These authors share last authorship
Healthcare-seeking decisions reflect personal assessments of illness etiology and severity to determine whether, how, and where to seek care. Understanding these contexts enables targeted interventions to address barriers and ensure timely access to quality services. We examined patterns of household healthcare seeking for emergent ill-health in Western Kenya.
As part of a multi-site mixed-methods study evaluating a patient-centered, digitally-facilitated self-testing intervention, we analyzed 16 in-depth interviews and two focus group discussions (
We found that a complex interplay of psychosocial, enabling, and need factors determined if, when, and where households sought care for ill members. Psychosocial influences included prevailing community norms that favored home remedies and self-medication prior to formal care as in the quote below: ‘
Household healthcare-seeking decisions were shaped by a dynamic interplay of psychosocial, enabling, and need factors. Addressing these multifaceted barriers through targeted, context-specific interventions, such as enhancing health infrastructure and integrating different providers, is crucial to ensuring timely access to quality healthcare.
Healthcare-seeking behavior (HSB), defined as any activity undertaken by individuals who perceive themselves to have a health-related problem with the aim of finding an appropriate remedy, is critical for better health outcomes (
More than half of the illnesses that households encounter, such as malaria, pneumonia, diarrhea, and measles, are preventable and treatable if management decisions are made in time (
Although interventions aim to enhance health infrastructure and capacity, their success depends on the active participation of the intended beneficiaries. For instance, if households do not make the right decisions to send members for care when indicated, the health facilities with all the capacity become dysfunctional and unlikely to achieve their health goals (
To date, few studies have addressed household healthcare-seeking patterns, particularly regarding emergent and chronic conditions. Understanding factors influencing household healthcare-seeking decisions allows targeted interventions to address barriers and ensure timely access to quality healthcare for improved health outcomes. Theoretical frameworks such as the Andersen Expanded Behavioral Model of Health Services Use (
This qualitative analysis was part of a multi-site mixed methods study to evaluate the acceptability and feasibility of a patient-centered, digitally facilitated integrated healthcare intervention for addressing acute infectious and chronic non-communicable diseases in rural and urban communities in Kenya, South Africa, and Zambia. Using qualitative methods, the Kenyan site sought to assess healthcare-seeking patterns and healthcare access utilization in-depth interviews and focus group discussions.
This study was conducted in rural, peri-urban, and urban areas of Migori County and targeted community members in different settings, including fishing, mining, and farming. Our population also included healthcare providers and managers, as well as people at policy-making levels. Migori County is one of the 47 counties of Kenya, bordering Lake Victoria in the southwest of the country. It has an estimated population of 1,116,436, comprising 536,187 males and 580,214 females, as per the 2019 National Census (
We conducted 16 in-depth interviews with community members and two focus group discussions with 14 community members. One FGD was reserved for women aged 25–45, and the second one consisted of young males aged 18–24. We purposefully sampled information-rich participants representing different economic activities in which they were involved (e.g., farming, mining, fishing), cadre/position in the communities (e.g., teachers, healthcare providers, religious leaders), and geographic areas (e.g., rural, peri-urban, and urban). This approach allowed us to capture a broad range of perspectives, ensuring that the voices of different socio-economic and occupational groups were represented in the study.
Research Assistants (RAs) worked with identified participants to schedule interviews and focus group discussions at a time and venue convenient to participants, ensuring privacy. We invited participants to community social venues, health facilities, and other conducive venues for interviews. Participants arriving at the venue were directed to a private room for consenting before participating in the in-depth interview or FGD. FGD participants were assigned numbers for use during the discussion instead of their names for confidentiality purposes. The FGD and IDI guides explored issues around the patient experience or health journey, including healthcare-seeking decision-making processes, healthcare utilization, and access for emergent and chronic diseases such as HIV and diabetes. Other topics covered included the healthcare system, provider roles, and their activities at the facilities. The discussions and interviews were audio-recorded and took an average of 2 h and 1 h, respectively. The interviews and discussions were conducted either in English, Kiswahili, or Dholuo, depending on participant preference.
