The authors have declared that no competing interests exist.
‡ These authors also contributed equally to this work.
Influenza virus infections cause between 291 243 and 645 832 deaths annually, with the highest burden in low-income settings. Research in high-income countries has examined public understanding of influenza, but there is little information on views and behaviours about influenza in low-income countries. We explored communities’ ideas about the severity, causes, prevention and treatment of influenza in Chikwawa district, Malawi. We conducted 64 in-depth interviews with parents of children aged <5 years, and 7 focus groups with community health workers, parents, and traditional healers. Data were analysed thematically and using a framework matrix to compare views between groups. Respondents held varied ideas about influenza, and many were uncertain about its causes and treatment. Some parents, traditional healers and health workers thought influenza was not severe because they felt it did not cause death or limit activities, but others disagreed. Many saw influenza as a symptom of other conditions, especially malaria and pneumonia, rather than as a disease of its own. Most mentioned dust as the main cause of influenza and believed influenza could be prevented by cleaning the home thoroughly. Treatment seeking for influenza followed different stages, usually starting with home remedies followed by purchasing drugs from groceries and then visiting a health centre. Seeking a clinician tended to be triggered by severe symptoms like high fever or difficulty breathing, and suspicions of malaria or pneumonia. Community health workers provide health education for communities, but some lacked understanding of influenza. Our findings suggest uncertainty about the causes and control of influenza among parents and varied levels of understanding among health providers. Strengthening the capacity of community health workers to provide relevant information about influenza prevention and treatment could address parents’ interest in further information and support informed health seeking and engagement with future influenza interventions.
Influenza is a global health concern, causing between 291 243 and 645 832 deaths annually [
Understanding public views about influenza is crucial for developing effective control strategies [
This paper examines community perceptions and behaviour about seasonal influenza in Malawi, a low-income country in Southern Africa. Preliminary research suggests knowledge and beliefs about influenza vary considerably in Malawi [
This study took place in Chikwawa, a rural district in Southern Malawi with an under-five mortality rate of 66 per 1000 live births [
Our study was conducted in parallel to a community-based Phase IV trial examining the impact of malaria exposure on immune response to influenza vaccination in children aged 6–59 months [
Data were collected over 10 months from March 2016. We conducted in-depth interviews with parents of study-eligible children in 41 households. We purposefully sampled parents of study-eligible children based on consent or refusal to enrol their children in the vaccine trial, and from villages with varied health centre access. We also conducted a further 20 in-depth interviews using critical incident narratives (CIN) with parents who had brought a child aged less than five years to a health centre with either severe acute respiratory infection or influenza-like illness. We used CINs to provide specific information about decisions that parents made in response to a recent experience of an influenza-like illness. CIN interviewees were purposefully selected to include those whose children had different severe acute respiratory infections and influenza like illnesses, and attending different levels of health centre (village, primary and secondary).
Seven focus group discussions (FGDs), each with 9–12 participants, were held: four with parents of children aged less than five years (split between men and women, and between parents from more socially included and excluded households because health seeking can vary with social economic status [
All interviews and FGDs included discussion about understanding of influenza: perceived severity, symptoms, causes, treatment and overlap with other conditions. In interviews, we asked about general approaches to management of influenza, and about specific recent individual episodes of illnesses that participants attributed to influenza. CIN asked parents to recount details of a recent illness episode, from initial recognition of symptoms to arriving at a health centre [
Data were collected by two authors (MP and SN) with support from a social scientist (KG). Debriefing meetings were held soon after every data collection episode to review emerging findings and identify areas needing further investigation. Interviews lasted 26–105 minutes (with most being 45–60 minutes), and FGDs averaged 90 minutes (ranging from 89–116 minutes). All interviews and FGDs were conducted in the local language (Chichewa) and audio recorded.
Audio recordings were transcribed verbatim, translated into English, and imported into NVIVO10 software (QSR, Melbourne, Australia) to facilitate organization and analysis. Transcripts were read and re-read for familiarization with the data. Transcripts were coded inductively. KG, MP and SN each coded an initial sample of transcripts, before comparing interpretations. After comparing the interpretations, we maintained some codes and merged others, then generated an integrated coding frame that was used to code further transcripts, iteratively adapting the frame as new data were analysed. A framework matrix was developed to compare views among parents, and between parents, community health workers and traditional healers.
