STRENGTHS AND LIMITATIONS OF THIS STUDY

This study was conducted in multicentre (six health centres), which increases generalisability.

Prone to recall bias due to the exact days of onset of signs and symptoms may not be remembered.

The data were collected at a single point in time; a lack of temporal relationship might happen between some independent variables and the outcome variable.

In addition, the private clinics were not included in this study.

Introduction

Malaria is a primary cause of disease and mortality worldwide. The WHO estimates that there have been 241 million cases and 627 000 fatalities from malaria, with 31 countries responsible for almost 95% of these deaths.1 The sub-Saharan African region faces a significant malaria burden,2 accounting for 95% of cases and 96% of deaths from the disease. It is estimated that children under the age of 5 account for 80% of all malaria deaths in Africa.1 Approximately 70% of the world’s malaria burden is concentrated in only 11 countries, of which 10 are in sub-Saharan Africa (Burkina Faso, Cameroon, the Democratic Republic of the Congo, Ghana, Mali, Mozambique, Niger, Nigeria, Uganda and the United Republic of Tanzania) and India.2

Between 2000 and 2019, there was a significant reduction in malaria incidence and fatality rates across the region, with a 29% decrease in incidence and a 60% reduction in mortality. Despite these improvements, the disease remains a significant public health challenge, with an estimated 1.2 billion cases and 7.1 million deaths during this period.3 In Ethiopia, the malaria burden was substantial, with more than 1.2 million cases reported in 2017/2018. By 2022, this number had increased to 1.6 million cases.

The prevalence of malaria varies across different regions within Ethiopia: in Bahir Dar city administration, Amhara, southern national and nationality and people region, and Oromia regional state were 13.61%, 24.7%, 12.4%, 16.17% and 13.11%, respectively.4 5 The transmission of malaria is generally unstable and seasonal, with fluctuations depending on various factors such as rainfall, temperature, altitude and other ecological factors.6

In Ethiopia, the highest prevalence of malaria typically occurs between September and December, following the main rainy season, June to August.7 Approximately 60% of Ethiopians are at risk of malaria infection, with the disease being prevalent in over 70% of the nation.8–10 Andabet district, located in the Amhara region of Ethiopia, malaria is the leading cause of morbidity, accounting for 17.7% of all patient cases. The healthcare infrastructure in the district includes 6 health centres, 26 health posts and 3 private clinics.11

The magnitude of delayed seeking treatment among malaria patients varies across different regions. For instance, studies have shown that the rate of delayed treatment is 79% in Thailand,12 51.5% in Nigeria13 and 55.4% in Tanzania.14 In Ethiopia, similar studies report delayed treatment seeking a prevalence of 74.5% in East Shewa,15 80.2% in the Southwest of Ethiopia16 and 47.6% in Northwest Ethiopia.17 Several factors contribute to the delay in seeking treatment among malaria patients. These include fear of side effects, having no formal education, lack of transportation access, marital status, who had a history of death related to malaria, fear of the high cost of treatment services, residence, knowledge status of malaria symptoms, a history of malaria, distance from the health centres and family size. These were significant factors.14–20

Early diagnosis and treatment of malaria should occur within 24 hours of the onset of symptoms to decrease the risk of serious illness such as renal failure, convulsions, mental disorientation, disability and death.21–24 To avoid the severe and fatal illness of malaria, the Ethiopian Federal Ministry of Health recommended that treatment must begin as soon as possible, generally within 24 hours after symptom onset,25 and delay in malaria diagnosis and treatment seeking is a major problem in Ethiopia.26

The Ministry of Health of the Federal Democratic Republic of Ethiopia has established a 5-year National Malaria Strategic Plan (NASP) with the ambitious goal of reducing malaria mortality and morbidity by 50% by the end of 2025. Delay in diagnosing and treating malaria is, therefore, crucial to meeting the NASP target because it reduces the disease’s severity, mortality and economic impact.8 The malaria burden is still being challenged by biological issues (eg, mosquito resistance to insecticides and parasite resistance to antimalarial medications) and a lack of consistent international and domestic funding.27

