather than breathlessness. The study observed that caregivers indicated that they perceived difficulty in breathing as a problem in the chest, but struggled to articulate the symptoms clearly. For instance, Okumu had difficulty describing the symptoms; therefore, he described breathing difficulties using sounds as illustrations, despite being multilingual (speaking English, Dholuo, and Swahili). He thus explained:
The child...when the child breathes, you get that... she loses air. This is not normal. Yeah, there is a sound that it produces like... [Imitates breathing] like someone siphoning something or a trumpet or like..., which is very abnormal. There was no smooth breathing. Okumu, IDI
Similarly, Atiela, another caregiver, expressed uncertainty over whether her child had pneumonia based on how she breathed while sleeping.
In the past, I used to suspect, because she (daughter) used to have...but I do not think she had it. She used to breathe as if she wanted to...like when she was inhaling [imitates heavy breathing] while she was sleeping. So I wondered if my child had pneumonia or if she was okay. Atiela, IDI
Discussion
Our findings showed that caregivers had a shared perception of pneumonia, indicating a common understanding derived from their cultural and social capital, which is a reflection of the environment in which they live. Similarly, past literature has detailed a range of culturally specific beliefs about pneumonia in various contexts10,27,37. These findings highlight the diversity in community-level perceptions of pneumonia causation across different regions influenced by social and cultural beliefs. In this study, we present a nuanced understanding of perceptions of pneumonia causation and cultural understanding of pneumonia, which includes the terminology of pneumonia and pneumonia symptoms.
Our study revealed that caregivers had to navigate competing explanations of pneumonia, balancing traditional cultural knowledge with an emerging, though limited, biomedical understanding. While traditional beliefs about the causes of pneumonia are prominent, biomedical explanations are gaining traction- supplementing rather than replacing traditional beliefs. The study findings indicate the dynamic nature of caregivers’ knowledge as they integrate new information, such as acknowledging infections into their existing frameworks and how cold causes pneumonia. However, the predominance of cold exposure in caregivers’ explanations demonstrates the resilience of culturally ingrained beliefs even when exposed to alternative explanations. For example, generational knowledge of the effects of cold weather on pneumonia reinforces the idea of protecting children with warm clothing. Their disposition is embedded in their collective worldview and everyday practices, and explains why this belief persists across caregivers of varied educational levels and social statuses.
Pneumonia causation through cold exposure has been documented in various studies across sub-Saharan Africa and South Asia. These studies consistently revealed that caregivers strongly believed that pneumonia was caused by cold weather, a belief that was carried from one generation to the next9,10,27,34,38,39. However, in the current study, consumption of cold foods and drinks as a cause of pneumonia was not established. In Lagos, Nigeria, most caregivers noted the dangers of exposing children to cold, specifically mentioning bathing practices, either at the wrong time of the day or with cold water33. In contrast, in Mozambique, the cause of an episode of probable pneumonia was sometimes unknown to caregivers for various reasons that were attributed to probable pneumonia. In this context, pneumonia was caused by disregarding prescribed sexual practices, asthma, and cold weather11. Despite these past studies' failure to use a theoretical framework, their findings align with Bourdieu's concept of cultural capital, particularly in how caregivers’ knowledge and beliefs are shaped by their social and cultural context. This helps to explain the predominance of exposure to cold as a cause of pneumonia across different communities.
Exposure to cold as a cause of pneumonia was constructed and negotiated within the cultural context of the study’s community10. This aligns with Bakare et al., who posited that illness knowledge is socially constructed37. Within this context, pneumonia is attributed to natural causes that are shaped by cultural beliefs. These natural causes of ill health were constructed from the effects of the physical environment, such as exposure to cold and infections9,10,27,34,38,39. Although infection is recognized to some extent as a cause of pneumonia, this aspect has not been fully explored in studies conducted with community members. Past literature similar to the present study does not fully reconcile biomedical explanations of pneumonia with deeply ingrained cultural beliefs about the cold as a primary cause.
