The effect of community-based health insurance on out-of-pocket expenditure among diabetic patients at hawassa university comprehensive specialized hospital: facility-based comparative cross-sectional study - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Health Serv Res . 2026 Apr 15;26:536. doi: 10.1186/s12913-026-14516-z Search in PMC Search in PubMed View in NLM Catalog Add to search The effect of community-based health insurance on out-of-pocket expenditure among diabetic patients at hawassa university comprehensive specialized hospital: facility-based comparative cross-sectional study Tsegaye Alemu Tsegaye Alemu 1 School of Public Health, College of Medicine and Health Sciences, Hawassa University, Hawassa, Ethiopia Find articles by Tsegaye Alemu 1 , Blen Asnake Blen Asnake 2 School of Public Health, Yanet-Liyana Health Sciences College, Hawassa, Ethiopia Find articles by Blen Asnake 2 , Mende Mensa Sorato Mende Mensa Sorato 3 Department of Pharmacy, College of Medicine and Health Sciences, Arba Minch University, Arba Minch, Ethiopia 4 Department of Pharmacy, School of Medicine, Komar University of Science and Technology, Qularaisi, Sulaimaniyah, KRI Iraq Find articles by Mende Mensa Sorato 3, 4, ✉ Author information Article notes Copyright and License information 1 School of Public Health, College of Medicine and Health Sciences, Hawassa University, Hawassa, Ethiopia 2 School of Public Health, Yanet-Liyana Health Sciences College, Hawassa, Ethiopia 3 Department of Pharmacy, College of Medicine and Health Sciences, Arba Minch University, Arba Minch, Ethiopia 4 Department of Pharmacy, School of Medicine, Komar University of Science and Technology, Qularaisi, Sulaimaniyah, KRI Iraq ✉ Corresponding author. Received 2025 May 19; Accepted 2026 Apr 5; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13088705 PMID: 41987133 Abstract Background Despite the global target of 80% glycemic control among people diagnosed with diabetes in 2030, diabetes treatment coverage and control rate were still low. Diabetes imposes a substantial economic burden on health systems, patients, and their families. Due to low health insurance coverage in developing countries, the expenses related to diabetes care often result in significant out-of-pocket costs for patients. Objective To assess the level of out–of–pocket expenditure and the effect of community-based health insurance (CBHI) on out-of-pocket (OOP) expenditure among diabetic patients on follow-up at Hawassa Comprehensive Specialized Hospital, Sidama region. Method A facility-based comparative cross-sectional was conducted among 314 randomly selected adult type 2 diabetics. Kobo Collect app and SPSS version 26 were used data collection and analysis respectively. Independent sample t-test and linear regression were used to compare OOP expenses between CBHI members and non-members, and assess the association between CBHI-enrollment and OOP expenses. Result The average monthly household expenditure among participants was 6,471.52 Ethiopian Birr (ETB) (SD ± 3,275.47). Of which, average monthly costs of 3,568.55 ETB for food and 2,902.97 ETB for non-food items. The average monthly expenditure for diabetic illness was 2,046.00 ETB (SD ± 3,173.50), of which 869.67 ETB (42.5%) were direct medical costs. Regarding incidence and severity of catastrophic health expenditure (CHE), 82.5% of patients faced CHE at the 10% threshold, while only 27.4% did so at the 40% threshold. The intensity of OOP was 67 (43.2%), and 120 (75.5%) among CBHI members and non-members, respectively ( p = 0.000). The CBHI enrollment is significantly associated with a reduction in OOP expenses, with a coefficient of -499.410 ( p = 0.000). Similarly, age and occupation of participants were associated with lower OOP expenses, with a coefficient of -8.756 ( p = 0.028) and − 58.221 ( p = 0.002), respectively. Educational status of participants was associated with higher OOP expenses, with a coefficient of 104.416 ( p = 0.004). However, marital status, household size, and wealth percentile group did not have a significant effect on OOP expenses among diabetic patients. Conclusion A significant proportion of diabetes patients experienced CHE. Enrollment in CBHI lowered OOP costs and reduced CHE. But there are still gaps in CBHI coverage, especially when it comes to the lowest and middle wealth quantiles. To enhance CBHI effectiveness, it is important to prioritize rural and low-income households to reduce financial strain. Introduce a subsidy program to make diabetes medications more affordable for uninsured households. Incorporate beneficiary feedback to refine policies and address the diverse needs of enrollees. Future research to evaluate the long-term impacts of CBHI on diabetic patient expenses on household financial stability by involving hospitals from rural and urban sectors is important to better understand the effect of CBHI on OOP expense and catastrophic health expenditure in the region. Supplementary Information The online version contains supplementary material available at 10.1186/s12913-026-14516-z. Keywords: Diabetes, Out-of-pocket expenditure, Community-based health insurance, Catastrophic health spending, Household expenditures on health, Ethiopia Introduction Diabetes mellitus is a chronic metabolic condition that leads to high levels of glucose in the blood when the body cannot effectively use or does not produce enough insulin [ 1 ]. According to a systematic review from the Global Burden of Disease study, 529 million people were living with diabetes worldwide, with global age-standardized total diabetes prevalence being 6.1% in 2021. This is projected to increase to 1.31 billion with an age-standardized prevalence rate above 10% in 2050 [ 2 ]. The estimated prevalence of diabetes in Ethiopia was 2 to 6.5% [ 3 ]. Similarly, the estimated prevalence of diabetes in Hawassa Zuria Woreda, Sidama region is 12.4% [ 4 , 5 ]. Despite the global target of 80% glycemic control among people diagnosed with diabetes in 2030, diabetes treatment coverage and control rate were still low. A pooled analysis of population-representative studies revealed that 445 million adults aged 30 years and above did not receive treatment in 2022. About 90% of these untreated patients were in low- and middle-income countries (LMICs) [ 6 ]. Health expenditures is defined as the provision of emergency assistance devoted to preventive and curative health services, family planning, nutrition, and hygiene, but does not include water and sanitation services [ 7 ]. Whereas, out-of-pocket expenditure refers to household expenses incurred when utilizing a healthcare service. Similarly, out-of-pocket expenses are defined as payments made directly to healthcare providers at the time of service, excluding insurance-related prepayments [ 7 ]. Diabetes contributes for significant financial burden to patients and their families. The global health expenditure due to diabetes for adults has increased from $232 billion USD in 2007 to $966 billion USD in 2021. This is projected to rise to $1.03 trillion USD and $1.05 trillion USD in 2030 and 2045, respectively [ 1 ]. Due to low health insurance coverage (≈ 17%) in developing countries, the expenses related to diabetes care often result in significant out-of-pocket expenditure [ 8 , 9 ]. Out-of-pocket (OOP) expenses are a key factor contributing to the financial hardship of patients and families [ 10 ]. For individuals with chronic conditions like diabetes, the burden of paying for medical services OOP can hinder access to care, leading to delayed treatment, and increased rates of morbidity and mortality [ 11 ]. In Ethiopia, OOP expenditure accounts 31% of its total health expenditures in 2024/2025, which is significantly more than the 20% global benchmark [ 12 ]. Research conducted in Bahir Dar found that catastrophic health expenditures (CHE) accounted for 59.6% of diabetes patients [ 13 ]. The average monthly cost of diabetes care was estimated to be $37.7 USD (76.2% direct and 23.8% indirect cost respectively) [ 14 ]. About 59.6% of diabetic patients incurred CHE at 40% non-food threshold and 5% impoverished at $1.90-a-day poverty line [ 13 ]. These high costs can limit access to necessary care, causing households to fall into poverty. Studies from South Africa, India, and the US highlight the significant impact of diabetes treatment expenses, particularly for the impoverished, leading to financial hardship and restricted access to care [ 15 – 17 ]. In Ethiopia, household income, proximity to healthcare facilities, and health insurance coverage play key roles in determining out-of-pocket costs [ 18 ]. As a result of the negative consequences that high OOP expenses have on patients suffering from diabetes, health insurance or pre-payment strategies are being developed