Non-communicable diseases have risen markedly over the last decade. A phenomenon that was mainly endemic in high-income countries has now visibly encroached on low and middle-income settings. A major contributor to this is a shift towards unhealthy dietary behavior. This study aimed to examine the complex interplay between people’s characteristics and the environment to understand how these influenced food choices and practices in Western Kenya.
This study used semi-structured guides to conduct in-depth interviews and focus group discussions with both male and female members of the community, across various socioeconomic groups, from Kisumu and Homa Bay Counties to further understand their perspectives on the influences of dietary behavior. Voice data was captured using digital voice recorders, transcribed verbatim, and translated to English. Data analysis adopted an exploratory and inductive analysis approach. Coded responses were analyzed using NVIVO 12 PRO software.
Intrapersonal levels of influence included: Age, the nutritional value of food, occupation, perceived satiety of some foods as opposed to others, religion, and medical reasons. The majority of the participants mentioned location as the main source of influence at the community level reflected by the regional staple foodscape. Others include seasonality of produce, social pressure, and availability of food in the market. Pricing of food and distance to food markets was mentioned as the major macro-level influence. This was followed by an increase in population and road infrastructure.
This study demonstrated that understanding dietary preferences are complex. Future interventions should not only consider intrapersonal and interpersonal influences when aiming to promote healthy eating among communities but also need to target the community and macro environments. This means that nutrition promotion strategies should focus on multiple levels of influence that broaden options for interventions. However, government interventions in addressing food access, affordability, and marketing remain essential to any significant change.
The online version contains supplementary material available at
People must eat to live, but beyond this basic biological function, food forms an integral part of our daily lives. Food consumption has evolved into a multifaceted social instrument. It is a principal social and cultural activity that people enjoy for aesthetic or communal reasons while connecting people across cultural boundaries [
Globally, non-communicable diseases (NCDs) are the leading cause of death. According to WHO estimates, these diseases contributed to 36 million deaths globally in 2008, accounting for 63% of 57 million total deaths [
Economic development in LMICs together with recent technological innovations and modern marketing techniques have modified dietary preferences. This has led to major changes in the composition of diet which contributes to the prevalence of NCDs [
This is particularly true in Kenya where the middle-class boom has resulted in a larger market for processed foods from supermarkets and a decline in fresh foods available in traditional markets. Supermarkets in urban Kenya have risen from a tiny niche a half-decade ago to a fifth of food retail, spreading well beyond the richer consumers to derive more than a third of their sales and half of their customers from low income and poor consumers. The United Nations Sustainable Development Goal (SDG) 2 seeks “to end hunger, achieve food and nutrition security, improve nutrition, and promote sustainable agriculture” [
The analysis described here draws on baseline findings from a larger ongoing mixed-method natural experimental study evaluating the impacts of a new hypermarket (supermarket combined with a department store) on dietary behavior and the local foodscape in Western Kenya [
The ecological model was adopted in the formulation of data collection tools. This model recognizes the complex interplay that exists between an individual and the various levels of interaction with the environment [
The study was conducted in Kisumu and Homa Bay counties, in Western Kenya. These settings have a population of 1,155,574, and 1,131,950, respectively [
This was a cross-sectional qualitative study involving members of households who participated in the initial quantitative household survey [
With the establishment of primary health care networks (PCNs) and subsequent implementation of the Kenya Primary Health Care Strategic Framework 2019 – 2024 [
Using the saturation model for qualitative data [
The focus group discussion guides (Additional file 1: Appendix
The study recruited experienced qualitative data collectors of bachelor’s degree level. Prior to the commencement of the study, a three-day training was conducted on understanding the study aims. A refresher training was also offered on IDI and focus group discussion techniques and the discussion guides. After the training, the tools were piloted in both sites and necessary adjustments were made. Both the FGDs and IDIs were conducted by a moderator who was in charge of steering the conversation, and a notetaker who took notes verbatim and in addition captured the non-verbal cues during the discussion. The interviews took an average of one and a half hours each for the FGDs and close to forty minutes for the IDIs. They were conducted in either Swahili or Dholuo. These discussions were held at local venues such as classrooms, community, and church halls and offices. The discussions were recorded using an audio recorder.