Additionally, a patient’s fatal prognosis can be improved, and the progression of their illness to a critical or severe stage can be avoided with early, correct diagnosis and risk assessment for malaria.28 Determining the degree of patient and health system delays can offer a standard by which delay-reduction initiatives can be evaluated.29

Delayed seeking of malaria treatment is a common problem in our study setting. However, limited studies have assessed the prevalence and associated factors of delayed malaria treatment in the study setting. Most previous studies on delayed malaria treatment in Ethiopia are restricted to specific under 5 age groups and do not state the overall prevalence of delayed malaria treatment. However, this study uses all age groups as the study population. Therefore, this study aimed to assess the magnitude and associated factors for delayed treatment seeking among patients with malaria in Andabet district, Northwest Ethiopia, 2022.

Methods and materialsStudy design and period

A multicentred institution-based cross-sectional study was conducted among patients with malaria in Andabet District Health Centres from 8 July 2022 to 21 August 2022.

Study setting

The study was conducted in the Andabet district, which is located in the South Gondar zone, Northwest Ethiopia. It is 192 km from the capital city of the Amhara regional state, Bahir Dar, Ethiopia. It shares a border with the Estie, Dera, Huletejunese and Gonji Kollela districts in the West Gojjam zone. Administratively, the district is divided into 26 kebeles and has an estimated population of 147 641 by 2021/2022. Of these, 75 445 (51.1%), 19 991 (13.54%) and 4976 were male, under 5 children and pregnant mothers, respectively.11

The climatic conditions of the district were 3%, highland, 91% temperate zone and 6% lowland. The annual rainfall ranges from 700 to 2500 mm, and the altitude above sea level ranges from 1800 m2 to 2608 m2 with 8.3°C–25°C average annual temperatures. Of the 26 kebeles, 81% are malaria-endemic, and the remainder are affected by imported malaria cases during the major and minor malaria transmission periods.11

According to the annual report for this study area, the proportion of malaria cases among children under 5 and pregnant mothers was 10.1% and 2%, respectively. Annual parasite incidences in the study were 11.33 per 1000 persons, and the level of endemicity is moderate. Plasmodium falciparum is the most prevalent malaria parasite in the Andabet district, accounting for 53.2% of cases, whereas Plasmodium vivax is 47.8%.11 In northwest Ethiopia, the most prominent malaria vectors are Anopheles arabiensis; it is highly adaptable and thrives in various ecological settings, including urban, rural and agricultural areas.

Population

All malaria patients under the catchment area were the source population, and randomly selected microscopically confirmed malaria patients who sought treatment in Andabet district public health centres during the study period were the study population.

Eligibility criteria

All residents who lived for more than 6 months and visited the health centres during the study period for the treatment of malaria infection were included. Patients who had been referred from other health centres or had relapsed malaria cases during the study period were excluded.

Sample size determination and sampling procedureSample size determination

For the first specific objective, the sample size was determined by the single population proportion formula with a 95% CI and 5% df, and the previous study shows in the Oromia special zone 61% prevalence (p),26 margin of error (d) (0.05). The formula to determine the sample size is shown below:

For possible non-response rates during the data collection period, the final sample size was increased by 10%. The final sample size for this study was 403.

Sampling technique and procedures

Systematic random sampling was used in this study. The average number of malaria patients in the preceding similar period was 847 per 2-month duration. The total number of patients with malaria receiving treatment in the previous year was divided to determine the sample interval. The Kth value was 847/403, which approximates 2, therefore, every other patient was selected by lottery methods after the first study participant, and the sample size for the six health facilities was distributed proportionally (figure 1).

Figure 1

Schematic diagram of sampling procedure prevalence and associated factors of delayed treatment among patients with malaria in Andabet district health centres, Northwest Ethiopia, 2022.

Variables of the studyDependant variable

Delay malaria treatment (Yes/No).

Independent variables

Sociodemographic variables: age, sex, religion, marital status, residence, educational level, occupation, family size and average monthly income.