According to Colvin et al., there are different theories of causation, such as humoral models, which describe illnesses resulting from an imbalance in natural forces, such as excessive heat or cold10. Other theories of illness causation are extramarital affairs and pollution, which are believed to cause diarrhea in children, while convulsions are believed to be caused by supernatural forces. Biomedical theories of health coexist and operate in complex and dynamic relationships with other local theories of health10. For example, caregivers acknowledged that a sunken fontanelle was linked to diarrhea, but misunderstood the causality, attributing it to a condition known as nhoua or chipande, which they believed was caused by the sunken fontanelle. However, in this study, pneumonia was attributed to natural causes, specifically environmental factors and infections, which were viewed as complementary rather than contradictory causes.
This study revealed an interplay between asthma as a cause of pneumonia and hereditary pneumonia. Participants navigated two distinct beliefs: that pneumonia is caused by cold weather, a socially constructed idea, and that it is linked to asthma, learned through personal experiences and social contexts. The distinction between asthma and pneumonia was unclear for the participants. The overlap in symptoms between the two conditions, particularly breathing problems, may contribute to this confusion. Without clear guidance from healthcare professionals, caregivers may struggle to differentiate between asthma (a chronic condition often influenced by genetic factors) and pneumonia. This confusion can lead to misattributions, where symptoms of one condition are mistaken for those of the other, potentially impacting the child’s care and treatment.
Previous studies have identified asthma as a perceived cause of pneumonia11,27. However, the findings from this study suggest a more nuanced relationship, which highlights parental asthma as a hereditary condition that may directly contribute to the causation of pneumonia in children. While caregivers in Abbey et al.'s27 study viewed the cause of pneumonia as straightforward, the present study revealed a more complex causal interaction that confuses caregivers. Caregivers who mentioned parental asthma as a cause of pneumonia were uncertain whether pneumonia was the same as asthma or whether asthma directly caused it, reflecting a lack of clarity regarding the connection between the two conditions. Similar overlaps between pneumonia and asthma were reported by King et al. (2022)40. In a study conducted in Malawi, caregivers often mentioned that their children were ‘born with asthma’. Additionally, there was frequent uncertainty regarding whether their children were suffering from asthma or pneumonia. This ambiguity was also evident in the current study, as caregivers’ descriptions made it difficult to determine whether the child had asthma or pneumonia or not. This confusion can be attributed to the lack of embedded knowledge about asthma and pneumonia, which despite being clinically distinct, may present overlapping symptoms that are difficult for caregivers to differentiate.
Explanations of pneumonia, such as its association with viral infections, reflect an attempt to apply biomedical reasoning. However, this understanding is incomplete and interwoven with local beliefs about illness causation, such as the idea that viruses hibernate in cold water or mutate through cold drinks. These beliefs demonstrate how caregivers’ understanding is influenced by a combination of biomedical knowledge and sociocultural perceptions mediated by their social environment. However, their sociocultural knowledge has taken a predominance in their explanation of pneumonia. The narratives reflect the influence of ingrained dispositions, shaped by limited formal education and cultural beliefs, along with access to social capital through interactions with medical professionals. Caregivers’ limited knowledge of the biomedical causes of pneumonia was similar to findings from previous studies28,34,37,41, where the majority of the study population was unaware of the biomedical causes of pneumonia. Inaccurate knowledge of the causes of pneumonia can have implications for prevention and actions taken by caregivers when a child becomes ill with pneumonia27. While some studies have identified poor biomedical knowledge of pneumonia, the underlying issue appears to be the disconnect between symptom recognition and recognition of its link to pneumonia. Therefore, the knowledge gap means that caregivers are unable to apply knowledge to effectively identify, understand, and address the illness in real-life situations.
Caregivers provided terms based on the perceived causation of illness. For instance, a few caregivers referred to pneumonia as tuo koyo (illness of the cold). The findings of this study are consistent with those of Kilimanjaro, Tanzania, Ghana, and Central Kenya27,42,43, where caregivers described pneumonia symptoms literally but did not have a local name for the illness. For instance, kore thung'/ridore refers to a congested chest that reflects a symptom-based understanding of the illness. These findings reveal how local terminology often focuses on visible symptoms. Hildenwell highlighted the complexity of respiratory symptom terminology, noting that many terms were literal translations that did not necessarily focus on visible terminologies of respiratory symptoms, which were mostly literal translations that did not focus on visible symptoms. For instance ‘enhonhi’ which literally means ‘bird’ was used because a sick child’s hand and foot positioning resembled a bird-reflecting the belief that the illness is caused by birds41.