to minimize immediate dependency [ 19 ]. Community-based health insurance (CBHI) program has been endorsed in Ethiopia since 2011 as part of new health financing reforms, and it is designed to pool risk and protect households from out-of-pocket expenditure [ 19 ]. Even though the Government’s efforts to address the challenge of high out-of-pocket spending for health services through the introduction of CBHI, the CBHI enrollment in the Sidama region remained low [ 20 – 22 ]. Different studies revealed the relationship between CBHI enrolment, out-of-pocket expenditure, and socio-demographic and socio-economic variables [ 13 , 23 ] (Fig. 1 ). In low- and middle-income nations like Ethiopia, there is little data regarding the effect of CBHI on OOP expense. Addressing the CBHI enrollment effect on OOP is important to reduce the soaring global and national burden diabetes of diabetes and financial hardship of patients and families [ 24 ]. This study aimed to assess the effect of community-based health insurance on out-of-pocket expenses and level of catastrophic health expenditure among diabetic patients attending Hawassa University Comprehensive Specialized Hospital. In addition, the study could provide crucial information for policymakers seeking to refine CBHI programs and highlight the need for further research into systemic barriers and the long-term economic impact of health insurance on managing chronic illnesses. Fig. 1. Open in a new tab Conceptual frame work for the study, the effect of CBHI on out-of-pocket expenditure among diabetic patients Methods and materials Study design, area, and period A facility-based comparative cross-sectional study was conducted at Hawassa University Comprehensive Specialized Hospital (HUCSH), Sidama regional state, Ethiopia, from June 16 to July 30, 2024. The Hospital serves as a pivotal healthcare institution in the Sidama regional state, which is 275 km south of Addis Ababa, the capital city of Ethiopia. Hawassa city has 32 kebeles and eight sub-cities with a total population of 399,461 [ 5 ]. The hospital was established in 2006 with the primary aim of offering specialized care and serving as a referral center for the region, The hospital is strategically positioned to address the healthcare needs of diverse communities. The hospital has over 400 beds and provides high-quality patient care in a broad range of services. HUCSH offers services at general and specialty levels, including internal Gynecology & obstetrics, ENT (Ear, Nose and Throat), orthopedic surgery, neurosurgery, colorectal surgery, maxillofacial surgery, neurology, urology, psychiatry, ophthalmology, dermatology, dentistry, radiology, and laboratory and pharmacy services. It is the first hospital in southern Ethiopia to start special services like screening for diabetic retinopathy with a retinal photo camera. There were 1323 registered diabetic patients at HUCSH [ 25 ], of which about 1240 were adults on follow-up. All adult diabetic patients aged 18 and above receiving care in Hawassa specialized hospital were candidates for the study. Population The source population was all diabetic patients attending a follow-up appointment at HUCSH. The study population was diabetic patients aged 18 and above who received care and met the inclusion criteria. Inclusion and exclusion criteria All Diabetic patients aged 18 years and above receiving follow-up appointments at HUCSH for at least a month and who agreed to participate in the study were included. However, critically ill patients, patients less than 18 years old, and clients who were pregnant (may be gestational diabetes, which is transient since it will be resolved after pregnancy) were excluded. Sample size determination Sample size for objective one The sample size to address the primary objective was calculated using a single population proportion formula by taking the magnitude of catastrophic health expenditure (CHE) as 59.6% [ 13 ] among diabetic patients, and the magnitude of CHE among chronic disease patients in the rage of 39% to 64.2% ADDIN EN.CITE [ 23 , 26 ]. Where n = is the sample size. Z 2 = standard normal deviation, set at 1.96, corresponds to the 95% confidence interval. d = is the desired level of precision/margin of error (0.05) p = Magnitude of CHE among diabetics (p=59.6%), and q is 1-p. By taking the magnitude of CHE among diabetic patients, the estimated sample size was 370. Similarly, the estimated sample size based on CHE among chronic disease patients was 283 and 275, based on 39% and 64.2% CHE magnitude. Due to the small source population (i.e., adult diabetic patients in follow-up at Hawassa City), less than 10,000 ( N = 1240) [ 25 ], the finite population correction was made by using the formula . Where : NC= corrected sample size. n=calculated sample size. N=diabetic population in Hawassa city. The sample size after applying finite population correction was 285 based on diabetic patients, CHE, 283 and 275 based on 39% and 64.3% CHE among chronic disease patients. Therefore, a larger sample based on diabetic patients CHE after applying 10% for non-response (285 + 28.5 = 314) was used in this study. Sample size for objective two The double population proportion formula was used to determine the sample size for objective two using Epi-info version 7.2.6 stat-calc. The proportion of CHE was taken from the effect of CBHI on catastrophic health expenditure among chronic patients, which is 7% and 19% among insured and noninsured patients, respectively [ 27 ], 95% CI, 80% power and 5% degree of precision with one-to-one ratio among insured and non-insured was taken. Where; r: = the allocation ratio of insured to uninsured i.e., n2: n1 (n2 = rn1). p1 = proportion of CHE in insured. P2 = proportion of CHE in uninsured. Za/2 = the quintile of the standard normal distribution for type I error. Zb = the quintile of the standard normal distribution for type II error/power. n1 = the sample size for insured. n2 = the sample size for uninsured. Based on this formula, the calculated sample size was 278. Considering 10% non-response rate to adjust sample size the final sample size for secondary objective would be 306 (153 insured and 153 uninsured). The final sample size for this study was determined by comparing the two computations and choosing the one with the largest sample size. As a result, 314 diabetic patients (155 CBHI members and 159 non-members) were recruited in this study. Sampling technique and sampling procedure Simple random sampling by using lottery method was applied to select eligible study participants from diabetic out- patient department of Hawassa University Comprehensive Specialized Hospital (HUCSH). Study variables Dependent variable Out-of-pocket expenditure for diabetes care. Independent variables Socioeconomic variables : Sex, age, religion, marital status, residence, family size, distance from health facility, educational status, occupation, monthly income, source of income, wealth index, food and non-food expenditure of households. Health insurance related variables : CBHI membership status, type of membership, means getting service, benefits of CBHI, payment systems, and ease of access to health diabetes care. Clinical characteristics : Duration of diabetes since diagnosis, presence of comorbidity. Direct medical costs : cost of registration, outpatient care, follow-up visits, laboratory, drugs/medicines, and imaging studies. Direct non-medical costs : Cost of food, transportation, caregiver/family member, and accommodation. Indirect costs : Lost workday wages of patients and caregivers due to diabetes care and follow-up visits. Coping strategies : Personal savings, borrowing, selling assets, family support, reducing household food consumption, CBHI, and fee waiver systems. Data collection instrument and data collection procedure The data collection questionnaire was adapted from related studies [ 13 , 18 , 23 , 28 ] and is available in supplementary material (Data collection tools). The questionnaire was developed in English and translated into Amharic (the most generally spoken local language in the study area) and back translated to check for consistency. Data were collected from diabetic patients who were attending follow-up appointments and their respective charts. The data were collected by using the Kobo Collect tool app on the data collector’s smartphone. Appropriate confidentiality (restricting server access to authorized personnel, protecting all KOBO collect passwords and using limited code information for creating tool box), and data availability and security details (daily backup of the database to a separate, remote location) were considered. Two data collectors (BSc nurses) were recruited and carefully collected the data. Catastrophic health expenditure (CHE), that is, household’s out-of-pocket payment that exceeds 10% of total household expenditure or 40% of non-food expenditure, was estimated by using the WHO catastrophic health expenditure estimation approach available at: https://www.who.int/data/gho/indicator-metadata-registry/imr-details/4989 . Out-of-pocket payment is defined as formal and informal payments made at the time of using any health care good or service provided by any type of provider; including user charges (co-payments) for covered services and direct payments for non-covered services; and excluding any pre-payment in the form of taxes, contributions or insurance premiums and any reimbursement by a third party. A household’s capacity to pay for healthcare is defined as per adult equivalent total household consumption minus a standard amount to cover basic needs. Costs of diabetes and associated comorbidities were computed by including direct and indirect costs. Direct costs, including direct medical (outpatient visit, follow-up, medication, laboratory, imaging studies) were abstracted from the patient chart in the index year 2025. The cost of medicines was taken from the Ethiopian Pharmaceutical supply service Hawassa Hub selling price, and the retail price of HUCSH in 2025. Direct non-medical costs included the cost of transportation, food, and accommodation. Indirect costs included lost work days of the patient, family/caregivers due to diabetes, and complications. Direct medical costs were estimated based on the facility cost list. Direct non-medical costs were estimated by taking the average cost of meals, accommodation, and round trip of patients to the hospital. Costs of laboratory procedures and imaging studies were also taken from the HUCSH laboratory’s service and imaging studies price list. The salary scale of the health workforce was not included because it is a common cost for insured and non-insured diabetic patients and will not create a change in the overall burden of diabetes among these patients. Data quality assurance To enhance the quality of the data, data collectors were trained for two days on the objective and methodology of the research and data collection approach. The questionnaire was pilot-tested among 5% of the sample (16 diabetic patients) attending follow-up in Adare General Hospital and was modified based on the findings of the pilot study. The data was collected using the Kobo Collect app was checked on the server every night of the data collection date, and a backup was taken. The principal investigators conducted continuous supervision and follow-up throughout the data collection period. Data processing and analysis The collected data underwent rigorous cleaning and preprocessing to ensure accuracy and reliability. This process included checking for missing values, outliers, and inconsistencies. Missing data were addressed using appropriate imputation methods or exclusion where necessary, and variables were recoded and categorized to enhance interpretability. For instance, demographic and economic variables were grouped into meaningful categories to facilitate analysis. Following preprocessing, the normality of the dependent variable, out-of-pocket expenses, was assessed using the Shapiro-Wilk test alongside checks for skewness and kurtosis. The Shapiro-Wilk test yielded a significant result ( p < 0.001), indicating that the data did not follow a normal distribution. Pearson Chi-Square test was used to assess the association between dependent variable and independent variables. Descriptive statistics provided an overview of the variables and summarized participants’ economic and demographic characteristics, guiding the selection of appropriate statistical tests. Additionally, linear regression analysis was performed to examine the association between CBHI enrollment and out-of-pocket expenses, adjusting for potential confounders such as age, income, and other socioeconomic factors. Statistical significance was set at p < 0.05 to ensure robust conclusions about the financial impact of CBHI. This comprehensive approach to data processing and analysis enabled the study to draw meaningful insights into the relationship between CBHI enrollment and the financial burden of diabetes care. Operational definition Diabetic patients Participants with documented evidence of a medical diagnosis of diabetes in their medical records at Hawassa Comprehensive Specialized Hospital will be classified as having diabetes. CBHI enrollment Households that have been members of community-based health insurance for at least a month and have renewed their membership for the fiscal year of 2016. Out-of-pocket expenditure Direct and indirect financial costs incurred by diabetic patients for healthcare services related to diabetes management not covered by any health insurance. Catastrophic health spending an out-of-pocket payment greater than 40% of capacity to pay for health care. Or > 10% or 40% of total household consumption and non-food expenditure, respectively [ 29 ]. Results Demographic and socio-economic characteristics of respondents In total, 314 study participants voluntarily participated in this investigation, yielding a 96.7% response rate. Among the participants, 165 (52.5%) (95 insured and 70 non-insured) were male. The age of the respondents ranged from 18 to 80 years, with a mean age of 43.79 (SD ± 16.12). Most participants were married (59.2%), identified as Orthodox Christians (35.7%), and lived in urban areas (84.1%). The mean household size of the respondents was 5.00 (SD ± 2.34), with a range of 1 to 13. The average distance to health facilities was 14.83 (SD ± 17.68) kilometers from the hospital. Regarding the educational status of participants, 34.4% of the participants had completed secondary education, and 28.0% had attained tertiary education. Employment status revealed that 28.7% of the participants were employed, while 16.6% were housewives. Wealth distribution showed that the majority (54.1%) of respondents fell within the middle quintile, while the lowest, second, fourth, and highest quintiles each accounted for 10.8%–11.8% of the sample (Table 1 ). Table 1. Demographic and socio-economic characteristics of diabetic patients, Hawassa Comprehensive Specialized Hospital, 2024 Characteristic Category CBHI enrolled ( n = 155) Not enrolled in CBHI ( n = 159) Pearson chi-square (p-value) Frequency Percent Frequency Percent Sex Male 95 57.6% 70 42.4% 9.383 (0.002) * Female 60 40.3% 89 59.7% Age (years) 18–30 34 36.2% 60 63.8% 17.259 (0.001) * 31–45 33 42.9% 44 57.1% 46–60 57 58.8% 40 41.2% 61–80 31 67.4% 15 32.6% Religion Orthodox 53 47.3% 59 52.7% 5.404 (0.145) Muslim 44 61.1% 28 38.9% Catholic 11 42.3% 15 57.7% Protestant 47 45.2% 57 54.8% Marital Status Single 37 37.0% 63 63.0% 11.708 (0.002) * Married 106 57.0% 80 43.0% Widowed 6 46.2% 7 53.8% Separated 2 28.6% 5 71.4% Divorced 4 50.0% 4 50.0% Residence Urban 124 47.0% 140 53.0% 3.799 (0.51) Rural 31 62.0% 19 38.0% Household Size ≤ 5 79 39.7% 120 60.3% 20.304 (0.000) * > 5 76 66.1% 39 33.9% Distance to Facility (km) ≤ 25 125 46.1% 146 53.9% 11.378 (0.010) * 26–50 19 63.3% 11 36.7% 51–75 4 66.7% 2 33.3% ≥ 76 7 100.0% 0 0.0% Educational Status Illiterate 27 73.0% 10 27.0% 16.676 (0.002) * Read and write 12 44.4% 15 55.6% Primary school (1–8) complete 33 62.3% 20 37.7% Secondary school (9–12) complete 40 44.9% 49 55.1% College and above (> 12) 43 39.8% 65 60.2% Occupation Farmer 30 69.8% 13 30.2% 26.996 (0.001) * Employee 58 64.4% 32 35.6% Housewife 25 48.1% 27 51.9% Self-employed 9 42.9% 12 57.1% Unemployed 1 14.3% 6 85.7% Merchant 14 40.0% 21 60.0% Daily laborer 2 25.0% 6 75.0% Student 16 33.3% 32 66.7% Retired 5 50.0% 5 50.0% Wealth Quintiles Lowest 18 52.9% 16 47.1% 75.252 (0.000) * Second 16 43.2% 21 56.8% Middle 54 31.8% 116 68.2% Fourth 31 86.1% 5 13.9% Highest 36 97.3% 1 2.7% Duration of diabetes since diagnosis in years Below one year 11 7.1% 26 16.4% 6.957 (0.031) 1–5 years 119 76.8% 114 71.6% 6–10 years 25 16.1% 19 12.0% Open in a new tab Community-based health insurance enrolment status Nearly half of the respondents, 155(49.4%), were enrolled in CBHI. The duration of CBHI membership varied widely, ranging from 2 to 78 months, with a mean of 32.98 months (SD ± 13.80). Among the members, the majority, 141 (91.0%), paid for their enrollment through household contributions, while the rest received financial support from local government sources. Nearly all CBHI members 154 (99.4%) reported benefiting from the program. The most commonly cited advantage, reported by 140 members (90.9%), was reduced concerns about expected healthcare costs. Other notable benefits included reduced healthcare expenses and improved access to healthcare services. Despite these benefits, a significant proportion 67(43.2%) of CBHI enrolled members still paid a portion of their healthcare bills out-of-pocket, indicating partial coverage by the scheme. In terms of ease of accessing medical services through CBHI, most respondents 60 (38.7%) described it as “easy,” while a smaller group 10 (6.5%) found it “difficult. Regarding clinical characteristics, 76 (40 insured and 36 non-insured) patients had comorbidity with Pearson chi-square value of 0.429 at p = 0.513. Hypertension was the most common comorbidity, affecting 40 (52.6%) followed by pneumonia, 12 (15.8%), and congestive heart failure (9.2%). The mean duration of illness among patients living with diabetes was 37.5 months (SD ± 22.6) (Table 2 ). Table 2. Clinical and CBHI status of respondents attending Diabetic follow-up in Hawassa Comprehensive Specialized Hospital, 2024 Characteristic Category Frequency Percent CBHI member Yes 155 49.4 No 159 50.6 Payment type ( N = 155) Household contribution 141 91.0 Local government 14 9.0 Benefited from CBHI Yes 154 99.4 No 1 0.6 Types of benefits ( N = 154) Increased access to healthcare 88 57.1 Reduced costs of health care 98 63.6 Reduced concerns about expected healthcare costs 140 90.9 Currently get the service in Hospital( N = 155) Free of charge 88 56.8 Partial OOP payment 67 43.2 Ease of getting medical Access using CBHI Difficult 10 6.5 Average 54 34.8 Easy 60 38.7 Very easy 31 20 Duration of diabetes since diagnosis Below one year 37 11.8 1–5 years 233 74.2 6–10 years 44 14.0 Type of comorbidity diagnosed ( n = 76) HTN 40 52.6 Liver CA 1 1.3 Neuropathy 2 2.6 Pneumonia 12 15.8 TB 6 7.9 Urinary tract infection 1 1.3 Uterine tumor 1 1.3 Breast CA 3 3.9 CHF 7 9.2 Chronic liver disease 3 3.9 Regular follow-up Yes 314 100 No 0 0 Follow-up appointment Every month 155 49.4 Every two months 86 27.4 Every three months 72 22.9 Every six months 1 0.3 Open in a new tab Cost of diabetic disease treatment and household expenditure The average monthly household expenditure among participants was 6,471.52 ETB (SD ± 3,275.47). This expenditure was divided between food and non-food items, with average monthly costs of 3,568.55 ETB (SD ± 2,036.42) for food and 2,902.97 ETB (SD ± 1,852.10) for non-food items. Regarding the costs associated with managing diabetes, the average monthly expenditure for diabetic illness was 2,046.00 ETB (SD ± 3,173.50), of which 869.67 ETB (SD ± 746.72) were direct medical costs. Most participants (229) incurred medication expenses, with an average monthly cost of 543.21 ETB, while a smaller number (69) paid for imaging, averaging 112.25 ETB monthly. For direct non-medical costs, 228 patients reported an average monthly expenditure of 227.75 ETB (SD ± 319.03). The most common non-medical expense was transportation, with 224 participants paying an average of 165.04 ETB (SD ± 194.44) for travel related to their follow-up visits. Only seven participants incurred accommodation costs, averaging 241.42 ETB. In terms of productivity loss, which reflects the indirect costs due to missed work or caregiving responsibilities, the average monthly loss was 1,010.95 ETB (SD ± 3,042.26) (Table 3 ). Table 3. Cost of diabetic illness and household expenditure of diabetic disease patients attending follow-up in Hawassa Comprehensive Specialized Hospital, 2024 Types of cost N MEAN STD.DEV Household Expenditure per month Food expenditure 314 3568.55 2036.42 Non-Food Expenditure 314 2902.97 1852.1 Total expenditure 314 6471.52 3275.47 Direct medical cost per month Registration/Card 177 39.45 34.77 Laboratory 168 174.74 254.4 Drug 229 543.21 503.74 Imaging 69 112.25 294.29 Total-direct medical 263 869.67 746.72 Direct non-medical cost per month Food and Drink 46 288.43 372.49 Transportation 224 165.04 194.49 Accommodation 7 241.42 108.07 Total-direct non-medical 228 227.75 319.03 Total direct cost 314 1035.05 778.97 Indirect cost per month 314 1010.95 3042.26 Total cost 314 2046 3173.5 Open in a new tab Regarding types of healthcare costs incurred based on insurance status, the most common types of health costs incurred by diabetic patients were found to be indirect costs, which accounted for 49.41% of the total cost of managing diabetes, followed by direct medical costs, 42.5% (Table 4 ). As illustrated in Fig. 2 a, on average, CBHI members incurred a lower monthly cost of 1,488.20 ETB compared to 2,589.77 ETB for uninsured households. The bar graph highlights the differences in direct medical costs, non-medical costs, and indirect costs between the two groups. The proportion of total expenses is also effectively demonstrated in Fig. 2 b. The pie chart showed that for both groups, indirect costs made up a significant share, with CBHI members incurring 48% and uninsured households 50.2%. This provides a visual representation of the weight of indirect costs in the overall financial burden of managing diabetes. Medication was the most frequently incurred expense for both groups, with 96.9% of uninsured patients and 72% of CBHI members purchasing medications. Registration fees were paid by 93.7% of uninsured patients, while laboratory and imaging services were utilized by 76.7% and 32.1%, respectively. Table 4. Cost of diabetic illness among patients attending follow-up by CBHI status at Hawassa Comprehensive Specialized Hospital, 2024 Types of cost CBHI enrollment = Yes CBHI enrollment = No N Mean Std.dev N Mean Std.dev Direct medical cost per month Registration/Card 28 12.64 27.01 149 65.59 17.04 Laboratory 46 58.16 117.20 122 288.38 292.48 Drug 75 428.76 596.14 154 654.78 361.83 Imaging 18 66.99 238.80 51 156.38 334.68 Total 104 566.57 735.50 159 1165.15 631.95 Direct nonmedical cost per month Food and Drink 9 34.12 223.62 37 50.17 106.58 Transportation 123 164.16 220.65 101 72.47 113.14 Accommodation 5 9.03 52.05 2 1.8 17.35 Total direct non-medical 124 207.32 357.96 104 124.47 196.03 Indirect cost per month 72 714.30 1815.13 79 1300.00 3866.94 Total out of pocket 155 1488.20 2048.77 159 2589.77 3906.02 Open in a new tab Std.dev. = standard deviation; CBHI= Community-Based Health Insurance Fig. 2. Open in a new tab Proportion of costs for diabetic patients based on insurance status ( A ). Proportion of main cost types for diabetic patients at Hawassa Comprehensive Specialized Hospital ( B ) Catastrophic health expenditure (CHE) Regarding incidence and severity of CHE, 82.5% of patients faced catastrophic costs at the 10% threshold, while only 27.4% did so at the 40% threshold. When overall household expenditure was used, the incidence of CHE decreased, from 59.6% at the 10% threshold to 41.7% at the 40% threshold. The severity of CHE, measured using overshoot, reflects how much health expenditures exceeded the threshold. For non-food expenditure, the overshoot was 20.01% at the 10% threshold and decreased to 17.17% at the 40% threshold. For total household expenditure, the overshoot was much higher, at 71.4% at the 10% threshold, and 50.3% at the 40% threshold. Additionally, mean positive overshoot (MPO), which indicates the average percentage by which health expenditures exceed the threshold for households experiencing CHE, varied depending on the threshold and expenditure type. At the 10% and 15% thresholds for total expenditure, the MPO was 16.51% and 14.99%, respectively. For non-food expenditure, the MPO was 26.7% at the 10% threshold and 18.74% at the 40% threshold. About, the incidence and severity of CHE between CBHI members and non-members. The incidence of CHE was consistently higher among uninsured patients across all thresholds. For example, 89 (57.4%) of CBHI members experienced CHE at the 15% non-food expenditure threshold, while 133 (83.6%) of non-members did the same. When total household expenditure was considered, the incidence at the 15% threshold was 54 (34.8%) for insured patients, whereas it was 77 (49.4%) for uninsured patients. The severity of out-of-pocket (OOP) expenses, as measured by the mean positive overshoot (MPO), was lower for uninsured households. At the 15% and 40% thresholds for total expenditure and non-food expenditure, uninsured households had a lower MPO compared to insured households. Specifically, at the 15% threshold for total expenditure, the MPO was 17 (10.7%) for uninsured households, compared to 23 (14.8%) for insured households (Table 5 ). Table 5. Incidence and intensity of CHE by CBHI status among diabetic patients at Hawassa Comprehensive Specialized Hospital, 2024 Out-of-pocket expenditure as percentage of total household expenditure Category Overall ( n = 314) CBHI enrolled ( n = 155) Not enrolled in CBHI ( n = 159) Pearson chi-square ( p -value) Frequency (%) Frequency (%) Frequency (%) > 10% No 88 (56.8%) 39 (24.5%) 33.881 (0.000) * Yes 187 (59.5%) 67 (43.2%) 120 (75.5%) > 15% No 101(65.2%) 82 (51.6%) Yes 131 (41.7%) 54 (34.8%) 77 (49.4%) 5.961 (0.015) * > 25% No 87 (56.1%) 74 (46.5%) Yes 153 (48.7%) 68 (43.9%) 85 (53.5%) 2.888 (0.089) > 40% No 115 (74.1%) 113 (71.1%) Yes 86 (27.4%) 40 (25.8%) 46 (28.9%) 0.385 (0.535) OOP > 25% of capacity to pay No 125 (80.6%) 122 (76.7%) Yes 67 (21.3%) 30 (19.4%) 37 (23.3%) 0.717 (0.397) OOP > 40% of capacity to pay No 136 (87.7%) 129 (81.1%) Yes 49 (15.6%) 19 (12.3%) 30 (18.9%) 2.604 (0.107) OOP as a share of non-food expenditure Head count (%) > 10% 259 (82.5%) 106 (68.4%) 153 (96.2%) > 15% 222 (70.7%) 89 (57.4%) 133 (83.6%) 268.611 (0.091) > 25% 153 (48.7%) 68 (43.9%) 85 (53.5%) > 40% 86 (27.4%) 40 (25.8%) 46 (28.9%) Overshoot (%) > 10% 63 (20.0%) 28 (18.3%) 35 (21.7%) > 15% 50 (16.1%) 23 (15.1%) 27 (17.1%) 265.325 (0.331) > 25% 32 (10.2%) 16 (10.0%) 16 (10.4%) > 40% 152 (47.0%) 75 (48.4%) 77 (48.7%) Mean positive overshoot (%) > 10% 77 (24.3%) 41 (26.7%) 36 (22.6%) > 15% 74 (22.8%) 41 (36.7%) 33 (20.5%) > 25% 66 (20.9%) 35 (22.5%) 31 (19.4%) > 40% 54 (17.2%) 29 (18.7%) 25 (15.8%) OOP as a share of total expenditure Head count (%) > 10% 187 (59.6%) 67 (43.2%) 120 (75.5%) 197.220 (0.209) > 15% 93 (41.7%) 54 (34.8%) 39 (25.5%) 126.276 (0.501) Overshoot (%) > 10% 239 (76.1%) 111 (71.6%) 128 (80.5%) > 15% 75 (23.8%) 44 (28.4%) 31 (19.5%) Mean positive overshoot (%) > 10% 43 (13.7% 26 (16.7%) 17 (10.7%) > 15% 40 (12.7% 23 (14.8%) 17 (10.7%) Open in a new tab OOP = Out-of-Pocket expenditure; CBHI= Community Based Health Insurance Coping mechanism with healthcare expenditure for the uninsured The intensity of OOP was 67 (43.2%), and 120 (75.5%) among CBHI members and non-members, respectively, with a Pearson chi-square value of 33.881 ( p = 0.000). To cope with the financial burden of catastrophic health expenses (defined as > 10% of total household expenditure), 98 (61.6%) non-CBHI member patients relied on their savings, followed by 52 (32.7%) received support from family members. Only a small fraction (3 or 1.8%) of respondents had to sell assets, such as land or a vehicle, to fund their medical costs. Similarly, 42 (27.1%) of CBHI-insured households still paid for their medical bills out of pocket by using personal savings, followed by family support, 22 (14.2%) (Fig. 3 ). Fig. 3. Open in a new tab Healthcare financing and coping mechanisms of diabetic patients at Hawassa Comprehensive Specialized Hospital, 2024 Effect of CBHI on out-of-pocket expenditure Regarding the effect of CBHI on out-of-pocket expenditure, DM patients’ out-of-pocket expenses were examined using a variety of statistical methods to assess the impact of CBHI enrollment on out-of-pocket expenses. The mean out-of-pocket (OOP) costs for diabetic patients enrolled in the CBHI scheme versus those without enrollment showed that CBHI enrollees incurred significantly lower mean OOP expenses, 773.90 (SD ± 807.08 ETB), compared to non-enrolled patients, 1289.64 (SD ± 659.59 ETB). To assess the equality of variances in OOP expenditures between the two groups, Levene’s test for equality of variances was conducted. The test result F (1,312) = 6.61, p = 0.009, was significant, indicating that the assumption of equal variances was violated. The unequal variance t-test revealed a statistically significant difference in OOP costs between the two groups, t (297.07) = 6.191, p < 0.001. Patients enrolled in CBHI experienced substantially lower OOP expenses compared to their non-enrolled counterparts. The effect size, calculated using Cohen’s d (d = -0.701), suggests a moderate practical significance for this difference. These findings highlight the substantial financial relief provided by CBHI enrollment, emphasizing its effectiveness in reducing the healthcare burden for diabetic patients. In addition, the impact of CBHI enrollment, age, marital status, household size, family size, education, occupation, and wealth on OOP expenses among diabetic patients was assessed using multiple linear regression analysis. The regression model was statistically significant (F = 7.005, p < 0.001) and explained 13.8% of the variance in OOP expenses (R² = 0.138, adjusted R² = 0.118). The results, as presented in Table 6 , indicate that CBHI enrollment is significantly associated with a reduction in OOP expenses, with a coefficient of -499.410 ( p = 0.000). This suggests that patients enrolled in CBHI incur significantly lower OOP costs compared to those who are not enrolled. Similarly, age and occupation of participants were associated with lower OOP expenses, with a coefficient of -8.756 ( p = 0.028) and − 58.221 ( p = 0.002), respectively. Educational status of participants was associated with higher OOP expenses, with a coefficient of 104.416 ( p = 0.004). However, the analysis revealed that variables like marital status, household size, and wealth percentile group did not have a significant effect on OOP expenses among diabetic patients (Table 6 ). Table 6. Linear regression results for factors influencing out-of-pocket expenses among diabetic patients Coefficients a Model Unstandardized Coefficients Standardized Coefficients t Sig. 95.0% Confidence Interval for B B Std. Error Beta Lower Bound Upper Bound (Constant) 2224.110 300.587 7.399 0.000 2224.110 300.587 Age of participants -8.756 3.962 − 0.171 -2.210 0.028* -8.756 3.962 Marital status of respondents -7.979 63.866 − 0.008 − 0.125 0.901 -7.979 63.866 Number of family members -11.810 19.750 − 0.034 − 0.598 0.550 -11.810 19.750 Educational status of participants 104.416 35.583 0.170 2.934 0.004* 104.416 35.583 The main occupation of the participant -58.221 18.398 − 0.189 -3.165 0.002* -58.221 18.398 Community-based health insurance membership -499.410 92.101 − 0.303 -5.422 0.000* -499.410 92.101 Wealth percentile group -27.357 40.876 − 0.035 − 0.669 0.504 -27.357 40.876 Open in a new tab a. Dependent Variable: > 40% Threshold for non-food expenditure; *= statistically significant at p-value < 0.05 Discussion General description of the study This comparative cross-sectional study evaluated the effect of CBHI on out-of-pocket expenditure among diabetic patients attending Hawassa Comprehensive Specialized Hospital. The average monthly household expenditure was 6471.52 ETB (51 USD), with diabetes-related costs accounting for 2046.00 ETB (16.12 USD), with an exchange rate of 1 USD to 126.89 ETB in July 2024. This is greater than a finding from a study conducted in Bahir Dar, where diabetes care costs 382 ETB each month [ 13 ], and a study on chronic disease patients in Assela (555 ETB) [ 23 ]. This is comparable findings from a systematic review of the economic burden of diabetes mellitus involving 42 published articles that showed the annual cost per patient ranged from USD 87 to USD 9,581 [ 30 ]. A cross-sectional study conducted among patients with diabetes in Odisha/India showed 147.78 USD average monthly expenditure per patient [ 31 ]. However, this lower than finding from a study that evaluated the cost and utilization of healthcare services for persons with diabetes in the US showed that, in 2018, out-of-pocket expense were $2,037.2 USD and $1,543.3 USD for patients with type 1 and type 2 diabetes respectively [ 32 ]. The discrepancy between studies may be due to differences in data collection time, with changing healthcare costs over time. The fact that our study included participants in CBHI plans, but their study does not identify whether these individuals had insurance, may account for this discrepancy. Furthermore, patients may seek medical attention more frequently in Addis Ababa, which would raise the total cost of treatment. The majority of diabetes costs, 49.4% and 42.5% respectively, are attributed to indirect and direct medical expenses. On the other hand, direct non-medical expenses make up about 11.1% of overall expenses. This implies that both patients and their caregivers have experienced a significant loss of productivity as a result of diabetic sickness since diabetes is incurable and requires frequent visits to medical institutions, this has a significant impact on the finances of diabetic patients. Out-of-pocket expenses accounts for 42.5% of total medical expenses for diabetes care. The majority of direct medical expenses are attributed to medication and diagnostics (62.5% and 33%, respectively). This is comparable findings from a cross-sectional study conducted to determine out-of-pocket expenditure 206 patients with diabetes in Odisha/India showed that expenditure on medicine constituted around 65% of total medical expenditure followed by diagnostics services 13.2% and transportation (11.8%) [ 31 ]. Similarly, a study that evaluated the cost and utilization of healthcare services for persons with diabetes in the US showed that the largest proportion of OOP expenses were medication cists regardless of diabetes status [ 32 ]. However, this is significantly more than the results of a household survey in Ethiopia, which indicated that medicines and diagnostics make up roughly 45% and 16% of out-of-pocket medical expenses, respectively, and research on patients with chronic illnesses in Assela revealed that these expenses make up 41% and 14%, respectively [ 23 , 27 ]. This disparity may be caused by inflation, which affects the rising costs of diabetes care, as well as regional and socioeconomic factors like income and urbanization. This study indicated that, on average, 15% of out-of-pocket medical costs for diabetes are related to transportation. Similar findings have been reported by other studies, which show that a sizable amount of the direct costs of chronic illness are related to medication and transportation [ 23 ]. According to insurance status, this study looked at the primary categories of diabetes illness expenses and discovered that, for insured patients, direct and indirect costs make up 68% and 46% of total costs, respectively, whereas for uninsured patients, they make up 45% and 50%. This finding demonstrates how insurance coverage lessens the cost of medical bills, allowing insured patients to receive more frequent and thorough healthcare services. This increased utilization leads to higher direct medical costs for insured individuals. Conversely, because of the high out-of-pocket costs, uninsured people frequently put off or avoid seeking medical attention, which lowers their direct medical expenses. Delays or avoidance of care, however, may result in poorly managed or unmanaged diabetes, which raises indirect costs for uninsured patients by increasing productivity losses. The effect of community-based health insurance on out-of-pocket expenditure The findings in this study showed that 38% of OOP spending among insured households and 27.7% of the overall cost of diabetic disease are related to direct medical expenses. The largest portion of this direct out-of-pocket spending (28.8%) was related to drug costs, suggesting that people are paying for their medications out of pocket, whether or not they have insurance. Additionally, this survey revealed that 76% of insured families paid for pharmaceuticals and 22.1% for diagnostics (10.3% for laboratory and 11.8% for imaging). These figures are higher than those of a study on chronic patients at Aselle Referral Hospital [ 23 ], which discovered that 48% and 40% of chronic insured patients paid for pharmaceuticals and diagnostics, respectively. This significant disparity may result from a lack of supplies in public health facilities, where prescription drugs are frequently less expensive, forcing patients to seek treatment at private clinics, which charge more. In this study, the percentage of catastrophic health costs ranged from 27% to 83%. Our data showed that when 15% of non-food was used, the incidence of CHE among diabetic diseases was 71%. The findings are higher than those studies carried out in Bahir Dar (59.6%) and Dessie Referral Hospital (64.2%) [ 13 , 17 ]. This disparity can be explained by the fact that Hawassa Comprehensive Specialized Hospital serves as a referral center for underserved and rural areas, which contributes to higher out-of-pocket expenses for meals, lodging, and transportation. Additionally, patients treated by Hawassa Comprehensive Specialized Hospital may find diabetes care to be substantially more expensive due to regional economic disparities and variations in household income. This finding, on the other hand, is comparable to 74.4% seen in an Addis Ababa study involving cancer patients [ 33 ]. Nearly 83% of patients in this study spend more than 10% of their monthly non-food expenses. This is greater than the findings of a study conducted at Assela Referral Hospital, which showed that 54% of chronic patients spend more than 10% of their household income [ 23 ]. The disparity may result from differences in the economy across regions. Patients at Hawassa could come from households with lower incomes or greater levels of poverty, which makes paying for healthcare expenses more difficult. Additionally, it can be explained by disparities in coping strategies, since 32% of participants in the Assela study used community-based health insurance as their main source of funding, whereas only 27% of participants in our study had access to health insurance as a coping strategy to pay for their diabetic illness expenses. This study also demonstrates the severity of catastrophic health expenditures, such as overshoot and MPO, among diabetic patients. Out of all study participants, the average percentage of out-of-pocket medical expenses that exceeded 15% of non-food expenditure was 16%; among diabetic patients who had experienced CHE, the average percentage of out-of-pocket medical expenses that exceeded a specified threshold 15 was 23%. In a study of diabetic individuals in Bahir Dar, overshoot and MPO were reported to be 23% and 39%, respectively [ 13 ], which is nearly similar to our findings. Since patients with chronic diseases spend over 60% of their monthly non-food expenditures on medical care, their illness is likely causing their financial difficulty. At 15% of the non-food expenditure criteria, 57% of CBHI members and 84% of non-members experienced CHE. This is higher than studies at Asela Referral Hospital, where the incidence was 31% and 47%, respectively, and the CBHI Evaluation Study, where the incidence was only 7% and 19%, respectively [ 23 , 34 ]. The nature of diabetes diseases, which require frequent trips to medical facilities and incur extra expenses that insurance plans don’t necessarily cover, including transportation, may be the cause of this discrepancy. The severity of catastrophic health payments is also more intense for uninsured people than for insured people. The overshoot and mean positive overshoot for CBHI members were 15% and 26%, respectively, while they were 17% and 20% for non-members. This suggests that members having a lower overshoot (15% compared to 17% for non-members), CBHI lessens the severity of catastrophic health payments. The greater mean positive overshoot for members (26%) than non-members (20%) indicates that those who face catastrophic payments still experience significant financial challenges. This might be the result of members using healthcare services more frequently or gaps in CBHI coverage, especially for costly treatments. However, non-members may neglect their health care because of financial constraints, resulting in a lower MPO. To improve members’ and uninsured populations’ financial security, CBHI coverage must be expanded and improved. Compared to their counterparts, CBHI members generally have a reduced incidence of catastrophic health expenses. There is a strong correlation between CBHI enrollment and lower OOP expenditure. Patients who were enrolled in CBHI spent, on average, -499.41 ETB less on healthcare expenses than those who were not, according to the coefficient of -499.41 ( P < 0.000). This demonstrated how enrollment in CBHI reduces the likelihood of catastrophic medical costs. The main reason for this is that the majority of out-of-pocket medical expenses for treating diabetic diseases are related to medical services, particularly drugs. This finding is similar to research conducted in Assela and Northeast Ethiopia, where membership in CBHI reduced catastrophic out-of-pocket costs by 19% and 23%, respectively [ 23 , 35 ]. Additionally, out-of-pocket payments for medical services and the prevalence of catastrophic health expenditures are significantly impacted by health insurance, according to numerous studies conducted in Nigeria, India, and China [ 36 – 38 ]. However, another study conducted in China found that health insurance does not lessen the financial burden on people with chronic illnesses [ 39 ]. Differences in health insurance benefit packages may be the cause of this discrepancy. While the CBHI benefit package in Ethiopia includes both inpatient and outpatient services, the insurance benefit package in China only covers limited outpatient care and inpatient care [ 34 , 40 ]. Similarly, younger age patients were on average 8.756 ETB less in OOP expenses when compared to older age, with a coefficient of -8.756 ( p = 0.028). This could be due to age-associated increases in complications and increased healthcare needs of older patients. These could increase the OOP expense for patients than their younger counterparts. Similarly, employed participants were on average 58.22 ETB less in OOP expense when compared to their unemployed counterparts, with a coefficient of-58.221 ( p = 0.002). This could be due to the relatively stable income of employed patients, which could improve health-seeking behaviors like prevention and early detection, and treatment of complications. Participants with lower educational status were on average 104.416 ETB more in OOP expenses when compared to their counterparts, with a coefficient of 104.416 ( p = 0.004). Lower educational status could contribute to low health literacy and reduced health-seeking behaviors. Low health literacy contributes to delayed treatment seeking, poor medication adherence, poor glycemic control, and increased complications among diabetic patients [ 41 ]. However, demographic factors such as marital status, family size, and wealth percentile were not found to significantly affect OOP spending in this study, aligning with a study done in Ethiopia noted that limited demographic influence on out-of-pocket costs under CBHI schemes in Ethiopia [ 42 ]. It could be due to CBHI effectively reducing financial disparities in healthcare costs, providing equal protection across different demographic groups. However, there is no correlation between wealth and out-of-pocket costs ( p = 0.504) in this study. This is against evidence from different studies that revealed out-of-pocket spending severely burdens lower-income households, accounting for one-third of overall healthcare expenditures, according to a study conducted in Ethiopia [ 35 ]. According to research conducted in sub-Saharan Africa, OOP payments also predominate in healthcare finance, disproportionately affecting lower-income households and preventing universal health coverage [ 43 ]. The discrepancy could be due to a small number of participants in the upper wealth class (31 and 36 in the fourth and the highest quantile among insured and only 5 and 1 in the fourth and highest quantile among uninsured). This suggests CBHI premium costs are more affordable for wealthier households than the lowest or middle quantile participants. This requires designing strategies to increase the CBHI enrollment of the lowest and middle-wealth quantile patients. These strategies could include expanding subsidization or re-evaluating the affordability and willingness of society to pay for the CBHI program. Strengths and limitations of the study The inclusion of lost productivity due to the sickness and both direct (medical and non-medical) and indirect expenses is the strength of this study. In addition, using Kobo Collect, an electronic data collection tool, was used to decrease missing data and enhance data quality. However, this study has its limitations, including being a single public health facility-based study; it could not accurately reflect the experiences of diabetic patients in other regions of Ethiopia, especially in rural areas with varying socioeconomic circumstances and healthcare access. The cross-sectional design of the study makes it difficult to determine the causal linkages between variables like reduced financial hardship and CBHI enrollment. Participants’ recall bias to medical bills and indirect costs like lost productivity might have resulted in inaccurate information. Moreover, the study’s sample may only reflect a subset of outpatient health spending; inpatient health services may also have been utilized. Conclusion and recommendation A significant proportion of diabetes patients experienced catastrophic health expenditure (CHE). Enrollment in CBHI lowered OOP expenditure and reduced CHE. But there are still gaps in CBHI coverage, especially when it comes to the lowest and middle wealth quantiles. To enhance CBHI effectiveness, reduce financial burdens diabetes: Policy makers should prioritize rural and low-income households to reduce financial strain; introduce a subsidy program to make diabetes medications more affordable and focus particularly on uninsured households to alleviate financial barriers to essential treatment. Ethiopian Health Insurance Agency should focus on integrating direct and indirect cost coverage to strengthen financial protection; and incorporate beneficiary feedback to refine policies and address the diverse needs of enrollees. Researchers should evaluate the long-term impacts of CBHI on diabetic patient expenses on household financial stability. Inclusion of hospitals in rural and urban sectors and inclusion of the cost of complications of diabetes are important to better understand the effect of CBHI on OOP expense and catastrophic health expenditure in the region. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 (232.6KB, pdf) Acknowledgements We want to express my gratitude to the department of health service management at Yanet-Liyana College of Health Science for providing me with this wonderful opportunity. Additionally, we are also indebted to all the data collectors whose contributions were essential during the data collection process, as well as the study participants for their valuable input. We also would like to sincerely thank the HMIS, the outpatient department, and the staff at Hawassa Comprehensive Specialized Hospital for their gracious cooperation and readiness to provide data. Finally, we would want to express our gratitude to families and friends who helped during this project work. Abbreviations BMC Biomed central BMJ British medical Journal BSc Bachelor of science CA Cancer CBHI Community- based health insurance CHE Catastrophic health expenditure CHF Congestive heart failure CI confidence interval DM Diabetes mellitus ETB Ethiopian birr HCSP Hawassa Comprehensive specialized Hospital HMIS Health management and information system IDF International diabetes federation MPO Mean positive overshoot INR Indian Rupee LMICs Low- and middle-income countries MPH Master of public health OOPE Out-of-pocket expenditure USD United states Dollar WHO World Health Organization Author contributions All the authors read and approved the manuscript. BA conceived the research, framed the formatted the design, and conducted the data analysis; TA participated in the data analysis, reviewed the manuscript writing process, and polished the manuscript. MS participated in the data analysis, reviewed the manuscript writing process, polished the manuscript, and developed the manuscript for publication. The guarantor of the study is MS. The authors accept full responsibility for the finished work and/or the conduct of the study, have access to the data, and control the decision to publish. Funding None. Data availability All the data reported in the manuscript are publicly available upon the official request of the corresponding author upon acceptance of the manuscript. Declarations Ethical approval Ethical clearance was obtained from the Ethical Review Committee of Yanet Liyana Health Science College (Reference number: LHC/YLCHS/OGL/1568) on May 10, 2024. The research was approved in accordance with the World Medical Association (WMA) declaration of Helsinki ethical principles for medical research involving human participants. Please find our policy via the link below: https://www.wma.net/policies-post/wma-declaration-of-helsinki/ . Following the approval, a formal letter from the college was submitted to Hawassa Comprehensive Specialized Hospital, and permission to conduct the study was granted. Before the interviews, each participant received comprehensive information about the study's objectives and methodology, which was provided in an attached information sheet. Participants were assured of their right to privacy, anonymity, and respect throughout the study. Verbal and written consent were obtained from each participant to ensure voluntary participation. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. Lipoproteins LDLL-D, Model LMML-M, Network LMNLM, ICHOM International consortium for health outcomes measurement ICP Integrated Care Package ID Identification code IDF International diabetes federation IMA Intermutualistic Agency. Faculty of medicine and health sciences department of family medicine and population health 2025:9. 2. Ong KL, Stafford LK, McLaughlin SA, Boyko EJ, Vollset SE, Smith AE, Dalton BE, Duprey J, Cruz JA, Hagins H, et al. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2023;402(10397):203–34. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Bishu KG, Jenkins C, Yebyo HG, Atsbha M, Wubayehu T, Gebregziabher M. Diabetes in Ethiopia: A systematic review of prevalence, risk factors, complications, and cost. Obes Med. 2019;15:100132. [ Google Scholar ] 4. Seifu Y, Tsegaw D, Haji Y, Ejeso A. Prevalence and Associated Factors of Diabetes Mellitus Among Adult Population in Hawassa Zuria Woreda, Sidama Region, Ethiopia. Diabetes Metab Syndr Obes. 2020;13:4571–9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Belete GK, Sithole HL. Prevalence and associated factors of diabetes among adult populations of Hawassa town, southern Ethiopia: A community based cross-sectional study. PLoS ONE. 2025;20(1):e0318081. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Zhou B, Rayner AW, Gregg EW, Sheffer KE, Carrillo-Larco RM, Bennett JE, Shaw JE, Paciorek CJ, Singleton RK, Barradas Pires A, et al. Worldwide trends in diabetes prevalence and treatment from 1990 to 2022: a pooled analysis of 1108 population-representative studies with 141 million participants. Lancet. 2024;404(10467):2077–93. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. The Global Health Observatory. Out-of-pocket expenditure as percentage of current health expenditure (CHE) (%). https://www.who.int/data/gho/indicator-metadata-registry/imr-details/4965 . 8. Carrin G, Waelkens MP, Criel B. Community-based health insurance in developing countries: a study of its contribution to the performance of health financing systems. Trop Med Int Health. 2005;10(8):799–811. [ DOI ] [ PubMed ] [ Google Scholar ] 9. Odipo E, Jarhyan P, Nzinga J, Prabhakaran D, Aryal A, Clarke-Deelder E, Mohan S, Mosa M, Eshetu MK, Lewis TP, et al. The path to universal health coverage in five African and Asian countries: examining the association between insurance status and health-care use. Lancet Global Health. 2024;12(1):e123–33. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Group WB. Tracking universal health coverage: 2017 global monitoring report (English). Washington D.C; 2017;1. 11. Xu K, Evans DB, Carrin G, Aguilar-Rivera AM, Musgrove P, Evans T. Protecting households from catastrophic health spending. Health Aff (Millwood). 2007;26(4):972–83. [ DOI ] [ PubMed ] [ Google Scholar ] 12. Ethiopia-7th-Health-Accounts-Report_2016-17.pdf. 13. Tsega G, Getaneh G, Taddesse G. Are Ethiopian diabetic patients protected from financial hardship? PLoS ONE. 2021;16(1):e0245839. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Zawudie AB, Daka DW, Teshome D, Ergiba MS. Economic Burden of Diabetic Mellitus Among Patients on Follow-up Care in Hospitals of Southwest Shewa Zone, Central Ethiopia. BMC Health Serv Res. 2022;22(1):1398. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Shrestha R, Shakya A, Sapkota VP. Health Expenditure Among the Outpatient of Type 2 Diabetes in Selected Hospital of Kathmandu District: A Cross-Sectional Study. Adv Public Health. 2024;2024(1):5571393. [ Google Scholar ] 16. Laxy M, Zhang P, Benoit SR, Imperatore G, Cheng YJ, Gregg EW, Yang S, Shao H. Trends in total and out-of-pocket payments for insulin among privately insured U.S. Adults with diabetes From 2005 to 2018. Diabetes Care. 2021. [ DOI ] [ PubMed ] 17. Zinash A, Birhanu Demeke W, Solomon Ahmed M, Moges WA. Economic Assessment of direct cost of illness of diabetes mellitus at dessie referral hospital, North East Ethiopia. Int J Diabetes Clin Res 2020;7. 18. Alemayehu M, Addis B, Hagos T. Out-of-pocket health expenditure and associated factors among patients with hypertension in Debre-Tabor Comphrensive Specialized Hospital, South Gondar zone, Northwest Ethiopia, 2020. Front Public Health. 2023;11:1014364. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. World Health O. Primary health care on the road to universal health coverage: 2019 monitoring report: executive summary. 2019. 20. Bifato B, Ayele A, Rike M. Dangura D: 2020. 21. Mulat AK, Mao W, Bharali I, Balkew RB, Yamey G. Scaling up community-based health insurance in Ethiopia: a qualitative study of the benefits and challenges. BMC Health Serv Res. 2022;22(1):473. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Spaan E, Mathijssen J, Tromp N, McBain F, ten Have A, Baltussen R. The impact of health insurance in Africa and Asia: a systematic review. Bull World Health Organ. 2012;90(9):685–92. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Degefa MB, Woldehanna BT, Mebratie AD. Effect of community-based health insurance on catastrophic health expenditure among chronic disease patients in Asella referral hospital, Southeast Ethiopia: a comparative cross-sectional study. BMC Health Serv Res. 2023;23(1):188. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. World Health O. Guidance on global monitoring for diabetes prevention and control: framework, indicators and application. World Health Organization. 2024. 25. Woyesa SB, Hirigo AT, Wube TB. Hyperuricemia and metabolic syndrome in type 2 diabetes mellitus patients at Hawassa university comprehensive specialized hospital, South West Ethiopia. BMC Endocr disorders. 2017;17:1–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Shumet Y, Mohammed SA, Kahissay MH, Demeke B. Catastrophic Health Expenditure among Chronic Patients Attending Dessie Referral Hospital, Northeast Ethiopia. Clinicoecon Outcomes Res. 2021;13:99–107. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Kiros M, Dessie E, Jbaily A, Tolla MT, Johansson KA, Norheim OF, Memirie ST, Verguet S. The burden of household out-of-pocket health expenditures in Ethiopia: estimates from a nationally representative survey (2015–16). Health Policy Plann. 2020;35(8):1003–10. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Sorato MM, Davari M, Kebriaeezadeh A, Sarrafzadegan N, Shibru T. Societal economic burden of hypertension at selected hospitals in southern Ethiopia: a patient-level analysis. BMJ Open. 2022;12(4):e056627. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Massito J, Hinju G. Out of pocket and catastrophic health expenditure in Tanzania: recent evidence on the incidence, intensity and distribution. BMC Health Serv Res. 2025;25(1):677. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Butt MD, Ong SC, Rafiq A, Kalam MN, Sajjad A, Abdullah M, Malik T, Yaseen F, Babar ZU. A systematic review of the economic burden of diabetes mellitus: contrasting perspectives from high and low middle-income countries. J Pharm Policy Pract. 2024;17(1):2322107. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Swain SP, Samal S, Sahu KS, Rout SK. Out-of-pocket expenditure and drug adherence of patients with diabetes in Odisha. J Family Med Prim Care. 2018;7(6):1229–35. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Reynolds EL, Mizokami-Stout K, Putnam NM, Banerjee M, Albright D, Ang L, Lee J, Pop-Busui R, Feldman EL, Callaghan BC. Cost and utilization of healthcare services for persons with diabetes. Diabetes Res Clin Pract. 2023;205:110983. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Kasahun GG, Gebretekle GB, Hailemichael Y, Woldemariam AA, Fenta TG. Catastrophic healthcare expenditure and coping strategies among patients attending cancer treatment services in Addis Ababa, Ethiopia. BMC Public Health 2020;22. [ DOI ] [ PMC free article ] [ PubMed ] 34. Evaluation of community-based health insurance pilot. Schemes in Ethiopia: Final report. 2015;5. 35. Mekonen AM, Gebregziabher MG, Teferra AS. The effect of community based health insurance on catastrophic health expenditure in Northeast Ethiopia: a cross sectional study. PLoS ONE 2018;13. [ DOI ] [ PMC free article ] [ PubMed ] 36. Habib SS, Perveen S, Khuwaja HMA. The role of micro health insurance in providing financial risk protection in developing countries- a systematic review. BMC Public Health 2016;22. [ DOI ] [ PMC free article ] [ PubMed ] 37. Li A, Shi Y, Yang X, Wang Z. Effect of critical illness insurance on household catastrophic health expenditure: the latest evidence from the national health service survey in China. IJERPH. 2019;16. [ DOI ] [ PMC free article ] [ PubMed ] 38. of national health insurance, nigeria.pdf. 39. Fu XZ, Sun QW, Sun CQ, Xu F, He JJ. Urban-rural differences in catastrophic health expenditure among households with chronic non-communicable disease patients: evidence from China family panel studies. BMC Public Health 2021;21. [ DOI ] [ PMC free article ] [ PubMed ] 40. Su M, Zhou Z, Si Y, Wei X, Xu Y, Fan X. Comparing the effects of China’s three basic health insurance schemes on the equity of health-related quality of life: using the method of coarsened exact matching. Health Qual Life Outcomes 2018;16. [ DOI ] [ PMC free article ] [ PubMed ] 41. Cravo M, Rosendo I, Santiago LM, Abreu J. Health Literacy and Complications in People With Type 2 Diabetes: An Exploratory Study. Cureus. 2023;15(9):e46064. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 42. Admasu BA, Gebremariam MY. Assessment of Health Insurance’s Role in Reducing Financial Barriers to Health Service Access: Perspective from a Resource Limited Setting. Inquiry. 2025;62:00469580251333636. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Eze P, Ilechukwu S, Lawani LO. Impact of community-based health insurance in low- and middle-income countries: A systematic review and meta-analysis. PLoS ONE. 2023;18(6):e0287600. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Material 1 (232.6KB, pdf) Data Availability Statement All the data reported in the manuscript are publicly available upon the official request of the corresponding author upon acceptance of the manuscript. Articles from BMC Health Services Research are provided here courtesy of BMC ACTIONS View on publisher site PDF (2.5 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top