The audio files from the discussions and interviews were transcribed into English, de-identified, and stored securely in a password-protected folder. To ensure transcription quality, a different study staff validated the accuracy of the transcript by directly comparing part of it with the audio file. Any changes needed were discussed and made jointly with the RA who conducted the transcription. We applied deductive and inductive approaches to develop a codebook based on themes in the guide and the insights from the transcripts when they became available. The codebook and transcripts were then imported into Nvivo for coding. To achieve inter-coder reliability, the coding team of three, led by AS, coded two transcripts jointly and discussed the differences in the application of codes to build consensus and consistency in code application. The remaining transcripts were shared among three coders for individual coding. After coding, the code reports were generated for framework analysis that explored households' perspectives on healthcare-seeking decision-making and healthcare outlet options.
Slightly over half of the participants (52%) involved in the survey were male, 41% were aged 25–49, and 85% of the adults reported owning a phone. Household surveys revealed that most participants (81%) had sought some kind of treatment in the preceding 12 months (
Socio-demographic and healthcare-seeking attributes of the participants.
| Male | 135 | 52 |
| Female | 126 | 48 |
| Children (1–17) | 66 | 25 |
| Young people (18–24) | 49 | 18 |
| Mid age (25–49) | 109 | 41 |
| Elderly (50+) | 43 | 16 |
| Farming | 47 | 18 |
| Boda boda | 28 | 11 |
| Fishing-related work | 33 | 13 |
| Mining-related work | 28 | 11 |
| Child/student | 83 | 32 |
| Other | 40 | 15 |
| Yes | 166 | 85 |
| No | 29 | 15 |
| Yes | 211 | 81 |
| No | 50 | 19 |
| Public health facility | 113 | 53 |
| Private health facility | 65 | 30 |
| Traditional healer | 3 | 1 |
| Community health promoter | 12 | 6 |
| Drug shop/pharmacy | 18 | 9 |
| 1–2 h | 25 | 12 |
| Less than 1 h | 186 | 88 |
| Vehicle | 14 | 7 |
| Walked | 85 | 40 |
| Motorcycle taxi | 99 | 47 |
Though different by household, participants reported that usually, households with a sick member initially try home remedies that are immediately available and less costly. The prioritized home remedies include herbal medicine, leftover medications, and first aid interventions such as lowering fever with a wet cloth. After home remedies, the participants may then go for over-the-counter medications, initially purchasing painkillers and other fever-lowering medications, before they decide to seek care in health facilities, along with other interventions, such as visiting traditional and religious healers, especially if the illness persists in the first few days of treatment.
Some participants reported visiting CHPs (previously referred to as community health volunteers—CHVs) as part of their initial interventions before visiting health facilities. They observed that, lately, CHPs who live within their communities have been trained by the government and equipped with basic medications such as painkillers for initial interventions. In describing the convenience of CHPs in the communities, participants likened CHPs to a fire extinguisher in a factory that can be used before firefighters arrive.
…
The participants' level of knowledge and awareness about certain health conditions and how to handle them helped them in making healthcare-seeking decisions for their households. The specific knowledge factors reported to compromise household healthcare-seeking decisions included lack of ability to recognize diseases, stigma, and poor knowledge of medicine. For instance, the household head's level of awareness of disease etiology and symptoms determines where and when to seek care for a household member. Some participants reported that they always prioritize visiting health facilities because that is where they were told they would find experts to provide appropriate interventions.
Participants described long waiting times as one of the main hindrances to seeking care, especially at public health facilities. The fear of spending an extended period waiting for treatment at the facilities, with the possibility of commodity stockout and being asked to purchase medications elsewhere, made people favor other interventions, including self-medication. While some people may understand if the long waiting time was due to large patient numbers, they were dismayed if the wait was caused by healthcare providers reporting late to duty or prioritizing their personal commitments over patient care.
Participants emphasized the ability to tolerate a 2–3 h wait, preferably in the morning, to get treatment, but not the whole day due to other competing responsibilities and being dependent on daily wages, which they only earn when they work.
