Wealth-based and rural–urban disparities in digestive diseases among indonesian adults: evidence from a cross-sectional analysis of the Indonesia Family Life Survey (IFLS-5)

Abstract

Background

Digestive diseases represent an important public health concern in Indonesia, with national surveys reporting notable symptom burden and frequent hospital presentations for conditions such as gastritis, dyspepsia, and gastroesophageal reflux disease (GERD). However, evidence on their social and geographic disparities remains limited. Understanding how socioeconomic status and place of residence are associated with digestive health can guide equitable policy interventions.

Objective

This study examined socioeconomic and rural–urban disparities in digestive diseases among Indonesian adults, using nationally representative data from the Indonesia Family Life Survey (IFLS-5).

Methods

A cross-sectional analysis was conducted among 29,817 adults aged 15 years and above using IFLS-5. Digestive disease was defined based on self-reported doctor-diagnosed disease diagnosed gastrointestinal conditions using IFLS-5 item ‘stomach or other digestive disease”. Socioeconomic status was primarily assessed using an asset-based household wealth index derived by principal component analysis. Multivariable logistic regression was used to estimate adjusted odds ratio (aOR) for digestive disease factors, controlling for sociodemographic, health related factors, behaviours, lifestyle, and environmental factors, including depressive symptoms.

Results

The overall weighted prevalence of doctor-diagnosed digestive disease was 13.11%. The prevalence was higher in urban residents (14.8%) than rural areas (11.4%), and among non-poor (14.2%) than poor adults (12.1%). These differences were statistically significant (P < 0.001). After adjustment, rural adults had lower odds of digestive disease (aOR = 0.88, 95% CI: 0.79–0.98), while wealth index was not significantly associated (aOR = 1.02, 95% CI: 0.94–1.12). Higher education attainment was strongly associated with increased odds of digestive diseases (aOR 1.78, 95% CI: 1.62–1.94) compared with lower education. Female sex, former smoking, comorbid conditions, depressive symptoms, and poor self-rated health were also positively associated.

Conclusions

In Indonesia, notable differences in digestive diseases were observed across educational and residence groups. Although crude prevalence was higher among urban and non-poor, adjusted analysis revealed persistent disparities mainly driven by education and place of residence rather than household wealth. Reducing the burden of digestive diseases requires policy initiatives to enhance access to healthcare, nutrition, and sanitation, especially in rural and lower-education communities.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-26811-9.

Introduction

Digestive diseases, ranging from enteric infections and upper gastrointestinal disorders such as gastritis, dyspepsia and gastroesophageal reflux disease (GERD) contribute substantially to global morbidity and mortality. In 2019, these conditions accounted for millions of deaths and hundreds of millions of disability-adjusted life-years (DALYs) worldwide [1–3]. Projections suggest that noncommunicable digestive diseases will remain dominant causes of health loss through 2050 without stronger interventions [4]. Asia bears a disproportionate share of digestive disease burden, driven largely by persistent enteric infections and rising noncommunicable gastrointestinal diseases linked to poor sanitation and hygiene [5–7].

These conditions intersect with modifiable behavioural and psychosocial risk factors, such as smoking, alcohol, obesity, dietary patterns, and mental health conditions such as anxiety and depression, all of which affect quality of life and healthcare utilization [2, 5]. Together, this burden highlights digestive disease as a priority for integrated prevention, diagnosis, and equitable care, especially in countries like Indonesia where infectious and chronic conditions overlap [5, 6].

In Indonesia, digestive diseases represent a significant yet often overlooked health concern. The Basic Health Research Survey [8] found that 9.6% of Indonesians reported digestive disease symptoms, with a higher frequency in middle-aged and urban areas. According to more current data from the 2023 Indonesian Health Survey [9], digestive diseases remain among the top 10 reasons for hospitalization, especially for disorders like dyspepsia and gastritis. Indonesia faces a dual burden of digestive diseases, with persistent enteric infections alongside rising noncommunicable digestive diseases, reflecting the country’s epidemiological shift from infectious to chronic causes of DALYs [10].

Indonesia has a relatively low Helicobacter pylori prevalence (22.1%), in contrast to many neighbouring South-east Asian and East Asian nations (e.g., Thailand 21–54%, Vietnam ~ 50–70%, Philippines ~ 34%, China ~ 44%, Japan and Korea ~ 38–51%). The historically lower rates of gastric ulcers and gastric cancer seen in Indonesia are partially due to this decreased infection prevalence [11–13]. However, enteric infections remain widespread, with diarrheal disease linked to sanitation deficits, crowding, and antimicrobial resistance [14, 15]. These challenges intersect with lifestyle risks such as high male smoking prevalence and poor diets, as well as unequal healthcare access across regions [10, 16]. Despite the expansion of the national insurance program Jaminan Kesehatan Nasional (JKN), disparities in coverage and service utilization persist, influencing timely diagnosis and management of digestive diseases [17, 18].

According to the World Health Organization's Social Determinants of Health framework, socioeconomic status and place of residence are fundamental structural and material conditions determining exposure, vulnerability, and access to care [19]. A lower socioeconomic status is consistently associated with limited health literacy, increased dietary risks, higher smoking prevalence, poorer sanitation, and lack of timely diagnosis all contributing factors to gastritis, dyspepsia, and other gastrointestinal disorders [20, 21]. Place of residence similarly shapes the risk for digestive disease: urbanization has increased consumption of processed foods, sedentary behaviour, and psychosocial stress that are linked to functional gastrointestinal disorders, while rural populations are experiencing persistent underdiagnosis due to limited health infrastructure and diagnostic capacity [8, 22–24]. Socioeconomic and geographic pathways into digestive disease are of particular relevance in Indonesia, where rapid urbanization, an unequal distribution of health resources, and environmental deficits persistently create heterogeneous patterns of digestive disease burden across communities.

Prior studies in Indonesia and the broader region consistently link lower socioeconomic status, measured by income, education, and housing conditions to higher risk of enteric illness, particularly childhood diarrhoea [25, 26]. Lifestyle risks, especially tobacco use and poor diet, are major contributors to digestive symptoms and lower quality of life [10, 27], and mental health comorbidities amplify the severity of functional gastrointestinal disorders and impair daily functioning [28]. Recent evidence from Indonesia, using IFLS-5 data shows that poor water and sanitation conditions is significantly associated with mental health outcomes, particularly depressive symptoms, across both urban and rural settings [29]. This highlights the interconnectedness of environmental and health determinants, whereby deficiencies in water, sanitation, and safe hygiene (WASH), and food safety further elevated the risk of diarrhoeal and other digestive diseases [30, 31]. Moreover, despite high nominal coverage under JKN, persistent disparities in healthcare utilization and financial protection by residence and wealth status affect timely diagnosis and treatment of digestive conditions [16, 32].