Behavioural, physical and environmental variables: history of death from family due to malaria, fear of cost, fear of side effects, distance to health centres, access to health facilities and type of transportation.

Knowledge of malaria cause, signs and symptoms, and prevention methods: cause of malaria, transmission of malaria from person to person, signs and symptoms, prevention, advantage of early diagnosis and treatment.

Data collection tool and procedures

The data were collected from malaria patients by using a structured questionnaire that was adopted from different articles.20 26 The variable consists of sociodemographic, behavioural, physical and environmental knowledge of malaria causes, signs and symptoms, and prevention methods variable. Data were collected by six healthcare providers assigned to public health emergency management. The structured questionaries used for data collection are provided in online supplemental file 1.

Data quality control

A structured questionnaire was pretested for 5% (18 malaria patients) of the sample size before the actual data collection period at Estie district health centre. Moreover, the data collectors were trained for 1 day on the purpose of the study, details of the questionnaire, data collection procedure and ensuring the confidentiality of the respondents. The structured questionnaire is translated into Amharic and retranslated back to English to ensure the accuracy of translation into Amharic language. The completed checklist was checked for consistency and completeness before entry by the principal investigator. The data entry was cross-checked and we clarified the missing data.

Data management processing and statistical analysis

Data were checked for completeness before entry, and a code was assigned. The data were then entered using EpiData V.4.6, and Stata SE V.14 was employed for analysis. Descriptive statistics, including means, frequencies, percentages and tables, were used to present the results. Bivariable logistic regression analysis was conducted to identify risk factors for delays in seeking malaria treatment. Variables with a p value of less than 0.2 in the bivariable analysis were included in a multivariable logistic regression to control for potential confounders. Adjusted ORs (AOR) with 95% CIs and p values less than 0.05 were used to identify significant factors associated with malaria treatment delay.

A multicollinearity test was conducted to assess the association between independent variables with a p value of less than 0.2 in the bivariate analysis, using the variance inflation factor (VIF), with a mean VIF value of 3.47. Model fitness was evaluated using the Hosmer-Lemeshow goodness-of-fit test.

Patient and public involvement

Patients and public were not directly involved in the design, conduct and dissemination of this study.

ResultsSociodemographic characteristics of respondents

A total of 403 malaria patients were interviewed with a response rate of 100%. The mean age of the respondents was 27.6±SD 14.26 years. Out of the 403 study participants, 80% were over 15 years old, and 60.55% were male. Regarding their residence, more than 88% were from rural areas. Two-thirds (62%) of the respondents were married and 60% had no formal education (table 1).

Table 1

Sociodemographic characteristics of study participants on delayed malaria treatment seeking at Andabet district, Northwest Ethiopia from 8 July 2022 to 21 August 2022 (n=403)

Behavioural, physical and knowledge of malaria-related characteristics

61% of participants had access to health centres, of which one-third (35%) of respondents were less than 5 km from the health canter’s radius. In this study, 38.71% of patients travelled more than 5 km to seek treatment, 6.18% travelled on foot and 29.78% had a family history of malaria-related deaths. More than half of our study participants (52.11%) were unaware that malaria can be transmitted from person to person (table 2).

Table 2

Behavioural, physical and knowledge-related characteristics of study participants on delayed malaria treatment seeking at Andabet district, Northwest Ethiopia from 8 July 2022 to 21 August 2022

Prevalence of delayed malaria treatment

In this study, the prevalence of delayed malaria treatment seeking was 78.41% (95% CI 74.11, 82.17). Of 316 who had delayed malaria treatment seeking, the majority 137 (43.4%) of the study participant had taken their treatment within 48–72 hours after the signs and symptoms. Of the delayed malaria patients, 131 (41.5%) were female and 286 (90.5%) were from rural residences.

The average time for malaria patients to seek treatment was 48.4 hours. When analysed by age categories, the mean time to treatment seeking was 44.6 hours for children under 5 years, 52.03 hours for those aged 5–14 years, and 47.8 hours for individuals aged 15 years and older.

Factors associated with delayed malaria treatment

In bivariable analysis, all variables under study were possible factors except marital status, family size, fear of side effects of malaria medication, knowing malaria signs and symptoms, and whether it can be malaria transmitted from person to person through mosquito bite showed that possible factors for delayed malaria treatment at a p value of less than 0.2.

In multivariable logistic regression analysis, participants who did not know the cause of malaria (AOR 2.24, 95% CI (1.07, 4.65)), did not know signs and symptoms of malaria (AOR 2.52, 95% CI (1.04, 6.07)), travel more than 5 km (AOR=5.37, 95% CI (1.22, 23.55)) and fear of malaria treatment costs (AOR=1.88, 95% CI (1.05, 3.36)) were significantly associated with delayed malaria treatment seeking (table 3).

Table 3

Variables associated with delayed malaria treatment seeking at Andabet district, Northwest Ethiopia from 8 July 2022 to 21 August 2022 (n=403)

The odds of delayed treatment seeking among participants who did not know malaria signs and symptoms were 2.52 times higher than those who knew malaria signs and symptoms.

In multivariate analysis, patients who did not know the cause of malaria were 2.24 times more likely to delay malaria treatment than those who knew the cause of malaria.

The participant who travelled more than 5 km had 5.37 times (OR=5.37, 95% CI (1.22, 23.55)) higher odds of delayed treatment seeking as compared with less than 3 km traveller.

The odds of delayed treatment seeking among participants who do not know the cause of malaria were significantly (OR=2.24, 95% CI (1.07, 4.65)) higher than their counterparts.

Discussion

Delayed malaria treatment seeking is a major problem. In this regard, this study aimed to determine the prevalence and associated factors of delayed treatment seeking among patients with malaria in the Andabet district, south Gondar zone, Amhara region and Northwest Ethiopia.

Our study showed that the prevalence of delayed malaria treatment was 78.41% (95% CI 74.11, 82.17). This finding was in line with studies conducted in the Oromia Region (80.2%),30 South East Nigeria (78%),13 Thailand and Myanmar (79%).12 However, it is higher than studies conducted in Dera district, in Amhara region, Ethiopia (47.6%),17 Oromia special zone, Ethiopia (61%),26 Sudan (40%),31 Eritrea (39.9 %),32 in 25 low-income and middle-income countries (72%)33 and Indonesia (53.2).34

However, the prevalence of delayed malaria treatment in our finding (78.41%) is lower than studies conducted in Tanzania (93.7%),35 Benin (87.8%),36 Southern Ghana (89%)37 and Shan State, Myanmar (84.2%).38 The rationale for inconsistency could be that the majority respondents (70.22%) comprised farmers, and the data collection time for our study corresponded with the main harvesting season for rural residents. It is possible that these individuals put their farming pursuits ahead of receiving prompt medical attention. In addition, the summer rainy seasons during which schools were closed may have hindered the spread of malaria information from different organisations, lowering awareness and delaying action.

Additionally, the dissimilarities in the prevalence of delayed malaria treatment from the previous studies are due to in study population. Many of the studies reporting lower prevalence focused primarily on children under 5 children, a group typically prioritised for prompt treatment interventions.14 36 In contrast, our study included participants from the entire population, which could account for the higher prevalence observed. Lastly, sociodemographic factors such as educational level may explain delays, 60.3% of participants were unable to read and write, hindering symptom recognition and timely treatment. Cultural differences between study population in health beliefs and practices may also influence treatment-seeking behaviour.39 40

The likelihood of delayed treatment was significantly higher among participants who were unaware of the causes of malaria, clinical signs and symptoms of malaria, those who had to travel longer distances to reach health centres and those fear of malaria treatment costs.

Accordingly, study participants who lived in an area 5 km or more away from the health centres were 5.37 times more likely to be delayed to take malaria treatment than those who lived near health centres (3 or less kilometres). This finding was supported by studies conducted in Dera district, Ethiopia,17 20 Oromia Special Zone, Ethiopia,26 Hadiya Zone, Ethiopia,41 Eritrea,32 Tanzania,35 Equatorial Guinea,42 Indonesia43 and Shan State, Myanmar.38 This association might be because of the inaccessibility of health centres due to geographical distance, making it difficult for people to visit these facilities and for responsible bodies to create awareness about malaria prevention and treatment.20 39 In addition, 156 participants (38.71%) lived far from the health centres, which likely reduced their exposure to essential malaria information and contributed to challenges in accessing timely care.

This study revealed that the study participants who did not know malaria sign symptoms were 2.5 times more likely to delay seeking treatment compared with those who knew malaria symptoms. This finding is supported by different studies conducted in Dera district, Ethiopia,20 Eritrea,32 Cameroon,44 Rwanda45 and Myanmar.12 This could be because participants who recognised malaria signs and symptoms were more likely to seek timely diagnosis and treatment.46 In contrast, those unaware of these signs and symptoms might pursue alternative options, such as visiting traditional healers, a common practice among caregivers in Ethiopia before seeking formal healthcare.47 Furthermore, a person not knowing the signs and symptoms may pay less attention to taking early treatment and may not consider malaria.

Delayed malaria treatment was more common among participants who were unaware of the cause of malaria, with higher odds of delayed treatment-seeking compared with those who knew the cause. This finding aligns with a study from Equatorial Guinea. This can be due to the higher probability that the patients did not pay attention to the early treatment of malaria at the health facilities and did not know the cause: who did not perceive that they could be infected by malaria was the reason for the delayed treatment. For other reasons, some of the respondents delayed some days because they wanted to ascertain the cause of the illness.

In this study, fear of malaria treatment cost was a significant factor for delayed malaria treatment seeking as compared with those who did not fear treatment-related costs. This study finding was supported by studies conducted in South West Ethiopia,16 Benin,36 Sekou-Touré Regional Referral Hospital in Mwanza, Tanzania,48 Cameroon,44 South-eastern Nigeria13 and Shan State, Myanmar.38 This concern may be particularly prominent in low-income settings, where individuals prioritise other financial needs or mistakenly believe that malaria treatment is too expensive. Additionally, the lack of awareness about the availability of free or subsidised treatment services, further contributing to delayed care-seeking.

As a limitation, prone to recall bias due to the exact days of onset of signs and symptoms may not be remembered. The data were collected at a single point in time; a lack of temporal relationship might happen between some independent variables and the outcome variable. In addition, the private clinics were not included in this study.

Conclusions

The majority of study participants delayed seeking their malaria treatment. Do not know the cause of malaria, do not know malaria signs and symptoms, a distance greater than 5 km travelled to reach health centres and fear of treatment cost were significant factors for delayed treatment seeking among malaria patients.

Recommendations

Based on our findings, recommendations for the Andabet district health office and health centres include increasing emphasis on health education about malaria’s signs, symptoms and causes, particularly through campaigns targeting high-risk groups and remote residents. Raising community awareness of malaria services and expanding access to treatment at local facilities is essential. Decision and policy-makers should focus on building standard health facilities and designing programmes that highlight the seriousness of malaria and the importance of early diagnosis and treatment.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information

Ethics statementsPatient consent for publication

Consent obtained directly from patient(s).

Ethics approval

Ethical approval was obtained from the institutional review board of the University of Gondar, College of Medicine and Health Sciences, on the behalf of Institute of Public Health with Ref No: IPH/2192/2014. The Andabet District Health Office provided a formal permission letter was given. Written (for literate) and oral (for those unable to read and write) informed consent was secured from the study participants before starting the data collection. For those under 15 years old, the ascent was sought from the parents for providing information about them. In the case of patients whose age was 15–17 years old, permission was sought from their families, and information for the survey was provided by the patients themselves. Participants’ information was kept private in a safe and secured place, and to keep the confidentiality names and other personal identifiers were not included in the data collection tool.

Acknowledgments

The authors acknowledge the University of Gondar for ethical clearance, Amhara Regional Health Bureau for financial support and for each health centre and data collector.