Participants reported residing in remote rural areas with poor road networks and lack of transport as reasons for the delay in accessing healthcare at public health facilities. Available means of transport, such as motorbike taxis, are withdrawn due to impassable roads or trucks when it rains or late at night, prompting the need for pragmatic decisions on where and when to seek healthcare.
Related to the lack of transport and poor road networks, long distances to public health facilities and lack of funds to pay for transport and user fee charges sometimes drive the decision to go for cheaper alternatives, including traditional healers and self-medication through drug purchase from local pharmacies.
…
Another key factor cited in care-seeking decisions is having a favorite healthcare provider and flexibility in accessing services. For instance, a healthcare provider on duty can provide telemedicine to individuals with busy schedules, such as miners and fishermen, and connect them with CHPs promptly. This can dissuade households from relying on home remedies or self-medication, which may waste precious time from symptom onset to receiving a diagnosis and adequate treatment.
Perceived quality of services at public health facilities, including staff attitude and stockouts, is considered when households decide how to respond to an illness. If anticipating poor quality of care, especially stockout of prescribed medications, households explore cheaper and promising alternatives, including supernatural and self-medication interventions, before trying medication perceived as costly or healthcare perceived as inferior in quality.
Participants reported that people with health insurance coverage, such as the National Health Insurance Fund (NHIF), now referred to as Social Health Insurance Fund (SHIF), often have several options available to them about where and when to seek care. The options available to them include accessing care at private health facilities where staff are friendly and have a better quality of care. However, it was pointed out that the cost of services in private health facilities is out of reach for households without health insurance, which require multiple payments for registration, tests, and medication, thereby indirectly discouraging visits. The cost of healthcare services acts as a critical enabler or barrier. As such, the possibility of sharing financial burdens within families or borrowing to afford healthcare or insurance makes healthcare decision-making for timely interventions a lot easier.
The participants reported that the action taken depended on perceived complications and intensity of the illness and whether it is a known chronic disease that requires regular clinic visits. Home remedies, including leftover medications and over-the-counter medications from nearby pharmacies, are usually considered for conditions perceived to be mild and not serious.
Participants reported that the action taken sometimes depended on who was ill within the household, with care being sought more quickly and from multiple sources for children, who are believed to be more vulnerable than adults.
Participants reported that where to seek care and the urgency depend on the perceived etiology of the disease. If the disease is perceived to be caused by supernatural powers, the focus is always to seek care from traditional or religious healers.
In this study, we identified healthcare-seeking patterns in the context of emergent ill-health within households. We found that psychosocial (e.g., community norms), enabling (e.g., distance and availability of transport), and need (e.g., perceived severity and complexity of the illness) factors informed the household decision of whether, when, and where to seek care for ill members. We also found that, in most cases, households started their interventions by administering home remedies that included the use of leftover medications, local herbs, and wet clothes to control pain and fevers. Lack improvement made the intervention to be escalated to seeking care at public health facilities or from traditional and religious healers, which sometimes overlapped depending on the rate of patient recovery.