Although Indonesia faces both infectious enteric diseases and rising noncommunicable gastrointestinal disorders, existing evidence is fragmented, largely paediatric, hospital-based, or locality-specific, leaving little nationally representative insight on adult, clinically diagnosed patterns [33]. Research has linked diarrheal morbidity to poor water, sanitation, and hygiene and separately associated lifestyle and demographic factors with upper-digestive symptoms, yet these domains are rarely analyzed together for adults [34, 35]. Evidence linking socioeconomic, behavioural, mental-health, and environmental factors within a single framework remains limited. The Indonesia Family Life Survey (IFLS-5) was carried out by the RAND Corporation and Universitas Gadjah Mada in 2014–2015 and provides a unique opportunity to address this gap through its comprehensive, population-based data on socioeconomic and health [36].

To our knowledge, this is the first nationally representative study using IFLS to examine doctor-diagnosed digestive diseases as a primary adult health outcome. Prior surveys only focus on diarrhoeal illness, children, or included digestive conditions as part of broader multimorbidity profiles. This study aims to assess socioeconomic and rural–urban disparities in the digestive diseases among Indonesian adults and to identify associated behavioural, mental health and environmental factors using IFLS-5 data. These findings are intended to inform equitable public health interventions and future national monitoring efforts.

Method

Study design and data sources

According to the guidelines for cross-sectional studies provided by Strengthening the Reporting of Observational Studies in Epidemiology (STROBE), this study has been reported. [See Supplementary File1: Table 1A, Supplementary data].

The Indonesia Family Life study (IFLS) is one of the longitudinal surveys in Indonesia, covering approximately over 30,00 individuals, across 13 provinces, four on Sumatra (North Sumatra, West Sumatra, South Sumatra, and Lampung), all five of the Javanese provinces (DKI Jakarta, West Java, Central Java, DI Yogyakarta, and East Java), and four provinces covering the remaining major island groups (Bali, West Nusa Tenggara, South Kalimantan, and South Sulawesi), representing around 83% of Indonesian population [37]. The survey collected information about individual respondents, their families, homes, communities, and the health and educational resources they utilise. In this cross-sectional study, multicentre data performed in 2014–2015 consists of data from interviews with 16,204 households and 50,148 individuals. IFLS-5 data was conducted by the RAND Corporation and Universitas Gadjah Mada in 2014–2015.

Study population

For this analysis, 34,251 respondents had complete data on digestive disease diagnosis. After excluding participants younger than 15 years and those with missing information on key study variables, 29,817 individuals were included in the final analytic sample as shown in Fig. 1.

Fig. 1.

Flowchart of Study Sample Selection from IFLS-5 Respondents

Variable classifications

Outcome measure

The assessment of digestive disease in the IFLS-5 was based on multiple-condition chronic disease question in which participants were asked whether they had ever been informed by a physician, paramedic, nurse, or midwife that they had one of the several chronic conditions which referred to as doctor-diagnosed digestive disease. Among listed conditions, one particularly item captured the gastrointestinal disease: “stomach or other digestive disease.” This condition is recorded apart from both liver and kidney diseases; thus, only the gastrointestinal component is taken as the outcome variable in the present study [36].

Although IFLS does not provide ICD-coded detail for “stomach or other digestive disease,” evidence from Indonesian primary care and hospital settings shows that this category predominantly reflects common gastrointestinal diagnoses such as gastritis, dyspepsia, gastroesophageal reflux diseases (GERD), and peptic ulcer disease conditions that constitute the majority of GI-related clinical encounters in Indonesia [33, 38–40]. The single diagnosis of doctor-diagnosed gastrointestinal morbidity item is indeed in line with the methodology practiced in healthcare surveys like the US National Health Interview Survey, where physician-diagnosed chronic conditions are recorded with the help of composite self-report items [41, 42].

This measurement indicating physician-diagnosed disease, it is more so chronic or recurrent gastrointestinal disorders than short-term gastrointestinal symptoms captured somewhere else in IFLS, which are represented by this measure. The classification of these chronic gastrointestinal conditions into a single binary outcome is justified methodologically since they share similar overlapping behavioral, dietary, infectious, and psychosocial risk factors and frequently coexist in Asian populations [43, 44].

The IFLS-5 health module also includes questions on acute gastrointestinal symptoms (stomach ache, nausea, diarrhoea in the past four weeks, for example). These questions, however, asses’ acute statuses more than diagnosed illness. Therefore, we used the doctor-diagnosed digestive disease variable for chronic or recurrent gastrointestinal illness. The presence of acute symptom data is noted but was not included in the outcome variable.

The IFLS-5 question that used to capture chronic conditions, including the digestive disease item, is available in Supplementary File 1 in both English and Bahasa Indonesia.

Independent variables

The primary independent variables were Socioeconomic status and residence location. Socioeconomic status was measured using a household wealth index constructed via principal component analysis (PCA) of asset ownership variables and categorized as “poor” and “non-poor.” Residence was classified as “rural” or “urban” based on IFLS community codes.

Covariate

Demographic data included age, sex, marital status, and education level was classified into less than high school or more. Life style factors such as history of smoking was categorized to (current smoking, former smoking, and non-smoking). A shortened version of the "International Physical Activity Questionnaire (IPAQ)" short version (IPAQ-S7S) was used to measure physical activity during the previous seven days. The IPAQ scoring procedure classified physical activity as (low, moderate, and high) [45].

Health related factor

Health related factor was measure by using Self-reported hearth which was categorized to (healthy and unhealthy). Multimorbidity was defined as the presence of two or more chronic conditions in an individual at the same time and it was assessed through a self-report questionnaire with the question, “Has a doctor/paramedic/nurse/midwife ever told you that you had the following chronic conditions of disease?,” with response options: hypertension, diabetes, asthma, chronic lung diseases, cardiac diseases (heart attack/coronary heart disease/angina or other heart diseases), liver diseases, stroke, cancer or malignancies, arthritis/rheumatism, uric acid/gout, hypercholesterolemia, prostate illness, kidney diseases. According to WHO, [46], body mass index (BMI) was categorized, underweight (less than 18.5 kg/m²), normal weight (18.5–22.9 kg/m²), overweight (23.0–24.9 kg/m²), and obese (more than 25.0 kg/m²). The Centre for Epidemiologic Studies depressive Scale (CES-D), a 10-question test that looks at depressive symptoms, was shortened and used to quantify the existence of depression. Significant depressive symptoms were indicated by a score of 11 or above on the CES-D, which was used in this investigation as a dichotomous categorical variable [47].

Dietary patterns

This study evaluated dietary patterns by the frequency of consumption of each type of food, and concentrating on four areas: protein vegetables, fruits intake and the consumption of unhealthy snacks and all of them were divided into two categories: fewer than seven times per week and equal to or more than seven times per week [48]. This suggests that these foods should be included in a daily diet. Fast food, fizzy drinks, fried snacks, quick noodles, and sugary snacks are all considered unhealthy snacks in this study.

