Determinants of Livable Houses in Magelang Regency for Households with Stunted Children under Five Years of Age

Stunting in children under five years of age is a nutritional issue that many countries, including Indonesia, have yet to successfully address. According to Basic Health Research or Riset Kesehatan Dasar (Riskesdas) data from 2018, Magelang Regency has the second highest prevalence of stunted children under five in Jawa Tengah Province, at 37.9 percent. Environmental factors such as the availability of healthy and livable houses have an influence on this level. The purpose of this study is to examine the relationship between the availability of livable houses and the prevalence of stunted children under five in Magelang Regency, as well as the factors that influence the issue. The data used in the study were obtained from interviews conducted in 2019 and 2023, as well as data from Statistics Indonesia (BPS). The research employed descriptive analysis combined with inferential analysis using binary logistic regression to achieve the objectives of the study. The findings reveal that livable houses in Magelang Regency had a negative relationship with households with incidences of stunted children under five years old. Regional classification, waste management sites, sanitation constraints, perceptions of waste handling, and the head of the household’s (KRT) field of business were significant determinants of the availability of livable houses.


Introduction
The second of the Sustainable Development Goals (SDGs) is the commitment to eliminate hunger, improve food security and nutrition, and promote sustainable agriculture [1][2][3][4][5].One of the targets of this goal is to eliminate all types of malnutrition, and especially to reduce stunting and wasting in children aged under five.This target became a priority because data from the World Health Organization (WHO) revealed that in 2017 global stunting constituted the largest prevalence of nutritional problems, at 22.2 percent [6][7][8].The environment is listed as one of the determinants that influences children's survival in the theoretical framework of Mosley and Chen [9][10][11][12].Children are more likely to become ill and eventually die due to poor environmental quality, whereas they will grow up healthily with good environmental quality.Blum proposed a similar theory about the conceptual framework of health, which proposes that environmental factors, whether physical, biological, or social, all affect the health status of the population at the same time [13][14].Many more studies also obtained the same result, that the environment has an influence on the prevalence of stunted children aged under five [15][16][17].
Stunting is one of the problems that hinders human development, especially amongst the young.Stunted children under the age of five are more of a concern because the condition makes them more susceptible to disease and puts them at risk of degenerative diseases when they become adults.As well as on health, stunting also has a negative effect on children's IQ, which will have an impact on their socio economy in the future.Therefore, it is very important to take preventive measures so that stunting does not occur in children aged under five.
The Ministry of Health published data from Riskesdas 2007-2018 which revealed that except for 2013, the prevalence of stunted children under five fell each year, with 30.8 percent of children under five experiencing stunting in 2018.The latest data from the integration of the March 2019 National Socio-Economic Survey (Susenas) and the 2019 Indonesian toddler nutritional status study atau Survei Status Gizi Balita Indonesia (SSGBI) showed that the prevalence of stunting was 27.67 percent, a decrease of 3.1 percent from the previous year.Despite the declining level, it still falls short of the national mediumterm development plan, or Rencana Pembangunan Jangka Menengah Nasional/RPJMN 2020-2024 target, which aims for a national stunting rate of 19 percent by 2024 [13][14].
According to Riskesdas data from 2018, Jawa Tengah Province has a higher prevalence of stunted children under five than the national average.In addition, with a prevalence of 37.9 percent, Magelang Regency ranks second in Jawa Tengah Province for the highest number of such children.This situation emphasizes the need for strategies in the region to prevent and control stunted children under five.
Children under five spend most of their time at home, so the availability of healthy and livable accommodation is vital to support their growth and development [15][16][17][18][19].According to BPS, a livable house should meet four criteria, namely a minimum area of 7.2 m2 or sufficient living space; access to proper drinking water; access to proper sanitation; and durability [20].
The purpose of this study is to determine the relationship between the availability of livable houses and the prevalence of stunted children under five, and to analyze the determinants of the availability of such houses for households with stunted children under five in Tegalrejo District, Magelang Regency.

Methodology
The study used data collected from interviews conducted in 2019 and 2023.The sample frame was taken from secondary data from the Magelang Regency Health Office and the sample was chosen using a twostage sampling method (multistage sampling).The number of population is 1.426 households and sample is 143 households.Each village was stratified in the first stage based on the proportion of stunted children under the age of five, and then villages were selected using a table of random numbers/TAR.The second stage then determined the number of households with stunted under-fives for each stratum by allocating a proportionate number of samples to each village.Other data used in the study were from BPS on the number of livable houses by province in Indonesia.
Descriptive analysis was used to explain the environmental conditions in households with stunted children under five years old, and inferential analysis employed to establish the determinants of a livable environment in such households in Tegalrejo District, Magelang Regency.Tegalrejo sub-district was chosen after considering toddler data from the Magelang Regency Health Office.
The independent variables used were regional classification, waste management sites/TPS, sanitation constraints, waste handling perceptions, KRT business fields, and officer's attention.The availability of livable houses is the dependent variable, with per capita area of residence, proper sanitation, proper drinking water, and building resilience indicators.
The logistic regression model used was as follows: The research hypotheses were: H0: There is no effect of area classification, availability of TPS, sanitation constraints, perception of waste handling, KRT business field, and officer's attention on the availability of livable houses.H1: There is an effect of regional classification, availability of TPS, sanitation constraints, perceptions of waste handling, KRT business fields, and the attention of officers on the availability of livable houses.

Results
Figure 1 shows that the relationship between the incidence of stunted children under five years old and the availability of livable houses is negative.In other words, the higher the number of livable houses, the lower the prevalence of stunted children under five years old.The results of the regression equation are y = 33.02-0.05x.This means that every one percent increase in the number of livable houses will reduce the incidence of stunted children under five years old by 0.05 percent.
Based on Figure 2, the majority (61 percent) of the households with stunted children under five years old in the sample already own a house with livable conditions.The remaining 39 percent, or 4 out of 10 households, do not yet have a suitable house to live in.This situation shows that the quality of housing in families with stunted children under five years old is quite good.
Furthermore, based on their socio-demographic characteristics, 66.30 percent of households with stunted children under 5 years old living in rural areas have access reside in livable houses, while the corresponding figure in urban areas is 48.60 percent.Based on the main occupation of the head of the household, 64.30 percent of households whose KRT works in the non-agricultural sector have access reside in livable houses, while the figure for those whose KRT works in the agricultural sector is 41.20 percent.Based on environmental characteristics, the number of households that have good knowledge of managing waste have access reside in livable houses is greater than those with no knowledge (64.90 percent compared to 33.30 percent).67.70 percent of households live in environments where there are no waste management facilities have access reside in livable houses, while 42.40 percent of households do have waste management facilities.In terms of support from health workers/instructors, 62.60 percent of households who received attention and support from them regarding the concept of basic sanitation and healthy homes had access reside in livable houses, while 50.00 percent received no attention or support.
Based on the obstacles faced in realizing basic sanitation and healthy homes, households with stunted children under 5 years old who do not experience obstacles in realizing basic sanitation and healthy homes have access reside in livable houses by 68.40 percent, while 50 percent of household do have obstacles to access reside in livable houses.It can therefore be said that based on the socio-demographic and environmental characteristics, the percentage of households with stunted children under 5 years old who reside in livable houses is greater in rural areas; household heads mainly have jobs in the nonagricultural sector; there is no misperception in handling waste; in the environment there are no waste management facilities; there are no problems in realizing basic sanitation; and attention is received from health workers regarding basic sanitation and a healthy environment.
Furthermore, the results of the logistic regression test, both simultaneously and partially, produce a synthesis that there are four variables that significantly affect access to livable housing at a significance level of 5 percent.In addition, one significant variable affects access to livable housing at a significance level of 10 percent.Table 2 shows that regional classification, availability of TPS, sanitation constraints, and perceptions of waste management significantly affect access to livable houses at a level of 5 percent.Meanwhile, the occupation of the head of the household significantly affects access to livable houses at a level of 10 percent.In Table 2, it can be seen that the direction of the relationship between each determinant of the existence of livable housing is negative for the variables of regional classification, waste management sites, perceptions of waste handling and KRT occupation.Only the sanitation constraint variable has a positive relationship.The resulting regression equation is as follows: g(x) = 0,926 -1,345 X1 -1,332 X2 + 1,078 X3 -1,487 X4 -1,081 X5 (2) worse housing problems than urban ones [16][17][18].Data related to this also strengthens the notion that urban areas have higher access to livable houses than rural areas [19].However, research indicates that there are many slum areas and inadequate housing in urban areas due to urbanization and poverty [20][21][22].
Moreover, in such locations, it is known that the lowest indicator of livable housing is access to safe drinking water; such access in rural areas is relatively easier.
( The existence of waste management sites and household perceptions of waste management also affect access to livable houses.Households with stunted children under five year who have a slight misperception of waste handling are almost five times more likely to access livable houses.This is thought to be related to household knowledge of waste management, which has an impact on the condition of the house they live in.Notoatmodjo (2007), cited in Kusumawati et al. [17], and Suwita et al. [9] argue that knowledge is the basis for a person to behave more lastingly.Yulianti et al. also state that knowledge is the main basis for creating a healthy home [23].Therefore, the perception of waste handling, which is influenced by knowledge, has a big influence on access to livable houses.This is in line with the findings of Puspita et al., who found that the behavior of residents who are not good at processing inorganic waste is influenced by a lack of knowledge and understanding of how to treat waste properly [24].In addition to perceptions of waste handling, the existence of waste management site also affects access to livable houses.
The next determinant of access to livable houses is the existence of obstacles to the realization of suitable sanitation.This is one of the indicators of livable houses.Therefore, it is reasonable that such obstacles are significant obstacle in the creation of livable houses.Households with stunted children under five who face obstacles to achieving proper sanitation are three times less likely to own a livable house.If explored further, the problem is mostly caused by the costs involved.This is in accordance with Gunawan (2006), cited in Puspita (2019), that sanitation management is influenced by several factors, including income [24].
The final determinant that significantly affects access to livable houses is the occupation of the head of the household.Those who work in the non-agricultural sector are three times more likely to have access livable houses than those who work in the agricultural sector.This is thought to be influenced by the income factor as well as the knowledge factor [9,23,24].Agricultural households tend to have the lowest incomes.BPS noted that the agricultural sector is that with the lowest incomes [19,25].Relatively low income, either directly or indirectly, will affect the ability of households to access livable houses.

Conclusion and suggestions
Based on the results and the discussion, the study empirically shows that the relationship between the incidence of stunted under fives and the existence of livable houses by province in Indonesia in 2018 is negative.The regression equation shows that for every increase of the presence of a livable house, there will be a 0.05 decrease in the incidence of stunted children under five.This means that the higher the availability of livable houses, the lower the prevalence of stunted in children under five.
Furthermore, the study also shows that the environmental conditions in households with stunted children under five in Magelang Regency are quite good.The determinants of the availability of livable houses that are significant indicate the direction of the negative relationship between the regional classification variables, waste management sites, perceptions of waste handling and KRT occupation, while the sanitation constraint variable shows a positive relationship.
Based on the various study findings, policy suggestions for the government, through its relevant agencies, include continuing to make efforts to increase knowledge amongst the population about environmental health, especially in the efforts to create livable houses.Providing information on the requirements for livable houses is important.Various programs, such as suitable sanitation counseling and sanitation surveys by health workers and village officials, for the development of adequate sanitation in households could constitute government intervention efforts in accelerating access to suitable sanitation in Magelang Regency.
In further studies, it would be interesting to make an in-depth analysis of the current conditions of households with stunted children under five in Magelang Regency.The use of a new analytical approach regarding non-environmental factors that influence the availability of livable houses and their relationship to the incidence of stunted under fives can be more interesting to study.

Figure 1 .
Figure 1.Relationship between Stunted Children Under Five Years Old and Access to Livable Houses by Province in Indonesia Source: Riskesdas 2018 and Susenas 2018 data (processed)

Figure 2 .
Figure 2. Percentage of Households with Stunted Children under Five Years Old According to the Number of Livable Houses in Magelang Regency Source: Results of the 2023 data processing

Table 1 .
Percentage of Households by Access to Livable Houses in Magelang Regency Based on the logistic regression, it was concluded that households with stunted children under five living in urban areas had a lower chance of accessing livable houses, whereas those in rural areas were almost four times more likely to have access.This is different from most studies who found that rural areas had

Table 2 .
Logistics Regression Test Results.