Assessment of Urban Health Extension Package Utilization and Healthcare Seeking Behavior Among Model and Non-model Households in Addis Ababa, Ethiopia: A Comparative Community Based Study

Background: Ethiopia has been implementing the Urban Health Extension Program (UHEP) at the community level since 2009. The program was a pro-poor and cost-effective approach that aimed to enhance the utilization of urban health extension program packages and empower community healthcare-seeking behavior. This study was conducted to compare the utilization and healthcare-seeking behaviors of urban health extension program packages and the healthcare-seeking behaviors of model and non-model households. Methods: A community-based comparative cross-sectional study was conducted among 594 female household heads (297 models and 297 non-models) using a structured face-to-face interview. A bivariate and multivariable logistic regression analysis was employed to identify associated factors. A p-value less than 0.05 and an adjusted odds ratio (AOR) with a 95% confidence interval were carried out to identify significant factors. Results: Urban health extension program package utilization was 78% among model and 64.2% among non-model female households. A total of 75.5% of model and 65.2% of non-model female household heads had appropriate healthcare-seeking behavior. Moreover, having information about UHEPs (AOR = 2.35, 95% CI = 1.08 - 3.42), the frequency of home visits by UHEWs (AOR = 2.12, 95% CI = 1.01 - 3.13), knowledge about UHEPs (AOR = 3.14, 95% CI = 2.43 - 4.47), and household graduation status (AOR = 3.052, 95% CI = 2.024 to 5.113) were significantly associated with urban health extension program package utilization and healthcare-seeking behaviors. Conclusion: In terms of utilization, the overall urban health extension package favors model female household heads over


Introduction
Globally, urban health is underutilized and neglected, and there are health inequalities, particularly in low-and middleincome countries [1] .By 2050, nearly 60% of Africa's population is expected to live in cities, which are home to 35-40% of the world's children and adolescents [2] .According to UN-Habitat, the proportion of the urban population living in slums in developing countries has decreased from 39.4% in 2000 to 29.7% in 2014 [3] .Since 1997, Ethiopia has been implementing successive health sector development plans and has made notable advancements in expanding access to healthcare services and enhancing health outcomes [4] .
The urban health extension program (UHEP) was implemented in Ethiopia, and the deployment of specially trained urban health extension professionals (UHE-Ps) began in 2009 with the goal of improving community utilization of urban health extension packages and healthcare-seeking behavior [5] [6] .Urban health extension professionals (UHE-Ps) spend more than 75% of their time in the community educating residents about urban health extension program packages as well as identifying and preparing model households [7] .Model households (HH) are those that complete at least 75% of the model family training out of 60 training hours and implement and use packages at the household level, implying that households have acquired the necessary knowledge, skills, and behavioral changes to help them have better control over their health.
Healthcare-seeking behavior (HSB), on the other hand, is the action of persons visiting any health facility for modern treatment rather than traditional medical care [8] .
Evidence from Ethiopia revealed that only 59.2% and 72.8% of participants use urban health extension at the household level among model and non-models, respectively [9] [10] .Several factors were reported for the poor utilization of urban health extension packages, including sociodemographic and economic factors, household factors such as occupation, household income, frequency of home visits, model household training, and graduation from a model household [11][12] [13] .
Evidence also showed that literacy, educational status, perceived illness, income, and treatment costs were some of the predictor factors for health-seeking behaviors [14][15] .
Moreover, health care policies and programs' planning requires knowledge about healthcare seeking behavior for early diagnosis, effective treatment, and appropriate intervention [16] .Besides, identifying gaps and having regular and up-todate data on model and non-model households are critical for evidence-based decision-making and baseline data for any stakeholders to take action.As a result, the aim of this research was to compare the utilization of urban health extension packages among model and non-model female household heads in Addis Ababa, Ethiopia, in 2022.

Study setting
The study was carried out in Bole sub-city, Addis Ababa, Ethiopia.Addis Ababa is composed of eleven sub-cities with an estimated population of 5,006,000.Among these, 47.5% were males, and the remaining 52.5% were females [17] .

Study design and population
A community-based comparative cross-sectional study design was used to assess the urban health extension program package utilization and healthcare-seeking behaviour among model and non-model household heads.

Eligibility criteria
The study included female model and non-model HH heads over the age of 18, as well as those who had lived in the study area for more than a year.The study did, however, exclude female household heads that were seriously ill and unable to communicate.