Environmental factors

Environment factors were assed using water source and water quality assesses the main source of drinking water as well as boiling of water prior to consumption. Water source classification follows WHO/UNICEF Joint Monitoring Programme (JMP) classification [49, 50], where water sources are identified under improved (e.g. piped, well with pump, spring, rainwater, bottled) or unimproved (e.g., unprotected wells, river water, pond, etc.) sources. Drinking water (water treatment) was divided as (boiled or unboiled). Sanitation facilities were classified by household toilet type, following JMP guidelines [49, 51], as basic/indoors, limited/shared, unimproved/outdoors, and open defecation. As mentioned in WHO guidelines [52], household drainage sewage was classified as improved (e.g., permanent pit, disposal in yard/garden) and unimproved (e.g., shallow drainage ditch, drainage to river, pond, field, and the sea). As per UNEP guidelines [53], garbage disposal was categorized as environmentally friendly (e.g., trash can, composting, pit disposal) or environmentally unfriendly (e.g., burning, dump into rivers, forests, and fields).

All covariates included in the models are defined in detail in Supplementary File, including coding strategy.

Statistical analysis

Distribution of Digestive diseases in Indonesia (IFLS_5) map was created using ArcGIS Desktop version 10.8 (Esri, Redlands, CA, USA), shown in Fig. 2.

Fig. 2.

Distribution of digestive diseases in Indonesia (IFLS_5)

Data were analysed using Stata MP 64-bit, version 17.0 (StataCorp LLC, College Station, TX, USA). All analyses were weighted using cross-sectional IFLS-5 survey weights. According to Strauss John et al., [36, 54], the Indonesian Family Life Survey (IFLS-5) uses a stratified multistage sample methodology to provide estimates that are representative of the country's population across provinces and urban–rural strata. In this study sampling weights were used because ignoring the sample selection procedure might provide biased or deceptive findings, especially when random selection probabilities are connected to response variable values even after taking into consideration all available design information. Model misspecification is avoided and consistent estimators of the model parameters are generated using probability weighting of the sample data [55].

The individual-level cross-sectional sample weights (pwt14usxa) from RAND were used to weight all descriptive, bivariate, and multivariable analyses.

Descriptive, bivariate, and multivariate logistic regression models were used to analyse the data in this study. The relationship between outcome variables and covariate factors was assessed using the chi-square test. The final multivariate logistic regression model was presented as P-value, OR, and 95% CI; a P-value of less than 0.05 was considered significant. Multicollinearity was assessed using Variance Inflation Factors (VIF), and model fit was assessed using the Hosmer–Lemeshow goodness-of-fit test. Stratified analyses were provided when appropriate, and interaction terms (e.g., gender × depression, residency × water source) were examined to investigate possible effect modification. Missing data were minimal across all analytical variables. The proportion of missing values did not exceed 5% for any variable. Given the low proportion and the survey design, complete case analysis was performed.

Result

Spatial distribution

Figure 2 presents the spatial distribution of digestive diseases in Indonesia based on the Indonesia Family Life Survey (IFLS-5). The choropleth shading illustrates the number of digestive-related diseases reported in each province, categorized from the lowest range (64–86 cases) to the highest (371–732 cases). The map showed marked geographic variation. The highest reported case counts fall within the range of 371–732, covering provinces such as West Java, Central Java, and East Java, which reflects their dense populations and urban expansion. Moderate levels are observed in the 235–370 range, including Jakarta Raya and Sumatera Utara, as well as in the 172–234 range, which covers Sumatera Barat, Sumatera Selatan, and Yogyakarta. By contrast, lower burdens are concentrated in the 87–171 range (Sulawesi Selatan, Bali, and Lampung) and the 64–86 range (Kalimantan Selatan and Nusa Tenggara Barat).

Sample characteristics

Table 1 shows that 29,817 participants completed the survey, of which 15,189 participants were poor, while 14,628 were non-poor. There were 17,755 participants living in urban areas and 12,062 in rural areas. The majority of the participants were between the ages 25—44 years (38.21%) with females representing 53.59% of the sample. Married people formed about three-quarters of the population (74.13%), while the low education level was common as about 74.66% had less than a high school education.

Table 1.