A study conducted in Mozambique has revealed a similar understanding of respiratory illnesses. In their context, Xifuva and tosse most closely resonated with respiratory illnesses: tosse, an illness itself, or a symptom of another illness such as xifuva (chest)11. An overlap was also observed in this study, whereby one term was used to describe an illness as well as a symptom of an illness. For instance, athung’a is an illness as well as a symptom of measles. These terminological overlaps point to a complex cultural and linguistic framework for understanding illness. Past literature11,27,41–43 and the current findings align with Bourdieu’s concept of cultural capital, which shapes current knowledge and practice. The lack of distinct pneumonia terminologies indicates a limitation in the cultural practices surrounding pneumonia disease classification and recognition.
The use of ang’ieu to refer to pneumonia has significant implications for health-seeking behaviors. As ang’ieu is traditionally associated with measles, caregivers may conflate the two illnesses and potentially misinterpret the severity or required treatment for pneumonia. In a previous study in Western Kenya, measles was perceived as an illness that should not be treated with hospital medications, particularly injections, because of the cultural belief that such treatments may result in death44. When caregivers refer to pneumonia as ang’ieu, chances are that the child will be treated with traditional medicine, which complicates its management and outcomes.
Caregivers’ understanding of breathing difficulties reveals a significant limitation in their recognition and interpretation of symptoms. Key clinical concepts such as fast breathing and chest wall in-drawing, which are essential indicators of difficulty in breathing, remain largely unknown or poorly understood. Our findings are similar to those of previous studies, which also found that there was no specific word to describe breathing difficulty27,43. However, Hildenwall et al. reports a literal translation ‘tawera bulundji’ to mean ‘difficult breathing’ that was used to refer to difficulty in breathing. They acknowledged difficulties for caregivers in establishing the term that refers to fast breathing and chest-wall in-drawing as caregivers were only able to provide terminology after viewing a video of a child with difficulty in breathing41. These findings point to a limitation of linguistic cultural capital, which constrains the present linguistic knowledge about critical pneumonia symptoms.
Study limitations
This study has a few limitations. There was a possibility of biased or skewed findings due to the subjective nature of the responses and partly because of participants’ desire to give socially acceptable answers. However, the presence of the researcher within the community for six months influenced the trusting relationship between caregivers and the researcher. This enabled the researcher to verify the information obtained from study participants. Furthermore, owing to the interactive nature and observations made during the study, the researcher was able to validate earlier responses provided by the study participants. The study conducted sufficient training for the research assistants. These strategies were sufficient to reduce potential biases in the data collection, analysis, and interpretation. The study used qualitative research methods; therefore, the findings from this research cannot be used to make inferences and causality about relationships in observed variables.
Conclusion
The findings highlight that caregivers’ limited ability to interpret pneumonia symptoms is rooted in the lack of a cultural and linguistic framework for the illness. The absence of a widely recognized term for pneumonia in the Luo language and the conflation of respiratory symptoms have contributed to this challenge. Poor understanding of critical clinical signs, such as chest wall in-drawing and difficulty in breathing, may shape how caregivers perceive and recognize this illness. To minimize the gap between biomedical and culturally mediated knowledge, culturally sensitive health education programs that emphasize recognition of key pneumonia symptoms such as chest wall in-drawing, rapid breathing, and difficulty in breathing using locally understandable language and concepts should be developed and initiated. These programs should be integrated into existing community health structures and tailored to address the current gaps in the cultural and linguistic frameworks of pneumonia.
Disclaimer
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors or funder. The authors are responsible for the results, findings, and content of this article.
Ethics and consent
Ethical approval was obtained from the Maseno University Scientific and Ethics Review Committee (MUSERC) ref number MSU/MUSERC/01330/24 and the National Commission of Science, Technology and Innovation (NACOSTI) ref number NACOSTI/P/24/34015. Written informed consent was obtained from all participants enrolled in this study.
