Socioeconomic inequalities in healthcare access among patients with type 2 diabetes in Iran: a cross-sectional study

Abstract

Objective

Socioeconomic inequalities significantly impact access to healthcare services for individuals with type 2 diabetes mellitus (T2DM). This study investigates these inequalities in Iran, focusing on factors such as asset, sex, urban–rural residence, age, education, employment status, and marital status.

Design

Cross-sectional study.

Setting

This study used data from the national ‘Diabetes Care (DiaCare)’ study, a population-based survey conducted from 2018 to 2020 in Iran.

Methods

Socioeconomic status (SES) was assessed using Principal Component Analysis (PCA) based on assets. Socioeconomic inequalities in access to physicians, pharmacies and laboratories were measured using the Concentration Index (CI) and Erreygers Corrected Concentration Index (ECI). Decomposition analysis was performed using a probit regression model to assess the contributions of various factors to the observed inequalities.

Results

Among 13 315 patients with T2DM, 5.8% lacked access to physicians, 6.8% to pharmacies and 8.7% to laboratories. The CI was positive and statistically significant for access to physicians (0.0614), pharmacies (0.0787) and laboratories (0.0875), indicating better access concentrated among higher SES individuals. Urban residents had the largest positive marginal effects on access to physicians (0.032), pharmacies (0.078) and laboratories (0.053), with percentage contributions of 13.21%, 23.23% and 17.39%, respectively. Higher asset quintiles showed substantial contributions to inequalities, with the highest quintile contributing 10.5% to physician access inequality, 9.68% to pharmacy access and 9.16% to laboratory access. Education level also positively impacted access, with high school education contributing 0.64% and college education 0.52% to access inequalities. Sex differences showed a negative marginal effect for women, indicating slightly lower access.

Conclusion

Socioeconomic factors, particularly asset, residence and education, significantly impact access to healthcare services for patients with T2DM in Iran. Policies should focus on reducing barriers to healthcare access, especially for lower SES and rural populations.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study benefits from a large, nationally representative sample, allowing robust assessment of socioeconomic inequalities in healthcare access.

• The absence of direct income data required the use of an asset-based socioeconomic index, which may have resulted in some residual misclassification.

• Certain health-related and psychosocial variables were not included, which may have influenced the findings.

Introduction

Type 2 diabetes mellitus (T2DM) is one of the most common and leading causes of death worldwide.1,3 T2DM occurs when the body’s cells become resistant to insulin, hindering glucose uptake. It accounts for approximately 90% of all diabetes cases.⁴ The prevalence of T2DM is rising, making it a growing global epidemic. The International Diabetes Federation (IDF) estimates that the number of individuals with diabetes worldwide will increase from 463 million in 2019 to 700 million by 2045.^{2 5 6} This increase in prevalence is attributed not only to a higher incidence of T2DM but also to a decline in mortality rates among individuals with diabetes.⁷ Additionally, diabetes-related healthcare costs are projected to increase from US$ 966 billion in 2021 to $1028 billion by 2030.⁸

Diabetes is associated with a wide range of chronic diseases. Over time, it can damage the eyes, kidneys and nerves, leading to a higher risk of amputation, vision loss and early death.^{5 9} It is also a significant risk factor for coronary heart disease, cardiovascular disease, and cerebrovascular disease, and hypertension is more prevalent among individuals with diabetes.^{4 10} These complications exert growing pressure on healthcare systems. Generally, the increasing prevalence of diabetes is linked to a continuous rise in its complications, contributing to greater disability and escalating healthcare costs, which account for between 2.5% and 15% of the annual healthcare budget.¹ Factors such as unhealthy diets, smoking, alcohol consumption, physical inactivity, genetic predisposition, population growth, ageing, urbanisation, overweight or obesity, and other demographic variables contribute to the rising prevalence of T2DM.^{8 11 12}