Overall, healthcare-seeking is a complex social process involving how symptoms are perceived, interpreted, and appraised to arrive at a decision that is often informed by people's ability and motivation to access care (
Prioritizing home remedies postpones access to a qualified healthcare provider, resulting in diagnostic delays with far-reaching implications, especially for infectious diseases such as tuberculosis (TB) or diseases such as pneumonia and malaria in children and pregnant women (
Due to various reasons, including community habits and norms, cost and distance to health facilities, perceived severity of the illness, and the timing of symptom onset, people often explore interventions that are within reach and available, even if they are not necessarily optimal. For these communities, responsive healthcare interventions may acknowledge sequential healthcare-seeking behavior and work within this framework to introduce early points of professional intervention. Enhancing interventions such as those envisaged in Kenya's community health strategy, which recognizes the role of Community Health Promoters (CHPs), is likely to speed up the process of disease diagnosis and linkage to appropriate care (
Health system contexts are key considerations for households when making decisions about seeking healthcare. Among the health system factors considered by households in our study were waiting times and stockouts at health facilities, which led households to opt for home remedies and self-medication. Other studies have also reported similar factors for poor access to or retention in chronic disease programs, especially at public health facilities (
This study has several limitations (1) we did not capture health providers' perspectives on health system-related challenges that preclude households from prioritizing seeking care at public health facilities. Considering the perspectives of healthcare providers would have provided a more complete picture of the context of healthcare in reference to health system resources and challenges. (2) We also acknowledge the challenge of social desirability and recall bias where participants may have given responses that conform with what they perceived to be right or were not able to remember clear how or the contexts they made certain healthcare seeking decision. (3) As an inherent constraint of qualitative studies, this work lacks generalizability and representativeness but we endeavored to achieve diversity in participant enrolment to make the findings applicable in similar settings elsewhere. Despite these limitations, this study still offers critical perspectives on factors that influence healthcare decision-making and the emerging patterns that are essential considerations for countries such as Kenya that are implementing universal health coverage.
In conclusion, Household healthcare-seeking in Migori County is shaped by intertwined psychosocial, enabling, and need factors that often delay timely access to formal care. Households first considered home remedies that included the use of leftover medications and local herbs, and control of pain before seeking care at public health facilities or from traditional and religious healers, with overlaps depending on the patient's recovery rate. Nurturing the success of the Kenyan revamped community health strategy with well-resourced and equipped CHPs is likely to minimize time to allopathic care by reducing the need for home remedies and self-medication. While strengthening health infrastructure remains critical, our findings suggest several additional policy directions. First, pharmacies and traditional or religious healers—who are frequently the first point of contact for households—could be integrated into a formal referral system through structured partnerships, training, and supervision to ensure timely linkage to biomedical care. Second, targeted financial mechanisms, such as low-cost micro-insurance schemes or conditional transport subsidies, could address financial and geographic barriers while improving equity in access. Third, expanding the role of Community Health Promoters to include patient navigation, health literacy, and referral follow-up would build trust and bridge gaps between households and facilities. By aligning health system reforms with the realities of household decision-making, we can accelerate earlier diagnosis, reduce preventable delays, and advance progress toward universal health coverage.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving humans were approved by Kenya Medical Research Institute Scientific and Ethics Review Unit (SERU). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
ZK: Formal analysis, Writing – original draft, Methodology, Conceptualization, Writing – review & editing, Investigation, Validation, Supervision. EB: Funding acquisition, Writing – review & editing, Resources, Conceptualization, Methodology, Investigation. CM: Writing – review & editing, Formal analysis. NS: Data curation, Methodology, Supervision, Writing – review & editing. FO: Formal analysis, Data curation, Writing – review & editing. JM: Conceptualization, Methodology, Investigation, Writing – review & editing, Funding acquisition, Resources. AW: Writing – review & editing, Methodology. DP: Methodology, Writing – review & editing. AV: Writing – review & editing, Methodology, Investigation. HH: Methodology, Writing – review & editing, Investigation. DR: Methodology, Writing – review & editing. MB: Methodology, Writing – review & editing. SC: Writing – review & editing, Methodology. AS: Investigation, Writing – review & editing, Formal analysis, Methodology. PD: Funding acquisition, Writing – review & editing, Methodology, Conceptualization, Resources, Investigation.
We wish to acknowledge the following for their support and leadership that enabled the study to be conducted (a) Director General, Kenya Medical Research (Nairobi) and (b) Director, Centre for Microbiology Research (Nairobi). We thank the Migori County Health Management Team, as well as their Sub-county and health facility counterparts, for their support. We would also like to acknowledge the hard work of the DASH Team, particularly the staff who worked on the DASH Study. We also appreciate our sister projects under the Research Care and Training Program (RCTP). In a unique way, we thank all healthcare providers, Community Health Promoters, and study participants who took the risk and volunteered to participate in our study.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author(s) declare that no Gen AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Edited by:
Reviewed by: