The factors influencing urban health in Jakarta Province during Covid-19 outbreak

The rapid urbanization in Jakarta has given rise to a wide range of urban problems, characterized by an imbalance between the population growth rate and urban resources. This imbalance has resulted in a decline in the quality of urban services, making the population highly venerable to diseases, particularly during and after the Covid-19 outbreak. Relevant studies have indicated that variables related to infrastructure, social and economic determinants can affect urban health. However, the literature exploring the relationship between health determinants and urban health from an urban planning perspective is still limited. The aim of the research is to examine the relationship between health determinants and urban health in Jakarta Province during the Covid-19 outbreak. The study employed Partial Least Square (PLS) path modeling using smartPLS software, while data were collected through questionnaires administered to 317 Jakarta residents. The results showed that urban health determinants, represented by the built environment/infrastructure and social determinants, significantly and positively impacted the physical health of Jakarta residents. However, the urban health determinants insignificant on the mental health of Jakarta residents during Covid-19. It was also found that built environment/infrastructure had a greater influence on physical health than other determinants. This impact was significant for both neighborhoods that are well-resourced and under-resourced. The influence of the social determinant on physical health significantly varied between these neighborhoods. This research emphasized the importance of stakeholders focusing on improving the population’s quality of life and health through the built environment/infrastructure and social determinants.


Introduction
High population density is a common occurrence in cities due to urbanization.The increase in urban population results in increased demand for urban services across various sectors.However, this creates an imbalance between the rate of population growth and urban resources, leading to a decline in the quality of urban services, such as limited and uneven accessibility to adequate infrastructure.According to the U.N., at least one-third of the global urban population faces inadequate living conditions, including high density, the spread of infectious diseases, and adverse environments that negatively impact the quality of life and health [1].Covid-19 is one of the infectious/communicable diseases that has had a global impact, specifically in Indonesia.Jakarta Province has the highest population density in Indonesia [2], with a population of 10,562,088 people, a population growth rate of 0.92% per year, and a population density of 14,555 people/km² [3].The increase in Covid-19 cases was also accompanied by a surge in cases in Jakarta Province, reaching its peak in July 2021 [4].To combat the spread of Covid-19, a range of policies has been implemented since 2020, including Indonesia's large-scale social restrictions (LSSR), Transitional LSSR, the Community Activities Restrictions Enforcement (CARE), Java-Bali CARE, Micro-scale CARE, Emergency Public Activity Restriction, and the New Policy of CARE levels 1-4.Apart from infectious diseases, the World Health Organization (WHO) reports that 63% of global deaths are attributed to non-communicable diseases (NCDs) like cardiovascular diseases, cancer, and diabetes, most of which are linked to risks associated with urban physical and social environments [1].In Indonesia, NCDs were the third leading cause of death in both men and women in 2017, with Ischemic Heart Disease, Stroke, and Diabetes Mellitus being the prominent diseases [5].Mental health disorders were also prevalent among urban residents [5].
Urban health entails the examination of urban attributes, which include both social and physical environmental elements, as well as the availability of resources and infrastructure that can impact health and disease within urban settings [6].The urban health framework for the Western Pacific region, spanning from 2016 to 2020, extends beyond the concept of "healthy cities" and underscores the significance of fostering "healthy and resilient cities."This approach takes a proactive stance toward promoting urban health [7].The determinants of urban health require a multilevel examination.Systematic literature reviews highlighted the impact of variables related to the built environment, social and economic determinants on urban health [6,[8][9][10][11].Although urban planning has the potential to influence urban health studies, the relationship between the two fields is relatively new and gaining attention from urban planners and health experts.Besides, the existing literature exploring the relationship between the scope/determinants of health, such as the built environment/infrastructure, social, as well as economic factors, and urban health in Jakarta Province from a planning perspective, is still limited.Therefore, this study aims to examine the relationship between health determinants and urban health in Jakarta Province during the Covid-19 outbreak.

Methods
This research utilized a quantitative method to investigate how health determinants associated with the built environment/infrastructure, along with social and economic factors, influence the health status of residents in Jakarta.This section offers a summary of the data collection and analysis.

Data collection
The primary data were obtained through questionnaires that were distributed online and offline in August 2022 -October 2022.The survey consisted of questions regarding individuals' self-evaluations of their physical and mental health statuses amid the Covid-19 pandemic, which were addressed independently.Respondents assessed the statements presented in Table 1 using a Likert scale, where they could choose from the following options: "1 = Strongly Disagree," "2 = Disagree," "3 = Neutral," "4 = Agree," and "5 = Strongly Agree".The availability and sanitation conditions affect my health condition I5.Availability and quality of clean water [6,10] The availability and quality of clean water affects my health condition I6.Availability of public transportation [6,[9][10]13] The availability of public transportation affects my health condition The respondents consisted of individuals between the ages of 15 and 64 who were residents of Jakarta Province and actively engaged in various activities within the region.The sampling technique used is simple random sampling [18].The sample size was determined using the Slovin formula (n = N / (1 + Ne²)) [19], with an error rate of 5.6%.The study included a total of 317 respondents (Table 2).The distribution of respondents is shown on Figure 1.The respondent profile showed that 19% of the respondents had contracted Covid-19, while 32% had contracted other communicable diseases, such as tuberculosis, malaria/dengue fever, and diarrhea during the pandemic.On the other hand, 49% reported they had not contracted any communicable diseases, and 26% confirmed they experienced stress/anxiety/depression during the pandemic.

Data analysis
This study employed the Partial Least Square (PLS) path modeling method, which served as an initial step in building theories through tests and validation of exploratory models [21].The analysis was conducted using smartPLS, a widely-used tool for PLS path modeling.PLS offers an advantage in generating more accurate predictions, specifically when the sample size is relatively small [21].The following were considered during the PLS path modeling analysis [21]: 1. Input of data and drawing of path models 2. Measurement model assessment a. Test validity: (a).Convergent validity: this is achieved when loading factors on the scale range from 0.5 to 0.6, indicating suitability for testing [22], and the Average Variance Extracted (AVE) should exceed 0.5 [22][23]; (b).Discriminant validity: assessed using the Fornell Larcker Criterion, where the square root of the AVE should be greater than the correlation value between latent variables, and cross-loading should be considered.b.Reliability test.A construct is deemed reliable when it possesses a composite reliability value exceeding 0.70 [21] and a Cronbach's alpha coefficient greater than 0.6 [24].These assessments confirm the reliability of the findings.

Results and Discussions
To determine the model's representation, several alternative model images were tested.The model can be further analyzed using Figure 2a, based on the results of the outer loading test.

Measurement of the outer model and inner model
This model met the outer model assumption, enabling subsequent analysis, while the reliability was confirmed by the test outcomes.The structural model underwent evaluation through the R-square value for the dependent variable and path coefficient values for the independent variables.Subsequently, it was scrutinized for significance based on the t-statistical value for each path.The R-square value signifies the determination coefficient for the endogenous construct, typically indicating strength at 0.67, moderate influence at 0.33, and weak influence at 0.19.The adjusted coefficient of determination (Rsquare adjusted) indicated the extent of influence the independent variable exerted on the dependent variable.
1.The adjusted R-square value of 0.213 revealed that the combined influence of the independent variable (built environment/infrastructure) and the mediation variable explained 21.3 percent of the dependent variable (physical health condition).The remaining variability was attributed to factors beyond the scope of the model.The relationship between urban health determinants and physical health condition was considered moderately weak.2. With an adjusted R-square value of 0.108, it was evident that the independent variable (built environment/infrastructure) contributed to 10.8 percent of the mediating variable represented by economic determinants.The remainder was accounted for by variables not encompassed within the model.3.An adjusted R-square value of 0.257 indicated that the independent variable (built environment/infrastructure) elucidated 25.7 percent of the mediating variable pertaining to social determinants.The remaining variance was attributed to external factors not considered in the model.4.An NFI value of 0.480 was obtained, indicating a 48.0%fit for the model in this study (Figure 3).

Generated model of factors influencing physical health of Jakarta residents
Based on the model (Figure 4 1.The built environment/infrastructure directly and positively affected the physical health of Jakarta residents with a coefficient of 0.332 and a p-value of 0.000 < alpha of 0.05.This indicated that a 1-point increase in the determinant improved physical health by 0.332, while other variables were held constant.2. The social determinant directly and positively affected physical health with a coefficient of 0.199 and a p-value of 0.007 < alpha of 0.05.This indicated that a 1-point increase in the determinant improved physical health by 0.199, while other variables were held constant.
3. The economic determinant had no effect on physical health with a coefficient of 0.007 and a pvalue of 0.920 > alpha of 0.05.This indicated the evidence was not enough to conclude that an increase in economic determinants improved the physical health of Jakarta residents.

Multi-group analysis between neighborhoods that are well-resourced and under-resourced
Multi Group Analysis (MGA) is used when research experts have categorical/dummy variables to differentiate between existing models.In the case of the mediation model mentioned above, experts were interested in examining how the relationship between variables affected the dummy variables of neighborhoods (well-resourced and under-resourced).The objective was to identify the potential differences in the influence of settlement categories on Health, as mediated by economic and social determinants.When examining the impact based on settlement categories, it becomes evident that the impact between neighborhoods that are well-resourced and under-resourced is significantly different.
The MGA results showed that the influence of built environment/infrastructure determinants on physical health was significant for both neighborhoods that are well-resourced and under-resourced (Figure 5).This implied that the provision of infrastructure was essential for improving the health condition of Jakarta residents.The influence of social determinant on physical health was significant for the neighborhoods that are well-resourced.The direct impact of social determinant on physical health between both neighborhoods was statistically proven to be different.This result holds significant importance in this study.On the other hand, the specific indirect effects between built environment, social and physical health were significant.This section serves as a note for urban planners and stakeholders, emphasizing the need to tailor the provision and maintenance of infrastructure to the specific needs of residents in both neighborhoods.

Measurement of the outer model and inner model
This outer model successfully fulfilled the assumption based on the rule of thumb for the convergent validity (identification of loading factor and average variance extracted).The coefficient of determination (R-square adjusted) showed the extent of influence the independent variable had on the dependent variable.
1.The adjusted R-square value of 0.021 indicated that the independent variable (built environment/infrastructure) and the mediating variables could collectively account for 2.1 percent of the dependent variable (mental health condition).The remaining variance was attributed to other variables not included in the model.This signified a very weak relationship between urban health determinants and mental health outcomes.2.An adjusted R-square value of 0.734 showed that the independent variable (built environment/infrastructure) could explain 73.4 percent of the mediating variable, which took the form of economic determinants.The remaining variance was ascribed to external variables beyond the scope of the model.3.An adjusted R-square value of 0.613 indicated that the independent variable (built environment/infrastructure) could account for 61.3 percent of the mediating variable, represented by social determinants.The residual variance was explained by factors not considered within the model.4. Based on the fit model, the NFI value of 0.872 indicated the model was fit at 87.2% (Figure 6).

Generated model of factors influencing mental health of Jakarta residents
Based on the model (Figure 7 Based on the outer and inner model tests, all variables from the built environment/infrastructure, social and economic determinants were included.The interpretation of the regression equation above is as follows: 1.The built environment/infrastructure determinant had no effect on the mental health of Jakarta residents with a coefficient of 0.202 and a p-value of 0.072 > alpha of 0.05.This indicated there was no enough evidence to prove that the increase in built environment/infrastructure improved mental health.2. The social determinant had no effect on mental health with a coefficient of 0.018 and a p-value of 0.439 > alpha of 0.05.This indicated there was no enough evidence to prove that an increase in social determinant improved mental health.3. The economic determinant had no effect on mental health with a coefficient of -0.050 and a pvalue is 0.351 > alpha of 0.05.This indicated there was no enough evidence to prove that an increase in economic determinant improved mental health condition.

Multi-group analysis between neighborhoods that are well-resourced and under-resourced
The results showed that the influence of built environment/infrastructure determinant on the economic and social determinants were significant for both neighborhoods that are well-resourced and underresourced (Figure 8).This indicated that the provision of infrastructure was crucial for improving the economic and social conditions of the community.On the other hand, the influence of infrastructure determinants on mental health was significant for the neighborhoods that are well-resourced.The direct impact of infrastructure determinants on mental health between both neighborhoods had been statistically proven to be significantly different.This result holds a significant importance in this study.However, the specific indirect effects of economic and social determinants were insignificantly different.

Discusssions
Urban resources, including features of the built environment, social environment, as well as economic are considered determinants that can influence justice in realizing the goals of urban health [25].
Previous studies have primarily focused on examining the statistical relationship of variables in one to two determinants of health outcomes in general or within a specific scope of health outcomes (communicable disease, non-communicable disease, or mental illness) [6,8,13,[15][16].Nonetheless, this study delves into the interplay among three health determinants, such as the built environment/infrastructure, social determinants, and economic determinants, with a focus on the physical and mental well-being of Jakarta residents amid the Covid-19 pandemic.The findings from both models unveiled a moderately weak connection between urban health determinants and physical health.However, the link between urban health determinants and mental health appeared to be statistically insignificant.This lack of significance may be attributed to the intricate and diverse nature of cities, where numerous factors can impact the overall health of the population.Evaluating how the urban environment influences health presented challenges and introduced complexities that cannot be easily resolved through the application of straightforward analytical methods [8].Additionally, cities are dynamic and undergo transformations, with the Covid-19 outbreak serving as a disruptive force in urban areas.This phenomenon had implications for the varying contributions of different factors in shaping health within cities [8].The built environment/infrastructure emerged as the most influential determinant of the physical health of Jakarta residents.This conclusion was in line with previous studies, highlighting the impact of housing conditions [6,10], availability of public transportation [6,[9][10]13], availability of pedestrian facilities [9], and conditions of place of activity [9,14] on urban health.The Covid-19 pandemic offers an opportunity to test and implement built environment and infrastructure interventions aimed at increasing health equity and decreasing health risks [26].

Conclusion
The outer model assumption was fulfilled, indicating the validity of the two models employed.Based on the inner model analysis, the direct influence of the urban health determinants on urban health was moderately weak (21.3%).The two determinants of urban health, namely the built environment/ infrastructure and social determinant, significantly and positively affected the physical health of the residents.The resulting equation is as follows: Y = 0.332 Built environment/Infrastructure + 0.199 Social + 0.007 Economic.Each increase in social determinant was influenced by 0.509 of the built environment/infrastructure determinants, while each increase in economic determinant was influenced by 0.333 of the built environment/infrastructure determinants.The built environment/infrastructure, represented by housing condition (I1), availability of public transportation (I6), availability of pedestrian facilities (I7), and conditions of place of activity (I9) variables, exhibited a greater influence on physical health than others.The social determinant, which is represented by variables of social network (S2) and education level (S3) also significant and positively affects the physical health of Jakarta residents.The impact of built environment/infrastructure determinant on physical health was significant for both neighborhoods that are well-resourced and under-resourced, while the influence of social determinant on physical health was significant only for the neighborhoods that are well-resourced.Regarding the inner model, the direct influence of urban health determinants on mental health was found to be very weak (2.1%), and none of the determinants significantly affected the mental health of Jakarta residents.
The MGA result revealed that the influence of built environment/infrastructure determinant on economic and social outcomes was significant for both neighborhoods.Moreover, the influence of this determinant on mental health was significant only for the neighborhoods that are well-resourced.This study statistically proved that the direct influence of built environment/infrastructure determinant on mental health significantly differed between neighborhoods that are well-resourced and underresourced.

Figure 1 .
Figure 1.Map of distribution of research respondents.

3 .
Structural assessment of the model Inner Model Test: examining the R-square (percentage of indicators successfully described by the model), Path Coefficients (direction of positive or negative relationships), T-Statistics (Bootstrapping), Predictive Relevance (where blindfolding above 0 indicates the model is relevant for predicting factors), and Model Fit (model quality).4. Forming equations based on models.

Figure 3 .
Figure 3. Outer and inner model results.

Figure 4 .
Figure 4. Generated model of factors influencing physical health of Jakarta residents.

Figure 6 .
Figure 6.Outer and inner model results.

Figure 7 .
Figure 7. Generated model of factors influencing mental health of Jakarta residents.

Table 1 .
Determining variables to build a model.

Table 1 .
Determining variables to build a model (cont.).