Expenditure and use on radiology imaging in urban and rural areas in a developing country Authors: Oscar Espinosa, Gabriela Puentes, Valeria Bejarano, Javier Romero, Martha-Liliana Arias, Sandeep Hedgire, Dania Daye, Suyash Mohan, Richard Duszak Published: 2025 ABSTRACT AbstractBackgroundChanges in the population pyramid and in the disease burden of patients can be observed through the expenditure and frequency of use of medical imaging in health systems; however, this topic has undergone little study in the international literature. Our objective was to analyze the per capita and total expenditure and the frequency of use of radiological imaging according to gender, geographic area and type of medical imaging in the Colombian population.MethodsIn this work, a national database was used to study temporal dynamics of annual expenditures and frequency of use on medical imaging of more than 19 million people (0-100 years old). Descriptive analytical techniques and statistical visualization were used for the development of the research.ResultsPer capita expenditure increased progressively with age and for almost all ages it was found to be higher in men than in women. In addition, rural areas had a higher per capita expenditure on medical imaging than urban areas. However, when analyzed by total expenditures, women and urban areas had the highest participation. Interventional radiology and magnetic resonance imaging procedures accounted for the largest share of total medical imaging spending. General radiology medical imaging has the highest frequency of use per capita and urban women are the profile with the highest use of these medical services.ConclusionOur findings enable further characterization of the evolution of the financial burdens of medical imaging which we anticipate increasing with future aging of the population.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12962-025-00681-1.Keywords: Medical imaging, Health spending, Frequency of use, Developing countries Introduction The United States has been a pioneer in studies on medical imaging spending, but it remains debated whether or not these medical procedures drive health care spending [1–7]. Demographic transitions and changes in disease burden imply a change in the frequency of use of health services and consequently in the utilization of medical imaging across age groups (e.g. in health conditions such as neoplasms, maternal-perinatal, abdominal, neurological, and musculoskeletal diseases, etc.). The characterization of such data is especially relevant in developing countries where it can provide evidence to promote further investment and development in this topic. In 2023, according to the Oxford Poverty and Human Development Initiative, 6.1 billion people lived in developing countries [8]. A key challenge for these nations is the limited health budget, which affects the acquisition of radiology equipment, the infrastructure and the workforce needed for implementation. This is relevant given that diagnostic imaging is a fundamental tool for the diagnosis and follow-up of communicable and non-communicable diseases. Their timely use allows for appropriate diagnoses and interventions that positively impact the quality of life of the populations [9]. According to the National Administrative Department of Statistics, the Colombian population is expected to grow by 10.5% between 2023 and 2050, from 52,215,503 to 57,713,467 in the middle of the century, with increases in the proportion of people over 60 years of age (from 14.6% to 25.9%) [10]. These changes could imply an exponential requirement in the expenditure on diagnostic imaging, especially in rural areas where equipment, infrastructure and trained personnel are scarce. Understanding spending trends can inform future investments in health systems. In this context, radiology is not only an essential diagnostic tool but also a cornerstone in clinical management and therapeutic decision-making, particularly in high-burden conditions such as cancer, cardiovascular diseases, and trauma [11, 12]. International literature has shown that the underutilization or delayed access to imaging services can lead to diagnostic delays, inappropriate or unnecessary treatments, and worse clinical outcomes, especially in rural areas and underserved communities [13–15]. Moreover, analyzing medical imaging expenditures provides insight into resource efficiency, helps identify disparities in service distribution, and supports the development of evidence-based health policies. In middle-income countries such as Colombia, where health resources are constrained and geographic inequalities are pronounced, characterizing the patterns of imaging use and spending is a key step toward promoting equity and ensuring the sustainability of the healthcare system. Our objective was to analyze the variation in spending by age, gender, geographical area of the patient and type of medical imaging in Colombia, as well as