The thematic analysis used in this study was informed by the blended approach to coding described by Graebner [
The study recruited 33 and 38 participants in Kisumu and Homa Bay Counties respectively giving a total of 71 respondents. This group constituted of males and females aged 20–69 years from different socio-economic groups. The majority of the participants constituted those between ages 30–39, 31.6% in Homa Bay while in Kisumu those between 40–49 formed the majority by 27.3%. On both sites, > 50% of participants were married and had attained at least primary school level education. A summary of the sociodemographic details of the focus group discussion participants held with community members in Kisumu and Homa Bay Counties is presented in Table Sociodemographic characteristics of FGD participants in Kisumu and Homa Bay Female 15 45.5 21 55.3 Male 18 54.5 17 44.7 20–29 8 24.2 9 23.7 30–39 6 18.2 12 31.6 40–49 9 27.3 9 23.7 50–59 7 21.2 6 15.8 60–69 3 0.1 2 5.3 None 1 0.03 0 0 Primary 22 66.7 23 60.5 Secondary 6 18.2 9 23.7 A-Level/college 3 0.99 6 15.8 Didn’t disclose 1 0.03 0 0 Formal employment 2 6.06 6 15.8 Business 17 51.5 11 28.9 Semi-skilled labor 10 30.3 6 15.8 Farmer/Agriculture 1 3.03 8 21.1 Unemployed 3 9.09 6 15.8 Didn’t disclose 0 0 1 2.6 Married 27 81.8 30 78.9 Single 3 0.09 2 5.2 Widow 3 0.09 6 15.8 In-depth Interview participants Role Number Trade 9 Health 2 Environment 2 Administration 2 Agriculture 2 Faith 1 Education 2 Total 20
Participants cited perceived satiety of some food types, age, occupation, taste, preference, and medical reasons as some of the influencing factors on what they would eat. The amount of money one has was mentioned by most of the participants.
“
It was interesting to note that majority of the participants were of the opinion that the choice of food was dependent on the person’s gender. The women were of the opinion that most men preferred traditional staple food like cassava and
“…
A majority of the participants especially in Kisumu mentioned local vegetables (e.g. sukumawiki). Other popular food items included:
Responses on the frequency of food purchases varied from participant to participant. For some, a weekly budget for the dry goods (cereals, flour) and daily purchase of perishable goods such as milk and vegetables was more feasible. Only a few suggested that they purchase foodstuff once a month. The majority of participants however reported making these purchases daily. Reasons provided for the daily purchase of food included: the need to ensure the family eats fresh food, a daily wage that only allows one to spend what is earned daily, and a lack of cold storage facilities (refrigerators).
For most families, special occasions include Christmas, when a child has done well in school when the family has guests or there is a family celebration. Meals provided on such special occasions include chapati (a round flat unleavened bread resembling naan usually made of whole wheat flour and cooked on a griddle pan) chicken, sweets, an assorted variety of store-bought baked goods, and food from the American fast food restaurant Kentucky Fried Chicken (KFC). The frequency of consumption of these foods also varied among participants with some quoting a weekly routine, others once every month, and others once or twice a year.
Both FGD and IDI participants stated that the choice of food and even its source had changed over time. One of the common intrapersonal level influences mentioned was convenience. Due to the nature of work, people are left with little time to prepare food and opt for store-bought options.
Location played a major role in the participant’s responses as to where they got their food. A clear dichotomy was discernable regarding sources of food between the two sites. In Homa Bay, a majority of the participants indicated they consumed food from their farms, including a variety of cereals, legumes, root tubers, vegetables, fruits, poultry, and dairy products.
This was in contrast to Kisumu where most participants reported that they get their food from an open market, small local retail stores (
Other community level influencers mentioned by the participants include Available foods in the market, regional staple foods, seasonality of produce, convenience,
The existence of taboos about food was mentioned as a cultural/ community influence. There are some parts of the chicken that women were prohibited from eating. In addition, mothers-in-law are not supposed to eat chicken in their son-in-law’s house as a sign of respect.
Some participants were of the opinion that food choice is also influenced by individuals wanting to be associated with a particular social class and wanting to fit in, therefore choosing to eat foods considered ‘classy’ even though they sometimes struggle to afford them.
Gender roles in the community played a major role in food purchase and preparation. Although some participants on both sites mentioned that both the man and woman participate in the purchase of food, the majority agreed that the women were solely responsible. Their reasons for this also varied.
The majority of the participants were of the opinion that it was the women who prepared meals in the home. Some of the reasons cited include the working hours of the man of the house and traditional expectations.
“
Some participants also observed that both men and women were involved in the cooking while others cited older children lending a hand in preparing meals.
Participants mentioned the change in the physical environment as a major source of concern. Climate change has affected the seasons making it difficult for farmers to plan planting seasons. This has also affected the production of fish in the lake.
The majority of the participants from both IDIs and FGDs mentioned distance to food markets as a major determinant of what people ate in households. This especially stood out from participants from Homa Bay.
Other factors mentioned as influencing the choice of food at the macro level include Food prices, political instability, health education, and road infrastructure.
It was noted that the export of locally available food e.g. fish led to a hike in prices of the product in the area of origin.
Participants were concerned about overpopulation and the lack of urban planning that has, in their opinion affected food security. This has in turn shifted the foodscape from traditional wet markets to refined foods in the supermarket.