Baseline characteristics of study population across socioeconomic and residence

Variables	Socioeconomic status		Residence		Total N = 29,817
	Poor n = 15,189	Non poor n = 14,628	Urban n = 17,755	Rural n = 12,062
	% (SE)	% (SE)	% (SE)	% (SE)	% (SE)
Sociodemographic
Age
15–24	6.60 (0.14)	9.66 (0.17)	8.93 (0.15)	7.33 (0.16)	16.27 (0.22)
25–44	19.63 (0.24)	18.58 (0.24)	19.43 (0.22)	18.77 (0.26)	38.21 (0.32)
45–59	16.52 (0.29)	13.95 (0.27)	14.70 (0.24)	30.47 (0.35)	30.47 (0.35)
≥ 60	9.24 (0.23)	5.81 (0.19)	7.02 (0.18)	8.03 (0.23)	15.06 (0.28)
Sex
Male	31.66 (0.32)	14.75 (0.23)	23.39 (0.26)	23.02 (0.31)	46.41 (0.35)
Female	20.33 (0.28)	33.26 (0.33)	26.71 (0.28)	26.89 (0.34)	53.59 (0.35)
Marital status
Married	40.47 (0.34)	33.66 (0.33)	35.68 (0.31)	38.45 (0.35)	74.13 (0.30)
Single	11.52 (0.22)	14.35 (0.23)	14.41 (0.21)	11.45 (0.24)	25.87 (0.30)
Education level
Non or low than high school	38.84 (0.35)	35.81 (0.34)	32.92 (0.31)	41.74 (0.36)	74.66 (0.28)
High school and high	13.15 (0.21)	12.19 (0.20)	17.18 (0.22)	8.16 (0.18)	25.34 (0.28)
Life style
Smoking staus
Non smoker	26.83 (0.31)	36.31 (0.33)	32.58 (0.30)	30.56 (0.34)	63.14 (0.34)
Former smoker	3.39 (0.12)	1.44 (0.08)	2.73 (0.10)	2.10 (0.11)	4.83 (0.15)
Current smoker	21.77 (0.29)	10.26 (0.20)	14.78 (0.22)	17.25 (0.288)	32.03 (0.32)
Physical activity
Low	11.54 (0.21)	11.66 (0.21)	12.77 (0.20)	10.43 (0.22)	23.19 (0.28)
Moderate	19.45 (0.27)	19.06 (0.27)	20.39 (0.25)	18.13 (0.29)	38.51 (0.34)
High	21.00 (0.29)	17.29 (0.27)	16.94 (0.24)	21.35 (0.32)	38.29 (0.34)
Health realted factor
Self-Reported Health
Healthy	40.71 (0.34)	37.20 (0.33)	39.18 (0.32)	38.73 (0.35)	77.91 (0.29)
Unhealthy	11.28 (0.22)	10.81 (0.21)	10.91 (0.20)	11.18 (0.24)	22.09 (0.29)
Comorbidities
No	37.12 (0.33)	34.37 (0.32)	34.53 (0.31)	36.96 (0.35)	71.49 (0.32)
Yes	14.87 (0.26)	13.63 (0.25)	15.56 (0.24)	12.94 (0.27)	28.51 (0.32)
BMI
Normal	30.12 (0.32)	25.29 (0.30)	26.24 (0.28)	29.17 (0.34)	55.41 (0.35)
Underweight	6.27 (0.17)	5.74 (0.15)	5.25 (0.13)	6.76 (0.19)	12.01 (0.23)
Overweight	11.90 (0.22)	12.30 (0.23)	13.52 (0.21)	10.68 (0.24)	24.20 (0.30)
Obese	3.69 (0.13)	4.67 (0.14)	5.07 (0.13)	3.29 (0.14)	8.36 (0.19)
Depressive symptoms
No	41.34 (0.34)	37.36 (0.34)	39.23 (0.32)	39.46 (0.36)	78.70 (0.28)
Yes	10.66 (0.21)	10.65 (0.20)	10.86 (0.19)	10.44 (0.22)	21.30 (0.28)
Food Frequency
Protein consumption per week
< 7	30.40 (0.32)	28.72 (0.31)	27.59 (0.29)	27.07 (0.34)	54.65 (0.35)
≥ 7	21.60 (0.29)	19.28 (0.27)	22.51 (0.26)	22.84 (0.31)	45.35 (0.35)
Vegetables consumption per week
< 7	28.12 (0.32)	26.53 (0.31)	27.59 (0.29)	27.07 (0.34)	54.65 (0.35)
≥ 7	23.88 (0.29)	21.47 (0.28)	22.51 (0.26)	22.84 (0.31)	45.35 (0.35)
Fruits consumption per week
< 7	30.87 (0.32)	27.92 (0.31)	30.07 (0.30)	28.71 (0.34)	58.78 (0.34)
≥ 7	21.13 (0.28)	20.09 (0.28)	20.02 (0.25)	21.20 (0.31)	41.22 (0.34)
Unhealthy Food consumption per week
< 7	29.64 (0.32)	26.77 (0.30)	28.35 (0.29)	28.06 (65.41)	56.41 (0.35)
≥ 7	22.35 (0.30)	21.24 (0.29)	21.84 (0.26)	21.84 (0.32)	43.59 (0.35)
Environment factors
Water source
Improved	45.53 (0.35)	41.69 (0.34)	46.25 (0.34)	40.97 (0.35)	87.22 (0.24)
Unimproved	6.46 (0.18)	6.31 (0.17)	3.84 (0.12)	8.93 (0.22)	12.78 (0.24)
Water treatment
Boiled	49.93 (0.35)	46.11 (0.35)	48.97 (0.35)	47.07 (0.35)	96.04 (0.12)
Un boiled	2.06 (0.09)	1.89 (0.08)	1.11 (0.05)	2.83 (0.11)	3.95 (0.12)
Sanitation facility (sewage)
Improved	12.52 (0.24)	12.04 (0.23)	9.86 (0.19)	14.70 (0.27)	24.56 (0.31)
Unimproved	39.47 (0.34)	35.97 (0.33)	40.23 (0.33)	35.21 (0.35)	75.44 (0.31)
Garbage disposal
Friendly	20.65 (0.27)	19.00 (0.26)	28.69 (0.29)	10.96 (0.23)	39.65 (0.33)
Unfriendly	31.34 (0.33)	29.01 (0.32)	21.40 (0.26)	38.95 (0.35)	60.35 (0.33)
Household toilet
Indoor	45.74 (0.35)	42.19 (0.34)	46.07 (0.34)	41.85 (0.35)	87.93 (0.24)
Outdoor	6.25 (0.18)	5.81 (0.35)	4.02 (0.12)	8.05 (0.21)	12.07 (0.24)

Percentages are weighted using IFLS-5 cross-sectional survey weights (pwt14usxa weights, Complete case analysis was applied; missing data were < 5%

In terms of lifestyle factors, more than half of the participants 63.14% were non-smokers, while 23.19% reported having low physical activity. Regarding the health status, 22.09% reported unhealthy, 28.51% had comorbidity, while 21.30% reported depressive symptoms. Most respondents had normal BMI (55.41%). The dietary assessment shows that less than half of the respondents would consume protein (45.35%), vegetables (45.35%), and fruits (41.22%) at least seven times weekly, while more than half of the respondents would consume unhealthy foods less than seven times a week. As shown, environmental indicators revealed disparities: while access to improved drinking water source was high 87.22% and almost everyone, 96.04%, reported boiled drinking water before consume. However, disparities persisted in sanitation and waste management, as not all population groups particularly the poor and rural residents- had equal access to improve sanitation facilities.

Table 2 presents the crude (unadjusted) associations between all covariates and digestive disease. The overall weighted prevalence of digestive disease was 13.11%. It was higher in the non-poor (14.18%) and urban (14.76%) populations than in the poor (12.11%) and rural (11.44%) populations. In the younger age group (15–24 years) within the age groups, the highest prevalence of the disease was found (14.08%), followed by those aged 25–44 years (14.05%). Women had a significantly higher prevalence which was 15.56% compared with men 10.27% (p < 0.001). In supplementary figures, we also present the weighted prevalence of digestive comorbidities by socioeconomic status and by urban–rural residence (Supplementary Figures S1).

Table 2.