Socioeconomic factors significantly impact the prevalence and management of diabetes.¹³ Socioeconomic status (SES) can affect access to and quality of care, social support and community resources. It also influences knowledge about diabetes, communication with healthcare providers, early intervention, medication adherence, exercise, and diet.^{1 2 14} SES encompasses factors such as access to healthcare and information, availability of healthy food and exercise opportunities, income levels, education, job opportunities and individual lifestyle choices.⁸ SES inequalities manifest as an unequal distribution of access to resources and opportunities, hindering the ability to achieve a healthy lifestyle.¹⁴

Studies indicate that individuals with diabetes who have lower levels of education and income tend to have poorer glycaemic control compared to those with higher education and income levels.⁹ It is evident that individuals with diabetes and low SES are less likely to access information about their condition, receive adequate diabetes care or engage in preventive measures.¹⁵

Understanding SES inequalities is essential to achieving equity in healthcare.⁸ Given the substantial burden T2DM places on healthcare systems, it is vital to identify and address these inequalities to reduce the inequitable distribution of healthcare services. Therefore, this study aims to examine how SES inequalities contribute to access to healthcare services among individuals with T2DM.

Methods

Study design

This study used a cross-sectional design based on data from an observational, population-based study titled ‘Diabetes Care (DiaCare)’.¹⁶ DiaCare was a national survey conducted to evaluate the status of care among patients with type 2 diabetes across all provinces of Iran from 2018 to 2020. The study employed systematic random, stratified and cluster sampling techniques to select participants. The sample (n=13 315) consisted of individuals aged 35–75 years living in both rural and urban areas, with a consistent sample size across provinces. The allocation between urban and rural areas was proportional to their population distribution within each province.¹⁶ The inclusion criterion for this national study was a diagnosis of T2DM in accordance with the recommendations of the American Diabetes Association (ADA).¹⁷ Data were collected through structured interviews using questionnaires. The variables collected included SES, demographic information, and access to healthcare services such as physicians, pharmacies, and laboratories. Additional details on the DiaCare study are available in the published study protocol.¹⁶

This study is observational and descriptive in nature; therefore, the findings reflect socioeconomic inequalities and associations in healthcare access rather than causal effects.

Patient and public involvement

Patients and the public were not involved in the design, conduct or dissemination of our research.

Statistical analyses

The exposure

SES was assessed based on household assets, including home ownership status, house area, number of rooms, and possession of items such as TV, LED/LCD TV, landline phone, mobile phone, laptop, internet access, refrigerator, washing machine, dishwasher, microwave and car. The SES index (asset index) was calculated using Principal Component Analysis (PCA) based on these asset variables. This approach assigns data-driven weights to each asset based on their variance and covariance structure, rather than relying on simple asset ownership or frequency. As a result, assets that better discriminate between different socioeconomic positions receive higher weights in the composite index. The use of a multidimensional asset index allows for capturing long-term economic status and reduces bias associated with short-term income fluctuations or reporting errors, particularly in large population-based surveys. In the Iranian context, where reliable individual income data are often unavailable in national surveys, asset-based SES measures have been widely used as a valid proxy for socioeconomic positioning.

The resulting asset index was then divided into five quintiles, with the first quintile representing the poorest and the fifth quintile representing the wealthiest.

The outcome

Access to healthcare services (physicians, pharmacies and laboratories) was measured using binary (yes/no) questions based on physical and geographical accessibility. Participants were asked whether they had access to each service in terms of distance.

To assess SES inequalities in access to healthcare services (physicians, pharmacies, and laboratories), the Concentration Index (CI) was employed. The concentration curve plots the cumulative percentage of the outcome variable (eg, access to a physician) against the cumulative percentage of the population, ranked by SES.¹⁸ If no SES inequalities exist, the curve lies along the equity diagonal line (45-degree line).¹⁹ A concentration curve below the diagonal indicates that wealthier individuals have better access to healthcare services, while a curve above the diagonal suggests that access is concentrated among the poorer segments of the population.