A participant cited improved road networks as an influencing factor in the change of diet in many households. The food not produced in various towns is easily distributed to other areas in demand.
This qualitative study provides a useful perspective on the relationship between food retail and dietary preferences across various levels of influence in Kisumu and Homabay Counties in Western Kenya. Consistent with other findings [
Evidence-informed approaches are increasingly prominent on national agendas for health policy and health research especially in LMICs in relation to NCDs [
Although this study stratified the focus group discussions by socioeconomic status, which is a major inter-personal influence on food choice [
Other interpersonal influences such as perceived satiety experienced with some foods in comparison to others, cost of certain foods, and transportation costs – all influenced participants’ choice and source of food. This is consistent with findings from other studies [
In our study, social pressure was seen as a barrier to healthy eating with participants mentioning that wanting to fit in would force others into unhealthy eating habits. This was in part in line with a study conducted in Germany [
The local-based food pattern of
The high price of food was iterated by both FGD and IDI participants as a major influence on dietary preference. These findings were consistent with other studies [
With participants in this study worrying about the chemical content and fertilizer in the groceries sold in wet markets, there is a demand for policies that protect the food supply through the protection of the natural environment. These could include the prevention of industrial contamination of food and water, which could have other potential macro-level impacts on opportunities for healthy eating.
This study is one of few qualitative investigations into food choices and practices in this context. However, this study was not without limitations. As described, CHV judgments on socioeconomic status may have biased the sample towards lower and middle socioeconomic status households; greater diversity in the social-economic status of the participants may have provided additional insight. Though efforts were made to stratify focus groups in such a way to promote frank discussion (e.g. males separate from females), it still may be that social factors prohibited the discussion of some topics or the expression of opinions perceived to differ from the norm. To further qualify the responses by participants, it would be beneficial to include a quantitative assessment of daily food consumption in households since studies have shown significant variations in reported dietary intake as compared to actual consumption. In addition, to fully appreciate the multifaceted nature of the influencing factors in dietary preference, future studies, especially in Africa, would need to incorporate detailed views of the participants with regards to cultural influences, family dynamics, and political influences that were not fully explored in this study but have acted as a backdrop to the responses from the participants.
In conclusion, this study demonstrated that dietary preferences are complex and require interpretation through many lenses. Future interventions should not only consider intrapersonal and interpersonal influences when aiming to promote healthy eating among these communities but also need to target the community and macro environments. This means that nutrition promotion strategies should focus on multiple levels of influence that broaden options for interventions. However, government interventions to address food access, affordability, and marketing remain essential elements of any significant change.
We appreciate the support from the Kisumu and Homa Bay County Ministry of Health, the County Commissioners, local administration (the Chiefs and village heads), and community health volunteers (CHVs)for their assistance with community mobilization. We register support from the study coordinator: Lilian Sewe who took charge of the logistics and various other tasks that contributed to the success of this study. We are grateful to the participants in the FGDs and all those who participated in the study for their cooperation. This work is published with permission from the Director-General, Kenya Medical Research Institute.
Social mobilization: R.M, V.W, C.L and C.O. Training of field teams: R.M, C.L, P.W, V.W, and C.O. Read and approved the manuscript: RM, C.L, P.W, L.F, O.F, E.T, V.W, and C.O. Conceived and designed the study: C.O, L.F P.W, V.W, O.F. Analyzed data: R.M, P.W, and V.W. Drafted the manuscript R.M.
This research was funded by the National Institute for Health Research (NIHR) (16/137/64) using UK aid from the UK Government to support global health research. The views expressed in this publication are those of the author(s) and not necessarily those of the NIHR or the UK Department of Health and Social Care.
The focus group discussion guides used in this study are provided as supplementary documents. The transcripts generated for the FGDs will be availed upon request. The corresponding author will be available to provide these and any additional information required.
As described, this analysis forms part of a wider mixed-methods study. The study was reviewed and approved by the Scientific Ethics Review Unit (SERU) of the Kenya Medical Research Institute (KEMRI, SSC ≠ 3730). Thereafter, additional permissions were obtained from the County administration: Ministry of Education, Commissioner, and Ministry of Health. The purpose of the study and its objectives were explained to local authorities, opinion leaders, village elders, and community members involved in the study. Before the commencement of data collection, a consent form was shared with the study participants. The moderator explained to them in detail, their level of involvement, the time it would take for the FGD, and the minimal risk involved. Participants were also informed that they could leave at any moment without coming to any harm. After it was established that the participant was willing to participate, written Informed consent was obtained from the participating respondents. Personal identifiers were omitted from the dataset prior to analysis. All methods were carried out in accordance with relevant guidelines and regulations.
I declare that the authors have no competing interests as defined by BMC, or other interests that might be perceived to influence the results and/or discussion reported in this paper.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.