Weighted prevalence of digestive diseases across socioeconomic and residence

Variables	Socioeconomic status		Residence		Total N = 29,817
	Poor n = 15,189	Non poor n = 14,628	Urban n = 17,755	Rural n = 12,062
	% (SE)	% (SE)	% (SE)	% (SE)	% (SE)
Proportion of Digestive diseases	12.11 (0.31)	14.18 (0.34)	14.76 (0.30)	11.44 (0.35)	13.11 (0.23)
Sociodemographic
Age
15–24	14.25 (0.76)	13.97 (0.62)	15.31 (0.62)	12.59 (0.75)	14.08 (0.48)
25–44	12.81 (0.44)	15.36 (0.51)	15.61 (0.43)	12.42 (0.52)	14.05 (0.33)
45–59	12.01 (0.65)	12.45 (0.70)	13.68 (0.63)	10.85 (0.71)	12.21 (0.48)
≥ 60	9.30 (0.76)	14.92 (1.20)	13.98 (0.95)	9.28 (0.91)	11.48 (0.66)
Sex
Male	10.03 (0.37)	10.80 (0.52)	11.15 (0.39)	9.38 (0.47)	10.27 (0.30)
Female	15.36 (0.55)	15.68 (0.44)	17.93 (0.45)	13.21 (0.51)	15.56 (0.34)
Marital status
Married	11.79 (0.35)	14.25 (0.41)	14.54 (0.36)	11.39 (0.40)	12.91 (0.27)
Single	13.25 (0.67)	14.01 (0.61)	15.31 (0.57)	11.62 (0.73)	13.67 (0.45)
Education level
Non or low than high school	10.53 (0.36)	12.33 (0.39)	12.52 (0.36)	10.51 (0.38)	11.39 (0.26)
High school and above	16.79 (0.64)	19.63 (0.69)	19.07 (0.54)	16.23 (0.91)	18.15 (0.47)
Life style
Smoking staus
Non smoker	13.69 (0.45)	15.25 (0.41)	16.52 (0.39)	12.53 (0.46)	14.59 (0.30)
Former smoker	14.65 (1.42)	16.63 (2.21)	15.40 (1.39)	15.03 (2.07)	15.24 (1.19)
Current smoker	9.77 (0.45)	10.07 (0.61)	10.78 (0.49)	9.08 (0.53)	9.86 (0.36)
Physical activity
Low	11.86 (0.65)	13.22 (0.64)	13.34 (0.55)	11.28 (0.54)	12.69 (0.38)
Moderate	12.46 (0.50)	15.29 (0.56)	15.91 (0.49)	11.56 (0.58)	13.86 (0.38)
High	11.93 (0.51)	13.61 (0.58)	14.46 (0.53)	11.57 (0.75)	12.55 (0.45)
Health realted factor
Self-Reported Health
Healthy	10.20 (0.32)	12.33 (0.36)	13.01 (0.32)	9.39 (0.36)	11.21 (0.24)
Unhealthy	19.03 (0.84)	20.56 (0.86)	21.06 (0.77)	18.54 (0.92)	19.78 (0.60)
Comorbidities
No	9.95 (0.33)	12.98 (0.36)	12.56 (0.32)	9.49 (0.36)	10.97 (0.24)
Yes	17.51 (0.72)	19.49 (0.78)	19.65 (0.65)	17.02 (0.87)	18.45 (0.53)
BMI
Normal	11.57 (0.40)	14.10 (0.46)	14.46 (0.41)	11.17 (0.44)	12.73 (0.30)
Underweight	12.51 (0.97)	13.70 (0.94)	15.29 (0.93)	11.35 (0.96)	13.08 (0.68)
Overweight	12.84 (0.67)	14.37 (0.69)	14.98 (0.60)	11.89 (0.78)	13.62 (0.48)
Obese	13.54 (1.27)	14.69 (1.20)	15.20 (1.01)	12.61 (1.58)	14.18 (0.87)
Depressive symptoms
No	10.93 (0.34)	12.94 (0.38)	13.38 (0.33)	10.40 (0.38)	11.89 (0.25)
Yes	16.69 (0.77)	18.52 (0.79)	19.74 (0.72)	15.38 (0.83)	17.61 (0.55)
Food Frequency
Protein consumption per week
< 7	12.22 (0.41)	14.57 (0.44)	15.20 (0.40)	11.44 (0.46)	13.36 (0.30)
≥ 7	14.57 (0.44)	13.60 (0.53)	14.09 (0.46)	11.45 (0.54)	12.74 (0.36)
Vegetables consumption per week
< 7	12.47 (0.44)	13.71 (0.45)	14.66 (0.41)	11.45 (0.48)	13.07 (0.31)
≥ 7	11.70 (0.44)	14.76 (0.52)	14.88 (0.44)	11.44 (0.51)	13.15 (0.34)
Fruits consumption per week
< 7	12.18 (0.41)	15.01 (0.46)	15.10 (0.39)	11.88 (0.48)	13.53 (0.31)
≥ 7	12.02 (0.48)	13.02 (0.50)	14.25 (0.47)	10.86 (0.51)	12.51 (0.35)
Unhealthy Food consumption per week
< 7	11.43 (0.40)	14.50 (0.45)	14.28 (0.39)	11.48 (0.46)	12.89 (0.30)
≥ 7	13.02 (0.50)	13.78 (0.52)	15.40 (0.47)	11.39 (0.54)	13.39 (0.36)
Environment factors
Water source
Improved	12.45 (0.34)	14.77 (0.37)	15.09 (0.32)	11.83 (0.39)	13.56 (0.25)
Unimproved	9.73 (0.83)	10.29 (0.86)	10.76 (0.98)	9.68 (0.75)	10.01 (0.60)
Water treatment
Boiled	12.44 (0.32)	14.45 (0.35)	14.90 (0.31)	11.85 (0.36)	13.40 (0.24)
Un boiled	4.33 (0.87)	7.70 (1.47)	8.93 (1.43)	4.77 (1.03)	5.94 (0.84)
Sanitation facility (sewage)
Improved	11.06 (0.62)	13,70 (0.70)	14.82 (0.70)	10.70 (0.62)	12.36 (0.47)
Unimproved	12.45 (0.36)	14.34 (0.39)	14.75 (0.33)	11.75 (0.42)	13.35 (0.26)
Garbage disposal
Friendly	12.99 (0.47)	15.72 (0.53)	15.57 (0.40)	10.98 (0.73)	14.30 (0.35)
Unfriendly	11.53 (0.42)	13.17 (0.44)	13.68 (0.46)	11.57 (0.40)	12.32 (0.30)

Weighted analyses using IFLS-5 individual cross-sectional survey weights (pwt14usxa), Chi-square significance assessed using survey design, Poor/non-poor based on PCA wealth index; urban/rural classification follows IFLS codes

Participants with higher education had higher prevalence (18.15%) than less education (11.39%). Lifestyle behaviours were also associated with digestive disease prevalence. Non-smoker (14.59%) and former smoker (15.24%) had higher prevalence than current smokers (9.86%, p < 0.001). Among the participants, poor self-reported health (19.87%), co-morbidity (18.45%), and depressive symptoms (17.61%) were strongly associated with a higher prevalence (all p < 0.001). For dietary factors, greater fruit consumption (≥ 7 times/week) showed a combined association with lower prevalence (12.51%) compared to lesser intake (13.53%, p = 0.030). Environmental indicators showed mixed patterns. Participants using unimproved water sources (10.01%) or unboiled drinking water (5.94%) had lower reported prevalence than those with improved sources and boiled water.

Table 3 presents adjusted association from the multivariable logistic regression identifying several predictors of digestive diseases. After adjustment for covariates, socioeconomic and place of residence disparities were noticeable. Participants residing in rural areas had significantly lower odds of reporting a digestive disease compared with urban residents (aOR 0.88, 95% CI: 0.80–0.97). Participants with higher education had significantly higher odds of digestive disease 1.78 times than those with primary education or less (aOR of 1.78; 95% CI: 1.62–1.94). In contrast, the household wealth index was not significantly associated with digestive disease (aOR = 1.02).

Table 3.

Multivariable logistic regression associated with digestive disease

Variables	aOR (95% CI)
Sociodemographic
Age [ref. ≥ 24]	1.00 (0.88–1.14)
25–44	0.79 (0.68–0.91)
45–59	0.71 (0.59–0.84)
≥ 60
Sex [ref. Male]
Female	1.73 (1.52–1.97)
Marital status [ref. Married]
Single	0.98 (0.87–1.10)
Residence [ref. Urban]
Rural	0.88 (0.80–0.97)
Education level [ref. less than high school]
High school and above	1.78 (1.62–1.94)
Socioeconomic status [ref. Poor]
Non-Poor	1.02 (0.94–1.12)
Life style
Smoking staus [ref. Non-smoker]
Former smoker	1.44 (1.16–1.80)
Current smoker	1.01 (0.88–1.16)
Physical activity [ref. High]
Low	0.84 (0.75–0.94)
Moderate	0.96 (0.87–1.06)
Health realted factor
Self-Reported Health [ref. Healthy]
Unhealthy	1.81 (1.64–1.99)
Comorbidities [ref. No]
Yes	1.67 (1.52–1.83)
BMI [ref. Normal]
Underweight	1.01 (0.95–1.25)
Overweight	0.91 (0.82–1.01)
Obese	0.81 (0.69–0.95)
Depressive symptoms [ref. No]
Yes	1.38 (1.26–1.52)
Food Frequency
Protein consumption per week [ref. < 7]
≥ 7	0.92 (0.84–1.00)
Vegetables consumption per week [ref. < 7]
≥ 7	1.01 (0.93–1.10)
Fruits consumption per week [ref. < 7]
≥ 7	0.92 (0.84–1.00)
Unhealthy Food consumption per week [ref. < 7]
≥ 7	1.01 (0.97–1.14)
Environment factors
Water source [ref. Improved]
Unimproved	0.81 (0.70–0.94)
Water treatment [ref. Boiled]
Un boiled	0.43 (0.31–0.58)
Sanitation facility (sewage) [ref. Improved]
Unimproved	1.02 (0.92–1.13)
Garbage disposal [ref. Friendly]
Unfriendly	0.97 (0.89–1.06)

Ref = reference category, no multicollinearity was observed (all VIF < 5), Models are weighted using IFLS-5 individual-level sampling weights

Older age conferred lower odds compared to the youngest age group: those aged 25–44 years had an aOR of 0.79 (95% CI: 0.68–0.91), and those aged 45–59 years had an aOR of 0.71 (95% CI: 0.59–0.84), indicating 29% lower odds. Women had significantly raised odds (aOR 1.73, 95% CI: 1.52–1.97). Former smoking posed a risk (aOR 1.44, 95% CI: 1.16–1.80), while current smoking showed no association. Low physical activity lower odds of reported digestive disease (aOR 0.84, 95% CI: 0.75–0.94), while moderate activity was not significant.

Poor self-rated health (aOR 1.81, 95% CI: 1.64–1.99), presence of comorbidities (aOR 1.67, 95% CI: 1.52–1.83), and depressive symptoms were some of the strongest predictors. The association of obesity with digestive diseases was inverse (aOR 0.81, 95% CI: 0.69–0.95). Higher intake of fruits and proteins showed a borderline protective association (around 0.92); while vegetable intake and intake of unhealthy foods showed no significant association. Environmental factors revealed paradoxical findings, with participants using unimproved water sources (aOR 0.81, 95% CI: 0.70–0.94) or untreated drinking water (aOR 0.43, 95% CI: 0.31–0.58), reporting lower odds of digestive disease.

Table 4 presents interaction analysis examining whether the association between key factors and digestive disease differed by sex, place of residence and comorbidity status. No statistically significant interaction was observed between depressive symptoms and sex (interaction aOR 1.07, 95% CI: 0.91–1.25), indicating the association between depressive symptoms and digestive disease did not differ meaningfully between male and female. The analysis demonstrates that depressive symptoms were associated with higher odds of digestive diseases in males (aOR 1.29, 95% CI: 1.13–1.46), and females (aOR 1.38, 95% CI: 1.25–1.52).

Table 4.

Interaction analysis of sex, residence, comorbidity, and depressive symptoms on digestive diseases

Variables	aOR (95% CI)	p-value
Male: Depression vs No	1.29 (1.13–1.46)	< 0.001
Female: Depression vs No	1.38 (1.25–1.52)	< 0.001
Interaction (extra effect in women)	1.07 (0.91–1.25)	0.400
Urban: Unimproved vs Improved	0.80 (0.66–0.96)	0.016
Rural: Unimproved vs Improved	0.89 (0.77–1.04)	0.150
Interaction (Rural vs Urban difference)	1.12 (0.88–1.42)	0.350
No comorbidity: Depressive symptoms vs Normal	1.14 (1.08–1.55)	< 0.001
With comorbidity: Depressive symptoms vs Normal	1.23 (1.08–1.41)	0.002
Interaction term (extra effect of depression in comorbidity group)	0.88 (0.75–1.03)	0.116

Models are weighted using IFLS-5 individual-level sampling weights, Ref = reference category

Similarly, there was no difference interaction was found between water sources and place of residence (interaction aOR = 1.12; 95% CI: 0.88–1.42). The chances of digestive diseases disorders were 20% lower among urban individuals who had not used unimproved water (aOR 0.80, 95% CI: 0.66–0.96), but this link was smaller and not statistically significant among rural participants (aOR 0.89, 95% CI: 0.77–1.04). The interaction term showed no meaningful difference between urban and rural groups (interaction aOR 1.12, 95% CI: 0.88–1.42), indicating no meaningful difference between urban and rural groups. Finally, among those with co-morbidities (aOR 1.23, 95% CI: 1.08–1.41), and those without either, depressed symptoms are linked to 41% higher odds of digestive diseases disorders (aOR 1.14, 95% CI: 1.08–1.55). Nonetheless, the elevated the odds of depressive symptoms was linked to individuals in the same manner.

Figure 3 (a) indicates the expected probabilities of digestive diseases stratified by sex and depressive symptoms. It is shown from the data that those with depressive symptoms are more likely to suffer from digestive diseases than those who do not show any such symptoms for both males and females. Although the expected probability was slightly higher for females with depression, yet the interaction term did not turn out to be important for statistical testing, meaning thereby that the association between depression and digestive diseases was uniform across both sexes.

Fig. 3.

Interaction effects of sex, comorbidity, and environmental factors with depressive symptoms on the probability of reporting digestive diseases

The predicted probability of digestive diseases by comorbidity status, as well as by depressive symptoms, is shown in Fig. 3 (b). Individuals with the highest predicted probability of digestive disease were those with both comorbid depression and features of comorbidity, while the next highest predicted probability individual had a comorbidity alone. Depressive features added to the risk in both groups, but the interaction term was not statistically significant. Hence comorbidity did not amplify the effect of depression on the likelihood of having digestive diseases.

Figure 3 (c) shows the predicted probability of reporting digestive diseases based on place of residence (urban versus rural) and association between source of water and digestive diseases (improved versus unimproved). Though unimproved water sources were still associated with lower predicted probabilities than improved water sources. There was only minor difference between urban and rural. The interaction term between residence and water source was not significant, indicating that the effect of water source on digestive diseases was not appreciably different between urban and rural populations.