Sample size determination and sampling procedure
The sample size was calculated using the two-population proportion formula by applying Epi-Info version 7.2.1 software with the following assumptions: 95% confidence interval, 5% margin of error, 80% power, 1:1 model-to-non-model ratio, design effect of 2, and 10% non-response rate.In the study conducted in Ethiopia, knowledge was 80.7% [11] and urban health extension package utilization was 66.6% [9] .Since studies were conducted only on model households and there is no literature on non-model households, we assumed 50% of the proportions for non-model households.Therefore, the final sample size was 594 households (297 model and 297 non-model household heads).The study participant model and non-model female household heads were selected by using a multistage sampling technique.The study area was divided into 15 districts at the initial stage; four districts were randomly selected by lottery methods.
In the second stage, because there was no kebele structure in the Addis Ababa city administration, four ketena (the lowest administrative units in a kebele) were chosen at random, and proportional sample size allocation was done in each ketena.The total number of model and non-model female household heads was then obtained from the woreda health office.A sampling frame was prepared for each model and non-model female household head, and the first households were chosen at random from a list of registrations listed by the names of household heads in each ketena.Using the first household as an index, a simple random sampling technique was used to obtain the required sample size.

Study variables and measurement
The dependent variables in this study were the use of urban health extension program packages and healthcare-seeking

Data collection procedures
The minister of health's urban health extension program implementation guidelines, literature, and the Ethiopian demographic health survey questionnaire were used to create a structured questionnaire [9][11] [17] .The data collection questionnaire was written in English first, then translated into Amharic.Before data collection, the data collection tools were pre-tested with 5% of the total sample size in Yeka sub-city, Addis Ababa, and modifications were made accordingly.Four data collectors and two supervisors participated, and one-day training was given.

Data management and analysis
All questionnaires were reviewed for completeness and errors before being entered into Epi Info version 7.2.1.0and SPSS version 26 software for analysis.Bivariate logistic regression analyses were used to identify potential factors related to the use of urban health extension packages.To control confounding factors and determine the relationship between independent and outcome variables, multivariable logistic regression analysis was used.The 95% confidence interval and a p-value less than 0.05 were used to assess the degree of association between dependent and independent variables.

Socio-demographic and economic characteristics of the respondents
A total of 587 participants were successfully interviewed, including 294 model and 293 non-model female household heads, for a response rate of 99%.The average age of the study participants was 36.46 years, with an SD of 8 years.
Both model and non-model female household heads had three children on average.The average monthly income of participants' model female household heads was 3671, with a standard deviation (SD) of 1184, while non-model female household heads earned 3524.91, with an SD of 1107 Ethiopian Birr (Table 1).

Knowledge status of households towards urban health extension packages
The majority of study participants, 264 (90%) model and 213 (73%) non-model female HH heads, had heard about the urban health extension program, and UHE-Ps were the source of information for 227 (77%) of model and 82 (28%) of nonmodel HH heads.
According to this, model female HH heads had good knowledge with a mean score of 221 (75%), whereas non-model female HH heads had moderate knowledge with a mean score of 181 (64%).non-model HHs had first aid kits in their homes, as did 199 (68%) model and 164 (56%) non-model HHs (Table 5).

Healthcare seeking behaviors of households
In general, the overall urban health extension package utilization was 416 (71%), of which 228 (78%) were model and 188 (64.2%) were non-model HH female HH heads.On the contrary, based on the study participants' responses, the main reasons for not implementing and utilizing UHE-Ps for the model HHs were 33 (50%) that some components are not important and for non-model HHs 45 (43%), which I do not know how to use (Table 5).

Factors associated with urban health extension program package utilization
The bivariate logistic regression analysis revealed that hearing (having information) about UHEPs, income, occupation, understanding of UHE-Ps, perception of service quality, being model HHs, home visits, and the frequency of home visits by urban health extension workers were all significantly related to the utilization of urban health extension program packages at a p-value of less than 0.25.However, variables like age, educational status, marital status, religion, and family size had no significant association with UHE-Ps utilization.
In the multivariable logistic regression analysis, only having information about UHEPs, frequency of home visits, understanding the UHE-Ps, and being model graduated HHs were predictors of UHE-Ps utilization at a p-value of less than 0.05.
As a result, participants in the study who were regularly contacted by urban health extension workers were more than twice as likely to utilize UHE-Ps (AOR = 2.12, 95% CI = 1.01 -3.13) than those who were not frequently visited.Model female HH heads who heard about urban health extension programs were more than two times more likely to utilize the UHEPs compared to their counterparts (AOR = 2.35, 95% CI = 1.08 -3.42).Model female HH heads who understood the urban health extension program packages were more than three times more likely to use the UHEPa (AOR = 3.14, 95% CI = 2.43 to 4.47) than those female HH heads who did not understand the packages.Moreover, model female HHs who graduated were nearly three times more likely to use the UHEPa than non-model HHs (AOR = 3.052, 95% CI = 2.024 to 5.113) (Table 6).