Every participant provided written informed consent. Ethical approval for the study was obtained from the Malawi College of Medicine (16–004) and Liverpool School of Tropical Medicine (P.11/15/1828) research ethics committees. The U.S. Centers for Disease Control and Prevention relied on the local review by the Malawi College of Medicine.
In examining perceptions of influenza and associated health seeking, we identified themes related to symptoms, severity, causes, prevention and treatment. In discussing influenza with participants, we used the Chichewa word ‘chimfine’, which is the usual translation of influenza in Chichewa [
Parents, community health workers and traditional healers all commonly described influenza as a condition that makes people sneeze and unable to breathe normally.
Participants also distinguished influenza symptoms from those of conditions such as malaria, pneumonia or cough (malungo, chibayo or chifuwa, respectively, in Chichewa), citing nasal congestion, sneezing and a runny nose as more associated with influenza.
However, other participants identified overlaps in symptoms between influenza and other conditions. For example, some parents and community health workers saw fever and body pains as common to both influenza and malaria, and fever and difficulty breathing as typically associated with influenza and pneumonia.
Among those parents who described overlapping symptoms for influenza and other illnesses, some identified influenza and other conditions as separate diseases but thought only a doctor could distinguish them.
However, for other parents this overlap in symptoms suggested that influenza was an early indicator of other conditions such as malaria and pneumonia, rather than a disease in its own right.
Those who saw influenza as an indicator of other conditions said they usually had influenza before diagnosis with malaria, leading to the perceived association:
Similarly, some community health workers believed influenza was a symptom of malaria because
However, while some parents described these understandings of influenza as a disease and how it is recognized, others indicated doubts about what influenza really was.
Parents described several types of influenza. Common flu (termed Chimfine cha wamba) was the kind that parents said normally made them sick. Several parents also mentioned bird flu (Chimfine cha mbalame) and swine flu (Chimfine cha nkhumba), but said they had only heard about these conditions and not experienced them.
Some parents also described malaria-like flu (“Chimfine cha malungo”), which referred to a form of influenza that resembled malaria in symptoms, and that could develop into malaria.
Perspectives on severity of influenza varied between respondents. Some parents felt influenza was not as dangerous as other conditions, such as malaria, measles and pneumonia. This was primarily because parents felt influenza did not seriously affect daily activities.
Some traditional healers shared this view that influenza was not serious because it had less impact on activities than other illnesses:
Alongside the minimal impacts on activities, some parents felt influenza was less serious than other illnesses because they perceived it as not causing death. Death was only considered a risk through other conditions related to influenza.
However, other parents felt influenza was just as serious as other conditions, significantly affecting activities or bringing a risk of death. In relation to impact on activities, some said influenza affected work either directly, or through caring responsibilities for others who are sick.
In relation to health risks, difficulty breathing due to nasal congestion was the main danger that parents associated with influenza.
While parents had different views about severity, nearly all health workers said influenza was dangerous, particularly because of its impact on breathing and potential to progress to other conditions.
Although most health workers thought potential to cause other conditions made influenza dangerous, others saw influenza as only dangerous in combination with other conditions.
Participants saw severity of influenza as varying between groups. Most parents and health workers thought influenza was particularly dangerous for infants because difficulty breathing due to influenza limited sucking and eating. Health workers also mentioned a risk that breathing problems could cause death for young children.
Older people, pregnant women and people with pre-existing health problems were also seen as at particular risk from influenza. Parents and health workers said older people were vulnerable to influenza because their natural defence against infection was impaired.
Health workers stated that people who already had serious conditions like cancer would find it harder to recover from influenza, and so would pregnant women:
Participants had several ideas about causes of influenza, and the same individuals often mentioned multiple causes. The most commonly identified cause was dust, often reflecting the experience of sneezing after inhaling dust:
Linked partly with the idea of dust, people also thought influenza was caused by lack of hygiene, which they defined as not cleaning the house or not washing clothes or beddings.
While most participants saw dust as causing influenza directly, others linked the role of dust to germs. They used the Chichewa word ‘kachirombo’, which covers viruses but also bacteria and parasites (hence our use of the broad term ‘germs’ in translation). A few parents thought germs responsible for influenza were in the dust, a view shared by one traditional healer and health worker.
Another cause identified by some participants was the idea of influenza as spread from other people, often expressed in relation to proximity and congestion.
Transmission between people was only explicitly linked to spread of germs by the more senior health worker (who used the English term virus), and one parent and traditional healer.
A further cause of influenza identified by many participants was the weather. However, there were different ideas about what kind of weather caused influenza, and about how different kinds of weather led to influenza. Many people associated influenza with winter (May–July in Malawi), when the weather is cold and dry, with typical temperatures of around 4 to 10 degrees Celsius [
This association with cold weather was partly because of an idea that being cold could in itself contribute to influenza.
However, the link between cold weather and influenza also related to the earlier idea of dust causing influenza because dry and windy conditions during the cold season meant increased levels of dust:
Cold weather was also linked to influenza through the idea of influenza as spread between people. One traditional healer suggested that cold weather made people lie closer together, fuelling the spread of influenza.
In contrast, others linked influenza to hot weather and saw the heat of the sun as leading to influenza.
Beyond these immediate causes, some people pointed to wider contextual factors that contributed to influenza, for example housing conditions were seen as increasing the risk of person-to-person contagion.
Although participants described these causes of influenza, as with the understanding about what influenza is, many parents indicated uncertainty about what causes influenza.
Methods of preventing influenza often followed perceived causes. For example, those who identified dust as a cause tended to mention steps to prevent exposure to dust, such as dampening earth floors before sweeping, regular maintenance of floors, not working in dusty conditions and keeping children away from dust. These strategies were mentioned by both parents and health workers.
As well as cleaning to remove dust, most people saw cleaning and hygiene as important for preventing influenza more generally, and this included laundry and bathing.
Linked with the idea of cold weather as leading to influenza, some parents felt wearing warm clothes could prevent influenza.
Regarding the idea of influenza as spread between people, a few parents mentioned keeping away from people infected with influenza, and one mentioned washing hands after nose-blowing.
Similarly, one health worker discussed advising people to cover their mouths when sneezing as a person-to-person transmission control measure.
In line with ideas about influenza as a symptom of other conditions, some health workers felt influenza could be prevented indirectly by preventing those other conditions, for example, through vaccination against pneumonia.
While some parents identified ways to prevent influenza, others felt incapable of preventing the disease. For some, this related to a feeling that transmission could not be stopped:
For others, the feeling that influenza could not be prevented stemmed from uncertainty about its causes and a perceived lack of information:
Participants reported several approaches for managing and treating influenza. Some parents visited a health centre, especially when influenza symptoms were severe, and often because of a suspicion that influenza may have progressed to other conditions perceived as more serious. This approach was supported by health workers. If influenza presented with high fever for instance, suspected malaria was often the reason for going to hospital. Respondents used the Chichewa term ‘chipatala’, which is translated as hospital but can also refer to primary health centres.
For situations when influenza was less severe, parents described using home remedies such as bathing in warm water, holding a warm cloth on the nose, taking a salt solution, and frequent fluid intake.
Health workers and traditional healers also mentioned drinking water.
…
There were different ideas about how drinking water could help. One set of ideas related to having enough water in the body, including the idea of cleaning organs mentioned by this healer above, and other ideas related to water dissolving things or through effects on the blood. Other ideas related to respiratory symptoms, including reducing itching or dryness in the throat and opening the nose.
Another home remedy, discussed by a small number of parents and some traditional healers, was using herbs to help manage influenza by clearing the nose.
Some participants mentioned going to traditional healers, especially when something originally presumed to be influenza persisted and so they suspected witchcraft instead.
Traditional healers also reported people visiting them about influenza, usually after having gone to the hospital.
However, most parents said they did not visit traditional healers, and perceived them as unreliable and causing conflicts in households.
As well as home remedies, many parents mentioned purchasing drugs from the grocery, such as paracetamol and antibiotics. Health workers also mentioned that parents often took antibiotics before seeing a clinician.
As the first parent’s comment suggests, paracetamol was seen as relieving the effects of influenza, rather than providing a cure. This view was linked to an idea that influenza has no treatment, described in relation to both drug and non-drug options.
Previous advice from health centres or clinicians contributed to this idea that influenza has no treatment, and encouraged home management. Parents and some health workers described clinicians advising patients that there was no treatment, and only prescribing paracetamol.
…
Only the more senior health worker explicitly linked the lack of treatment to the idea of influenza as caused by a virus.
Related to the idea of influenza as having no treatment, people talked about influenza as self-limiting, saying “it ends on its own”. This contributed to some parents saying they just wait when they have influenza, rather than looking for any medication or treatment.
The uncertainty seen in relation to causes and prevention of influenza was also found in relation to treatment. Some parents and health workers described not knowing how to treat influenza.
Most parents mentioned friends and relatives, especially elder parents, as the usual source of information about influenza, including information on home-based management.
Parents mentioned health workers as another source of information on health matters, both through hospital visits and public health information sessions in the community. However, most parents felt health workers prioritized information about other diseases and did not disseminate information about influenza.
This lack of information may have contributed to the uncertainty about influenza indicated in previous sections.
Parents had not heard about an influenza vaccine before learning about it through the influenza vaccine trial or our meetings with them. Having learnt about it through the trial, most parents were very positive about the vaccine and described vaccination as providing protection and as the most effective means of stopping influenza.
Health workers were confident that communities would welcome an influenza vaccine. However, they felt there should be adequate sensitization so that people had information before the vaccine arrived.
There was also demand for adults to be vaccinated, not just children:
We found a broad spectrum of community views about the nature of influenza, including symptoms, types, severity, causes, prevention and treatment, and areas of both overlap and difference with biomedical understandings. Here we summarise these views and compare them to findings from other countries and biomedical understandings.
Participants often defined influenza as a condition characterized by nasal congestion or difficulty breathing, consistent with findings from South Africa [
In relation to severity, some saw influenza as less dangerous than other conditions, because it was not fatal or did not disrupt work; similar to perceptions in Kenya [
In relation to causes of influenza, most people discussed dust as the cause, but hot and cold weather, poor hygiene and transmission between people were also mentioned. The association of dust or dirt with influenza was also found in Kenya and South Africa [
Approaches to preventing influenza sometimes followed perceived causes, supporting the idea that health seeking reflects understandings of etiology [
Treatment approaches included various home remedies such as hot water or herbs, with hospital visits made primarily when symptoms suggested potential malaria or pneumonia; similar findings are reported in Kenya [
Participants’ uncertainty about influenza and varied understanding of severity suggest that providing information for primary health workers and communities could support more confident influenza management. Health education could build on existing understanding. For example, while many participants described hygiene as a way to prevent influenza, this related primarily to household cleanliness, with only one participant mentioning hand washing and little attention to other steps to prevent spread of infection such as covering a sneeze. Beyond education, views from our participants and other research indicate the importance of addressing social determinants of health, such as overcrowded housing, within influenza control strategies [
The study had limitations. First, the broad definition of chimfine, encompassing influenza but also a runny nose, cold and sneezing more generally, complicates interpretation of the findings; parents may see dust as a cause of sneezing rather than of influenza. Research in English-speaking countries shows that the term ‘flu’ is used similarly, with the common cold or gastrointestinal symptoms sometimes described as flu [
Our study provides the first detailed description of views about influenza in Malawi, and contributes to the literature on perceptions of influenza in low-income settings. Programmatically, the findings indicate the potential value of incorporating information about influenza within health education messages and providing information about influenza for community health workers. Our findings also suggest high acceptability of influenza vaccination for children, but acceptability of vaccination for other target groups including pregnant women and people with HIV requires further research. Finally, our research suggests the need for careful investigation of the broad meaning of chimfine and investigation of terminology used for influenza, to support clear communication between researchers, clinicians and communities.
(DOCX)
We would like to thank the study participants for their contributions through participation and the FLUVAC study for permission to speak to their participants.