Discussion

Our findings align with broader global evidence showing that digestive disease is patterned by sociodemographic and lifestyle factors. Using nationally representative data IFLS-5 data, this study demonstrates significant educational and residence disparities in doctor-diagnosed digestive diseases in Indonesia. Even after adjusting for other factors, residence and education had higher odds reporting digestive diseases. These gradients likely reflect a combination of lifestyle differences, health awareness, and unequal access to diagnostic services.

Although nationally representative data on digestive disease prevalence are limited in Southeast Asia, findings from community-based studies in comparable countries show similar patterns of common gastrointestinal symptoms. For example, Malaysian-based research among national populations has identified prevalence rates for dyspepsia within 11%−24% respectively, and significantly more so among urban dwellers and among middle-aged groups within that society [56]. Also, within Thailand, community-based research has identified prevalence rates among 10%−20% for so-called ‘functional gastrointestinal symptoms’ including abdominal pain, bloating, and indigestion among adult populations, with significantly more women and more city dwellers affected within Thailand [57]. Clearly enough, then, prevalence rates among our Indonesian research sample at 13.11% are consistent with regional rates within Southeast Asia.

Educational and rural–urban differences in digestive diseases in Indonesia are demonstrated nationally by this study. Even after controlling for behavioural and demographic variables, people in urban areas were more likely than those in rural areas to report digestive disease diagnosed by medical staff. Adults with higher levels of education had significantly higher odds of digestive disease, following a similar pattern across educational levels. These findings suggest that disparities in digestive health outcomes among population groups may be associated with variation in lifestyle, awareness, and access to healthcare services. The findings are in line with other studies that show a rise in metabolic and gastrointestinal disorders in Indonesian cities, such as Jakarta, Surabaya, and Yogyakarta due to urbanization and shifting lifestyles [58].

A number of factors that were considered significant in the crude models were no longer significant after adjustment, which most probably indicates the presence of confounding and the differences in healthcare access. In the crude results, the association of wealth, marital status, and some behavioral measures with education, age, or place of residence key social determinants that dramatically affect health literacy and care-seeking may have been the reason they were considered as associated with the outcome. When these underlying determinants were included in the multivariable model, the independent effect of these variables decreased. Access to diagnosis might also play a role: urban, educated, or more health system engaged persons are the ones most likely to receive formal digestive disease diagnosis, while those from rural or less-resourced areas may underreport due to lack of access to diagnostics. Health-seeking behavior further complicates these patterns, as higher-education or urban residents are more likely to identify symptoms earlier, seek professional advice more often, and participate in diagnostic tests. In conclusion, the attenuation of crude associations in the adjusted models probably points to a mixture of confounding factors, unequal access to diagnosis, and differences in symptom recognition and care-seeking behavior among the different population groups as the main causes. The crude association for wealth, in particular, attenuated after adjustment because wealth is highly associated with education and living in an urban area, both of which have a more direct impact on healthcare access, health literacy, and the probability of being diagnosed with a digestive disease. Consequently, once these confounders were taken into consideration, wealth was no longer an independent predictor.

Education may reflect health literacy and healthcare access, which impact disease diagnosis and treatment. Although household wealth was not significantly associated with digestive disease in the adjusted models, education remained a consistent predictor. In line with the WHO Social Determinants of Health framework [19], education is a core dimension of socioeconomic status, capturing health literacy, behavioural resources, and awareness pathways that are different from material assets as measured by the wealth index. This pattern is consistent with IFLS-based analyses showing that education often captures socioeconomic gradients more strongly than wealth [59]. Residence and education, therefore, reflect social and geographic inequalities relevant to digestive disease in Indonesia. These findings suggest that socioeconomic differences in digestive disease work primarily through educational rather than financial pathways.

Women consistently report more digestive diseases than men, aligning with meta-analytic evidence that functional dyspepsia and Irritable Bowel Syndrome (IBS) are more prevalent in women and impose greater symptom burden [60, 61]. Urban residence and higher schooling in our sample were linked to more diagnosed digestive problems, echoing IFLS-5 results showing higher multimorbidity in urban and better-educated groups and likely reflecting greater health-care access and diagnosis in cities [62, 63].

Urban dwellers may be more likely to experience gastrointestinal issues due to their increased exposure to processed food, stress, and more sedentary lifestyle patterns that have been well documented in Indonesia’s rapidly urbanizing communities [64, 65]. Due to limited access to healthcare, underdiagnosis may occur among rural inhabitants. Evidence supporting this pattern comes from studies showing that hospital utilisation rates among urban dwellers in Indonesia are significantly higher among urban residents comapred to rural resient in indonesia, because urban area have better access to refeerral service and outpatient care [63]. In addition, individuals living in the rural areas often need to travel longer distances to get specialist treatment, which in turn exacerbates the geographical imbalance in availability and provides less access to health services [66]. As a result, variations may represent both actual variations in the burden of disease and discrepancies in diagnosis. The results highlight the necessity of context-specific public health interventions that go beyond clinical risk factors to address geographical and socioeconomic obstacles to gastrointestinal care.

Since the measure of health outcome is entirely dependent on self-reported physician diagnoses, it is very likely that there will be a case of misclassification due to the differences in the diagnoses. People living in rural areas or belonging to lower-income groups are most probably getting less healthcare and are being less diagnosed and less reported with digestive diseases than people living in the urban area who are better educated and therefore more likely to seek care, recognise symptoms, and obtain formal diagnoses. These differences in access to healthcare may lead to biased associations where the disparities in the groups with good healthcare access are potentially exaggerating gradients in groups with better diagnostic access while masking the true burden in underserved communities. This bias related to diagnostic access is a significant limitation of IFLS and should be taken into account when drawing conclusions about the prevalence of digestive diseases.

By contrast, household wealth was not a strong predictor in our data, and the literature is mixed—some studies find little or no clear socioeconomic status gradient in Irritable Bowel Syndrome and related Disorders of gut-brain interaction (DGBIs), while others report heterogeneous patterns across settings [67, 68]. This adult pattern differs from childhood digestive illness in Indonesia, where poverty strongly elevates diarrheal risk, underscoring that infectious drivers dominate in children while adult digestive morbidity increasingly reflects non-infectious conditions [69]. Lifestyle-related exposures that cluster with urbanization—such as higher intake of ultra-processed or take away foods and rising obesity are plausible contributors to adult gastrointestinal burden and are repeatedly associated with DGBIs in contemporary cohorts [70, 71]. Taken together, the gender, urban, and education gradients we observe likely reflect both true lifestyle-linked risk and higher detection from care-seeking, while the weak wealth gradient suggests that adult digestive diseases now cut across income strata in Indonesia [62, 63].

Behavioural factors emerged as strong predictors of digestive disease in our analysis. Notably, smoking displayed a clear association with dyspeptic symptoms. In a 2024 cross-sectional study among Indonesian adults in Malang, 67% of individuals with dyspepsia were active smokers, pointing to a substantial behavioural contributor to digestive distress [27]. International evidence further supports this link: a study of nearly 2,000 Iranian men found that smokers had an 83% higher risk of clinically defined functional dyspepsia (FD), with significantly elevated 1.57 times for postprandial fullness and 1.92 for epigastric pain [72]. A global review also identified smoking as a robust modifiable risk factor for FD, possibly due to changes in the duodenal microbiome and mucosal integrity [73].

Dietary habits specifically irregular meal timing and intake of trigger foods play an important role in digestive symptoms. In an Indonesian study, individuals eating at intervals over six hours or consuming high-risk “trigger” foods showed significantly increased dyspepsia risk [27]. Similar patterns have been observed globally. a Chinese cohort showed that skipping meals and dining out were positively associated with dyspepsia, while dietary advice often includes smaller, frequent meals and reducing spicy, fatty, or acid-stimulating foods [74]. A longitudinal Japanese study elucidated that skipping breakfast or lunch was independently associated with higher prevalence of functional dyspepsia, reinforcing the importance of regular eating habits [75]. At the same time, healthier dietary elements appeared protective. An Indonesian cohort study revealed that high consumption of fruits and vegetables corresponded with lower dyspepsia risk [27] a finding that aligns with broader nutritional research suggesting antioxidant-rich, fiber-dense diets mitigate digestive symptoms. Together, these findings highlight the consistent influence of meal regularity and dietary patterns across different populations.

Depressive symptoms also showed strong and independent associated with digestive disease in our analysis. A large body of evidence revealed that psychological distress and mental health conditions can influence gastrointestinal function through the bidirectional interactions between anxiety, depression, and functional digestive disease mediated through the gut–brain axis, leading to altered motility, visceral hypersensitivity, changes in gut microbiota, and increased inflammatory responses [73]. According to a meta-analysis, the frequency of intestinal and extra-intestinal symptoms in IBS patients is closely correlated with the existence and intensity of depression [76]. These mechanisms help explain why depressive symptoms remained a significant predictor in our adjusted models, independent of socioeconomic, behavioural, and physical health factors. The cross-sectional nature of the data makes it impossible to establish temporal relationships, so the observed associations should be interpreted cautiously.

Strengths and limitations

This analysis draws on a very large, nationally representative Indonesian survey (the IFLS), which covers roughly 83% of the population. Such a sample provides high statistical power and generalizability. Moreover, the IFLS collects rich individual-, household- and community-level data, including socioeconomic status, lifestyle behaviours, environmental infrastructure, and health measures, allowing a broad, multilevel assessment of factors associated with digestive diseases. Although the data from 2014–2015, their use offers two important, long-lasting values that go beyond a straightforward prevalence snapshot. Firstly, it creates an important longitudinal baseline; our results are used as a standard for subsequent research that uses IFLS-6 to monitor patterns and prove causation over time. Second, and perhaps more significantly, our main contribution is to clarify the basic structural patterns and relationships of disparity (e.g., education gradients, rural–urban relationship) and their correlation with important risk factors (e.g., smoking, depression). Since these underlying biological and societal relationships are extremely stable and unlikely to have undergone significant change in the past ten years, our examination of these mechanisms is extremely pertinent for gaining fundamental policy insights.

On the other hand, several limitations should be noted. First, the study is cross-sectional, meaning exposures and outcomes were measured at one time point; therefore, causality can’t interfere. Second, the outcome measurement relied on self-reported doctor diagnoses, which may underestimate the true prevalence of digestive disease, particularly among rural and lower-SES groups who have limited access to diagnostic services. This underdiagnosis may have introduced differential misclassification. Third, using data from 2014–2015 may not take into consideration more recent changes in healthcare access and health-related behaviour. Moreover, the digestive disease outcome combined several gastrointestinal conditions into a single category (“stomach or other digestive disease”), meaning the findings may mask heterogeneity across specific disorders. In addition, some potentially significant variables, such as alcohol use, stress, or detailed dietary intake, were measured only crudely or were unavailable, which might have caused residual confounding. Finally, symptom-level data were limited; hence, the study relied on self-reported physician diagnosis as a proxy measure rather than clinical or laboratory confirmation. Future studies with clinical, longitudinal or symptom data are needed to validate these findings.

Policy implications and conclusions

These findings suggest that strategies for digestive disease prevention and management should adopt an integrated and comprehensive approach. The analysis showed consistent educational and rural–urban differences, whereas wealth was not an independent predictor after adjustment; therefore, policy efforts should prioritize addressing inequalities related to education, health literacy, and place of residence rather than income alone. Routine screening and clinical care for digestive diseases should also include mental health assessment and support, since depressive symptoms frequently co-occur with digestive disease.

Programs should explicitly address geographic and social disparities: rural areas may continue to experience a greater burden of infectious digestive diseases, while urban populations are increasingly affected by lifestyle-related conditions. Health promotion programs should emphasize modifiable behaviours, such as improving diet quality, encouraging smoking cessation, and promoting physical activity, as these are key drivers of chronic disease. In addition, improving water, sanitation, and hygiene (WASH) infrastructure is essential; poor sanitation and unsafe water are well-known risk factors for gastrointestinal infection.

Authors’ contributions

All authors contributed substantially to the conception, design, analysis, and interpretation of data. N.A and A.A led the data analysis and manuscript drafting. N.A.D.M.H contributed to GIS mapping, interpretation of findings. NA & AA and N.A.D.M.H manuscript revision, and final approval of the version to be submitted. All authors read and approved the final manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

This study used IFLS-5 data which are publicly available, conducted by RAND. Researchers can access the data upon registration at the RAND Corporation website: [https://www.rand.org/well-being/social-and-behavioral-policy/data/FLS/IFLS.html].

Declarations

Ethics approval and consent to participate

This study is based on secondary analysis of data from the Indonesia Family Life Survey Wave 5 (IFLS-5). This data collection protocol was reviewed and approved by the Institutional Review Boards (IRBs) of the RAND Corporation (Santa Monica, USA) and Universitas Gadjah Mada (Yogyakarta, Indonesia) with approval number: s0064-06–01-CR01. Written informed consent was obtained from all IFLS-5 participants by the original survey team. The IFLS-5 dataset is publicly available for research use through the RAND Corporation website https://www.rand.org/labor/FLS/IFLS.html following user registration and data-use agreement. Additional ethical clearance for the secondary use of this dataset was obtained from the Health Research Ethics Committee of Dr. Moewardi General Hospital, Indonesia (No. 790/IV/HREC/2025).

Consent for publication

The authors consent to the submission of this manuscript, including all text, figures, tables, and supplementary materials, to BMC Public Health. All authors have read and approved the final version of the manuscript and agree to its publication.

Competing interests

The authors declare no competing interests.