Discussion
This study attempted to assess and compare the utilization of urban health extension program packages by model and non-model female household heads.The overall knowledge status on UHEPa was moderate among participants, with model female household heads having a higher knowledge status than non-model female household heads.This disparity could be attributed to the presence of frequent home visits, during which UHE-Ps provided health education among model households.
In the current study, the most frequently mentioned UHE-Ps components by both model and non-model female household heads were immunization, latrine and excreta disposal, and solid waste disposal; on the other hand, the least frequently mentioned packages by both model and non-model female household heads were first aid, mental health, and malaria prevention and control activities.The findings are similar to those of a study conducted in Addis Ababa, Ethiopia [7][11] , in which solid waste, immunization, and latrine and excreta disposal were the three most mentioned packages, while first aid, malaria prevention and control, and mental health were the least mentioned.Furthermore, this consistency could be due to similarities in the study setting, socio-demographic characteristics, and a lack of attention.
The findings are also consistent with a systematic review conducted in Ethiopia from 2003 to 2018 on the success and challenges of health extension programs, which revealed that family planning, immunization, solid and liquid waste disposal, and latrine utilization were the most frequently mentioned packages [14] .The current knowledge status of households on urban health extension packages was higher than study findings from Gondar, and Hosanna town in Southern Ethiopia, where 65.3% and 42% of participants had good knowledge of UHE-Ps [15][16] .This inconsistency could be explained by differences in study settings and socio-demographic characteristics.
The study's findings were also consistent with a study conducted in the Hadiya Zone, South Ethiopia, where 68.3% of participants had good knowledge of UHE-Ps [18] , but lower than in Addis Ababa [11] .The absence of model HH training, a low commitment, and the current COVID-19 situation in AA restrict UHE-Ps' frequency of home visits to given health education activities, which could be reasons for this difference.The overall urban health extension program prioritizes the use of model female HH heads over non-model female HH heads.This finding was consistent with a systematic review conducted in Ethiopia from 2003 to 2018, which found that model HHs used more health extension packages than nonmodel HHs [19] .This disparity could be explained by the presence of frequent home visits, health education, and demonstrations of various packages at the household level during home visits.
The current study found that 29.6%, 59.5%, and 42% of participants in AA, Gondar Amhara region, and Hossana town, Hadiya Zone, South Ethiopia, use UHE-Ps [3][15] [16] .The current study's findings were nearly consistent with a study conducted in Bishoftu, Oromia region, which found that 72.8% of participants used UHE-Ps [20] , but lower than a study conducted two years ago in AA, where 86% of participants used UHE-Ps [11] .The absence of model HH training, the restriction of UHE-P home visits, the low commitment, and the lack of supportive supervision and feedback could all be reasons for this inconsistency.Evidence also indicated that one of the challenges to implementing and using HEP was the presence of limited supportive supervision [19][21] .Furthermore, the current study identified that, having information, understanding different package components, frequency of home visits, and model household graduation status were predictors of UHEPa utilization.
The current result is supported by the study conducted in AA [7][11] , west Gojjam zone, Amhara region [22] , Ambo town, Oromia region [23] , and the systematic review done in Ethiopia [14] , in which the understanding of the packages, frequency of home visits, being model graduated HHs, and monthly income were significantly associated with UHEPa utilization.The finding is also consistent with the other study carried out in Gondar, Amhara region, and the Sebeta Hawas district, Oromia region, which indicated that the understanding of packages was significantly associated with urban health extension services and maternal and child health package utilization [10][15][24] [25] In this study, the frequency of home visits was higher in model female HHs than in non-model female HHs.The results in the model female HH heads were consistent with the MOH UHEP implementation guideline [26][27] but lower than the results in Addis Ababa [11] and Hosanna town, Hadiya zone, south Ethiopia [18] .This disparity could be attributed to COVID-19's restriction of UHE-P home visits and the presence of a large disparity in the proportion of UHE-Ps to HHs; one UHE-P is expected to cover 500 HHs [26] .
According to the study participants' responses, the main reasons for not using the UHE-Ps were a lack of knowledge about some of the package components, some of the packages being unimportant, and some costing or requiring money.
According to a study conducted in the AA and Akaki districts of the Oromia region, the main reasons for not using packages were some components that were not important, were not prepared well, and required money [28][29] .

Limitations of the study
Because the study used a cross-sectional study design with only one point in time, observation and interview recall bias were possible, and it was difficult to identify a cause-and-effect relationship.

Conclusion
Based on the findings of this study, it can be concluded that household status, both model and non-model households, had an effect on UHE-Ps utilization.Understanding packages, frequent home visits, income, and being a model household graduate were significantly associated with UHE-Ps utilization.Therefore, providing model household training, frequent home visits, awareness creation on different components of packages, and giving more attention to disease prevention and control packages are essential to increasing UHE-Ps utilization of HHs.
Tables Table 1.Socio-demographic characteristics of the study participants Qeios, CC-BY 4.0 • Article, January 3, 2024 Qeios ID: H3F4OU • https://doi.org/10.32388/H3F4OU2/18 Based on community perception, the majority of model HHs study participants rated the quality of health services as very good (49.3%) and non-model HHs as good (49.5%)(Table3)..6.Disease prevention and control package utilization in households UHE-Ps' role in this package was to provide health education and make referral connections.Based on this, 229 (78%) model and 161 (55%) non-model HHs participants received tuberculosis health education.During coughing for more than two weeks, the majority of model HHs (217%) and non-model HHs (132%) visited the HF for diagnosis and treatment.