The CI is defined as twice the area between the concentration curve and the line of equality (the 45-degree line)¹⁸ and is calculated using the following formula:

$$C=\frac{2}{\mu }*cov(h,r)$$ (Equation 1)

where ℎ represents the value of the dependent variable related to healthcare (eg, access to a physician, laboratory, or pharmacy), μ is its mean and r is the fractional rank of an individual according to their SES, from the poorest to the richest.¹⁸ The values of CI range from −1 to +1, with the sign indicating the direction of the relationship between SES and access to physicians, pharmacies, and laboratories.²⁰ Positive values occur when the concentration curve lies below the line of equality, indicating greater accessibility for individuals with higher SES.18,21

Given the dichotomous nature of the outcome variables, the Erreygers Corrected Concentration Index (ECI) was used. It is defined as²²

$$ECI=4*\mu *C$$ (Equation 2)

Where μ is the mean of the health variable (the proportion of the population with access to physicians, pharmacies, and laboratories).

The decomposition method proposed by Wagstaff²³ was applied to evaluate the contributions of SES and other covariates to inequalities in access to healthcare services. Due to the dichotomous nature of the outcome variables, a probit regression model was used to decompose the CI. The probit decomposition regression model can be expressed as follows:

$$y_i=\delta +{\sum }_k{\gamma }_k{z}_k+{\epsilon }_i$$ (Equation 3)

Where $y_i$ indicates whether a participant has access or not, $z_k$ represents covariates, $\delta$, ${\gamma }_k$ and ${\epsilon }_i$ represent the constant term, marginal effect and disturbance terms, respectively.²⁴ The decomposition formula is:

$$C={\sum }_k\frac{{\gamma }_k\stackrel{-}{z_k}}{\mu }{C}_k+\frac{{GC}_{\epsilon }}{\mu }$$ (Equation 4)

where $\stackrel{-}{z_k}$ denotes the mean of each covariate, $C_k$ is the CI for each covariate, $\frac{{\lambda }_k\stackrel{-}{z_k}}{\mu }$ represents the product of the marginal effect and the mean of each covariate, as used in non-linear decomposition models and $\frac{{GC}_{\epsilon }}{\mu }$ is the error term.

To compare access to healthcare services across different groups, Pearson’s χ² test was used. Statistical significance was evaluated at p<0.05.

Missing data were minimal across study variables, and analyses were performed using complete-case analysis.

All analyses were performed using Stata software V.17 (College Station, Texas, USA: StataCorp LP).

Results

The study included a total of 13 315 patients with T2DM, of whom 50.2% were women. The frequency distribution of patients according to access to various healthcare services (physician, pharmacy, and laboratory) is presented in table 1. Overall, 5.8%, 6.8%, and 8.7% of patients did not have access to physicians, pharmacies, and laboratories, respectively. Among those without access to physicians, 57.8% were women, 46.5% lived in rural areas and only 2.4% had a college degree. For those lacking access to pharmacies, 52.4% resided in rural areas, 38.2% were illiterate and 37.3% belonged to the lowest SES quintile. Of the patients without access to laboratories, 56.1% were women, 34.5% were in the lowest SES quintile and 31.0% were employed.

Table 1. Baseline characteristics of the study population.

	Physician		Pharmacy		Laboratory
Characteristic	No access, n=780	Has access, n=12 535	No access, n=900	Has access, n=12 415	No access, n=1164	Has access, n=12 151
Sex
Women	451 (6.8%)	6229 (93%) 6306 (95%)	507 (7.6%)	6173 (92%) 6242 (94%)	654 (9.8%)	6026 (90%) 6125 (92%)
Men	329 (5.0%)		393 (5.9%)		510 (7.7%)
Age
35–45	113 (5.8%)	1848 (94%) 4491 (94%) 3880 (94%) 2255 (94%)	116 (5.9%)	1845 (94%) 4452 (93%) 3852 (93%) 2204 (92%)	160 (8.2%)	1801 (92%) 4347 (91%) 3772 (91%) 2170 (91%)
45–55	276 (5.8%)		315 (6.6%)		420 (8.8%)
55–65	248 (6.0%)		276 (6.7%)		356 (8.6%)
65–75	140 (5.8%)		191 (8.0%)		225 (9.4%)
Asset Index
1 (poorest)	276 (10%)	2393 (90%) 2478 (93%) 2518 (95%) 2539 (95%) 2605 (98%)	336 (13%)	2333 (87%) 2448 (92%) 2489 (93%) 2543 (95%) 2600 (98%)	401 (15%)	2268 (85%) 2388 (90%) 2436 (91%) 2486 (93%) 2571 (97%)
2	177 (6.7%)		207 (7.8%)		267 (10%)
3	145 (5.4%)		174 (6.5%)		227 (8.5%)
4	126 (4.7%)		122 (4.6%)		179 (6.7%)
5 (richest)	56 (2.1%)		61 (2.3%)		90 (3.4%)
Area
Rural	363 (9.6%)	3431 (90%) 9104 (96%)	472 (12%)	3322 (88%) 9093 (96%)	614 (16%)	3180 (84%) 8971 (94%)
Urban	417 (4.4%)		428 (4.5%)		550 (5.8%)
Education
No education	282 (8.3%)	3118 (92%) 8179 (95%) 1238 (98%)	344 (10%)	3056 (90%) 8118 (94%) 1241 (98%)	422 (12%)	2978 (88%) 7945 (92%) 1228 (97%)
High school or less	468 (5.4%)		529 (6.1%)		702 (8.1%)
College	30 (2.4%)		27 (2.1%)		40 (3.2%)
Employment
Employed	233 (5.4%)	4062 (95%) 1985 (97%) 6320 (93%)	271 (6.3%)	4024 (94%) 1970 (96%)	362 (8.4%)	3933 (92%) 1948 (95%)
Retired	64 (3.1%)		79 (3.9%)		101 (4.9%)
Other (unemployed/homemaker)	471 (6.9%)		542 (8.0%)	6249 (92%)	687 (10%)	6104 (90%)
Marital status
Single	114 (8.1%)	1298 (92%) 11 233 (94%)	125 (8.9%)	1287 (91%) 11 124 (93%)	157 (11%)	1255 (89%) 10 892 (92%)
Married	666 (5.6%)		775 (6.5%)		1007 (8.5%)

Table 2 displays the CI for different access variables (physician, pharmacy, and laboratory). The C values for all variables were positive and statistically significant: 0.0614 for physician access, 0.0787 for pharmacy access, and 0.0875 for laboratory access. This indicates that access to all these services is concentrated among individuals with higher SES.

Table 2. Concentration Index for different access variables.

Variable	Concentration Index	SE	P value
Physician	0.0614	0.004	<0.001
Pharmacy	0.0787	0.004	<0.001
Laboratory	0.0875	0.006	<0.001

Figure 1 illustrates the concentration curves for different access variables. All curves lie below the equity line, suggesting a higher concentration of healthcare access among individuals with higher SES.

Figure 1. Lorenz curves (concentration curves) for (A) physician access, (B) laboratory access and (C) pharmacy access.

Tables35 present the results of the decomposition analysis for access to physicians, pharmacies, and laboratories, identifying the contributions of different SES factors to overall inequality. The findings indicate that asset is a significant determinant of access to healthcare services. Individuals in the highest asset quintile (group 5) have the largest positive marginal effects for access to physicians (0.047), pharmacies (0.060), and laboratories (0.052). These contribute to the inequality in access to physicians (10.5%), pharmacies (9.68%) and laboratories (9.16%). Conversely, lower asset quintiles (groups 2 and 3) have negative percentage contributions, suggesting that reduced asset correlates with reduced inequalities.

Table 3. Result of the decomposition analysis for physician access.

Variable	Marginal effect	CI	Contribution	Decomposition component	%
Asset Index
1
2	0.0182259	−0.31878	−0.00123	0.003862	−2.00447
3	0.0245973	−0.00012	−6.3E−07	0.005226	−0.00102
4	0.0257862	0.320337	0.001756	0.005481	2.858425
5	0.0474709	0.639712	0.006447	0.010078	10.49675
Sex
Men
Women	−0.0098607	−0.15844	0.000833	−0.00525	1.355513
Area
Rural
Urban	0.0321497	0.332236	0.008113	0.024419	13.20865
Age
35–45
45–55	−0.0006332	0.129985	−3.1E−05	−0.00024	−0.05121
55–65	0.0020765	−0.04571	−3.1E−05	0.000687	−0.05116
65+	0.0089173	−0.12154	−0.00021	0.001711	−0.33863
Education
No education
High school or less	0.0034023	0.167217	0.000393	0.002347	0.639052
College and more	0.0146033	0.215886	0.000319	0.001477	0.518999
Employment status
Employed
Retired	0.008169	0.178731	0.000242	0.001354	0.393915
Other (unemployed/homemaker)	0.0027634	−0.26756	−0.00041	0.001518	−0.66112
Marital status
Single
Married	−0.0092031	−0.12252	0.000127	−0.00104	0.206765

CI, Concentration Index.

Table 5. Result of the decomposition analysis for laboratory access.

Variable	Marginal effect	CI	Contribution	Decomposition component	%
Asset Index
1	Ref
2	0.020708	−0.31878	−0.00141	0.004431	−1.79274
3	0.025816	−0.00012	−6.7E−07	0.005538	−0.00084
4	0.036669	0.320337	0.002521	0.007869	3.199601
5	0.052631	0.639712	0.007217	0.011282	9.160791
Sex
Men	Ref
Women	−0.00283	−0.15844	0.000241	−0.00152	0.306049
Area
Rural	Ref
Urban	0.053773	0.332236	0.0137	0.041237	17.3902
Age
35–45	Ref
45–55	−0.00769	0.129985	−0.00039	−0.00297	−0.48939
55–65	−0.0016	−0.04571	2.44E−05	−0.00053	0.030989
65+	−0.00663	−0.12154	0.000156	−0.00129	0.198323
Education
No education	Ref
High school or less	0.003886	0.167217	0.000453	0.002707	0.574504
College and more	0.025192	0.215886	0.000555	0.002572	0.704764
Employment status
Employed	Ref
Retired	0.003917	0.178731	0.000117	0.000655	0.148674
Other (unemployed/homemaker)	−0.0028	−0.26756	0.000415	−0.00155	0.527363
Marital status
Single	Ref
Married	−0.00235	−0.12252	3.27E−05	−0.00027	0.041542

CI, Concentration Index.

Table 4. Result of the decomposition analysis for pharmacy access.

Variable	Marginal effect	CI	Contribution	Decomposition component	%
Asset Index
1	Ref
2	0.021411	−0.31878	−0.00149	0.00468	−1.70428
3	0.028631	−0.00012	−7.5E−07	0.006275	−0.00086
4	0.036379	0.320337	0.002555	0.007976	2.918672
5	0.060546	0.639712	0.008483	0.01326	9.689711
Sex
Men	Ref
Women	−0.01419	−0.15844	0.001236	−0.0078	1.412102
Area
Rural	Ref
Urban	0.078142	0.332236	0.020342	0.061227	23.23589
Age
35–45	Ref
45–55	−0.00906	0.129985	−0.00046	−0.00357	−0.53037
55–65	−0.00071	−0.04571	1.11E−05	−0.00024	0.012618
65+	−0.0021	−0.12154	5.05E−05	−0.00042	0.057631
Education
No education	Ref
High school or less	0.003619	0.167217	0.000431	0.002576	0.491989
College and more	0.026935	0.215886	0.000607	0.002809	0.692819
Current smoking
No	Ref
Yes	−0.01105	0.006182	−9.3E−06	−0.00151	−0.01067
Employment status
Employed	Ref
Retired	0.008264	0.178731	0.000252	0.001413	0.288404
Other (unemployed/homemaker)	0.005177	−0.26756	−0.00078	0.002933	−0.89645
Marital status
Single	Ref
Married	−0.0049	−0.12252	6.97E−05	−0.00057	0.079635

CI, Concentration Index.

The marginal effects for women in access to physicians, pharmacies, and laboratories are negative (−0.0099 to –0.01419, and −0.00283, respectively), with positive percentage contributions for physicians (1.36%), pharmacies (1.41%), and laboratories (0.30%). This indicates that sex differences slightly exacerbate inequalities in access to physicians.

Urban residents show greater access to physicians, pharmacies, and laboratories, as indicated by the positive marginal effects of 0.032, 0.078 and 0.053, respectively. The percentage contributions are the highest among all variables: 13.21% for physicians, 23.23% for pharmacies, and 17.39% for laboratories. These findings suggest that urban–rural disparities are a significant driver of inequalities in healthcare access.

Age shows minimal effects on access to physicians, with all age groups having small percentage contributions. Notably, the 65+ age group has a slightly negative contribution (−0.34%), suggesting marginally better access for older individuals. The marginal effects for all age groups regarding access to pharmacies and laboratories are negative, except for the 65+ age group, which shows a small positive contribution for pharmacies (0.05%) and laboratories (0.19%), indicating slight exacerbation of inequalities in access to these services due to age differences.

Higher levels of education are associated with better access to physicians. Individuals with a high school education or less contribute 0.64% to inequality, while those with a college education contribute 0.52%. Additionally, higher education levels have positive marginal effects of 0.026 and 0.025 for access to pharmacies and laboratories, with percentage contributions of 0.69% for pharmacies and 0.70% for laboratories.

Retired individuals have a positive marginal effect (0.0081) on access to physicians, contributing 0.39% to inequality. Conversely, unemployed or homemaker individuals show a negative contribution (−0.66%), suggesting a reduction in inequality. For pharmacy access, retired individuals have a marginal effect of 0.0082, with a positive percentage contribution of 0.28%. For laboratory access, retired individuals have a marginal effect of 0.0039, contributing 0.14% to inequality, while unemployed or homemaker individuals show a more substantial positive contribution (0.52%), indicating that employment status differences exacerbate inequalities in access to pharmacies.

Marital status has a marginal negative effect on access to healthcare services: −0.009 for physicians, −0.004 for pharmacies and −0.002 for laboratories. These account for 0.20%, 0.07% and 0.04% of the inequality in access to these services, respectively, suggesting that marital status does not significantly contribute to inequalities.

Discussion

This study aimed to analyse SES inequalities in access to healthcare services for patients with T2DM in Iran. Our findings indicate that asset is a critical determinant of healthcare access, with individuals in higher asset quintiles enjoying significantly better access to healthcare services. This observation is consistent with existing literature that underscores the role of economic resources in ensuring timely and effective medical care.^{8 13 14 25 26} Tatulashvili et al reported that low-income patients with T2DM face barriers such as high costs, limited access to outpatient services and greater distances from health centres, which impairs their ability to receive care.¹⁴ Similarly, other studies have shown that low-income patients are less likely to undergo essential health checks, such as blood sugar tests, compared with their higher-income counterparts.⁹ Evidence suggests that individuals with higher incomes are more likely to see specialists, even if they require less frequent care, due to better access and fewer financial constraints.^{25 26} While people with lower income due to lack of access to health services, over time, they will face unfavourable conditions of the disease.²⁷ Conversely, individuals with lower incomes often face significant barriers to accessing healthcare services, such as transport costs and lack of access to care facilities, which exacerbates their disease conditions over time.⁸ In these situations, poorer people are less likely to seek care and they are more likely to turn to traditional healers.²⁶ Low-income individuals are also more likely to adopt unhealthy lifestyles such as smoking, alcohol consumption, lack of exercise and poor diet, to cope with stressful life situations.⁷ These individuals tend to be less engaged in follow-up and less likely to adopt healthy behaviours such as eating adequate amounts of fruits and vegetables or quitting smoking.¹⁴ High SES is also associated with better health literacy, which improves the utilisation of health services.⁹ Wealthier individuals can afford private health services, which typically offer quicker and more reliable access to care. Additionally, financial resources may enhance an individual’s ability to navigate the healthcare system, including covering transportation and ancillary services.

Our analysis also revealed sex disparities in healthcare access, although the contribution of sex differences was relatively modest. The negative marginal effect for women suggests they have less access to physicians compared with men. This contrasts with findings from Maiti et al, who reported that men had poorer control and awareness of diabetes, leading to fewer healthcare service utilisations.⁵ One explanation for the lower healthcare service utilisation among women could be the higher psychosocial burden and reluctance to seek care for diabetes management.²⁸ This highlights the need for targeted interventions to address sex-specific barriers in healthcare access, which may be contributed by systemic biases, differences in health-seeking behaviour or cultural norms.

The study also underscores the significant impact of urban residence on healthcare access. Urban residents have better access to healthcare resources compared with their rural counterparts, aligning with previous research.^{5 9 11 12 15} This disparity is often attributed to limited healthcare infrastructure in rural areas, income disparities and differences in health-promoting behaviours.^{9 12 26} While some studies have reported better diabetes control in rural areas due to effective primary healthcare programmes,¹³ our findings indicate that urban areas generally offer superior health facilities and specialised care. This urban–rural divide underscores the need for policies aimed at improving healthcare access in rural areas, such as investing in telemedicine, rural health workforce, and transport infrastructure.

Age showed minimal association with healthcare access concentration in our study, with small contributions from different age groups. The slightly negative contribution of the 65+ age group suggests marginally better access, possibly due to age-related health conditions or enhanced health coverage for the elderly. Maiti et al found that younger individuals with diabetes are less likely to seek treatment, which aligns with our finding of minimal age-related differences in access.⁵ The small magnitude of age-related associations suggests that age is not a major determinant of healthcare access in this context.

Education, as a marker of SES, was positively associated with healthcare access. Higher educational levels correlated with better access, consistent with previous research.^{3 5 12 27 29} Individuals with a higher level of education tend to be more users of health services and are more likely to accept rehabilitation services and seek specialised care when diagnosed with diabetes.²⁹ A higher level of education indicates the ability to transform information into practical behaviours to manage and control chronic diseases on a regular basis.¹² On the other hand, individuals with a higher level of education experience less cultural distance from the doctor and, as a result, may have fewer problems interacting with the physician and following up their health condition.³⁰ This issue is not true for individuals with lower levels of education. These individuals avoid seeking healthcare due to stigma or limited knowledge, which increases their risk of diabetes.¹⁵ This finding is consistent with the notion that education enhances health literacy, enabling individuals to better navigate the healthcare system, advocate for themselves and use available health services more effectively.² Educational interventions that improve health literacy across all levels of society could be a key strategy in reducing inequalities in access to healthcare.

Employment status also contributed to access to healthcare, particularly for retired individuals. Pension benefits and increased time availability may facilitate healthcare access for retirees. Conversely, unemployed or homemaker individuals may face fewer barriers due to social support systems or more flexible schedules. Sortso et al reported less access to health services among retired individuals, which contrasts with our findings.²⁹ This discrepancy might be due to differences in income levels and willingness to seek care among retirees.

We found that the minimal association of marital status on access to health services suggests that marital status is not a significant determinant of access to healthcare in this context. Evidence shows that married individuals are more likely to access healthcare services.^{5 12} It can be said that marital status may be one of the factors affecting access to health services. While being married might offer some advantages, such as emotional and financial support that can facilitate access to care, these associations appear to be relatively minor. This finding suggests that other SES factors, such as asset and education, play a more pivotal role in determining access to healthcare services.

Strengths and limitations

A major strength of this study is the use of a large, nationally representative sample, which allows for robust estimation of socioeconomic inequalities in healthcare access. One important limitation of this study is the lack of direct income data, which necessitated the use of an asset-based index to measure SES. Although asset-based SES indices are widely accepted and reflect long-term economic conditions, they may not fully capture differences in wealth where individual assets vary substantially in value or social meaning. Therefore, some degree of residual socioeconomic misclassification cannot be entirely ruled out. Combining income and assets could have resulted in a more accurate and precise classification. Additionally, psychological factors and variables such as family size, smoking, body mass index, and other factors were not taken into account, potentially impacting the study result.

Conclusions

Our study reveals that higher SES, urban residence, retirement status, and higher education levels are associated with better access to healthcare services. Additionally, financial, geographical, knowledge, and cultural barriers contribute to the lack of access among lower SES individuals, worsening diabetes outcomes. These disparities may exacerbate diabetes outcomes. The observed socioeconomic inequalities highlight potential areas for policy attention, including disparities related to education and access to health information.

Data availability statement

Data are available upon reasonable request.