the frequency of use of the health service, to better understand recent trends and distribution of spending and use on medical imaging in populations1. Materials and methods This study did not require the approval of an Ethics Committee since it did not constitute research with human subjects. The database used was Gestión de la Demanda, which registers all claims of health insurers (called Health-Promoting Entities in Colombia) of the contributory scheme, corresponding to formally employed individuals (and their dependents), who actively contribute to the health care system in proportion to their salaries. This database is freely accessible after registration in the information system of the Ministry of Health and Social Protection. This database contains disaggregated information by gender, single-year age, municipality, and other individual features, for millions of patients who received health technologies financed under the contributory health insurance scheme between 2013 and 2021. Regarding data quality, the Colombian Ministry of Health and Social Protection applies multiple validation layers throughout the data collection and processing stages, ensuring high standards of reliability and accuracy. After rigorous data cleaning procedures, the dataset labeled Gestión de la Demanda captures over 80% of patients enrolled in the contributory scheme. For categorization of urban or rural area (where imaging was done), we use the definition constructed by the National Planning Department, taking as urban area the municipalities classified as ‘cities and agglomerations’ and ‘intermediate’, and as rural area the municipalities classified as ‘rural’ and ‘dispersed rural’ [16]. Table S1 of the Supplement shows the respective allocation for the more than 1,100 municipalities in the country. We identified claims related to radiological imaging using the Unified Classification of Health Procedures (CUPS, for its acronym in Spanish) of Colombia. Specifically, codes starting with 870, 871, 872, 873 (general radiology), 874, 875, 876, 877, 878 (special and interventional radiology), 879 (computed tomography), 881 (ultrasound), 882 (noninvasive vascular studies), 883 (magnetic resonance imaging), 886, and 887 (other imaging techniques and studies) were taken [17]. Table S2 of the Supplement shows the breakdown of all selected health procedures (6-digit CUPS codes). Our research adopts an analytical-descriptive approach complemented by statistical visualization techniques based on Tufte’s principles. Then, to find the per capita expenditure on medical imaging, we calculated the total expenditure on imaging as the sum of all the monetary values paid by these insurers to different health care providers and divided it by the total number of people to whom these health care services were provided. In addition, to review the weight of radiological imaging on the total monetary value of medical care, the total expenditure of all health services and technologies was calculated, and then the percentage invested in medical imaging was estimated, according to age, gender and geographic area. To calculate the per capita frequency of use of radiological procedures, the total number of medical images was taken and divided by the total number of people to whom these health services were provided. When analyzing monetary values for different years, to standardize this metric, we converted Colombian pesos to purchasing power parity (PPP) US dollars at 2017 prices using the index calculated by the World Bank [18]2. The programming language used was R, version 4.3.2 for statistical analysis. Results A total of nine years were studied, from 2013 to 2021 (publicly available information), including more than 19 million people on average, from 0 to 100 years, with representation of more than 99.5% of the geographic territory. Table 1 presents the per capita expenditures and their standard deviation, for the different life cycles -defined by the Ministry of Health and Social Protection [20]- and rural and urban areas. Table 1Average annual per capita spending on medical imaging by life cycle and geographic area, 2013–2021 (US dollars PPP)Life cycleRural areaUrban areaNational totalInfancy (0–11 years)114.4 [21.7]79.3 [4.8]81.0 [5.2]Adolescence-Youth (12–26 years)115.5 [19.3]92.1 [7.7]93.4 [7.9]Adulthood (27–59 years)145.2 [22.7]126.9 [11.6]127.6 [11.6]Old age (60 years and over)229.7 [33.9]207.6 [25.9]208.6 [25.5]Note: standard deviation in square brackets Average annual per capita spending on medical imaging by life cycle and geographic area, 2013–2021 (US dollars PPP) Note: standard deviation in square brackets Figure 1A shows the annual per capita expenditure on medical imaging per year by age, gender and geographic area for the different years of analysis. A constant increase is observed as age increases, for both men and women, being higher in the case of rural areas3. The age with the highest per capita expenditure is between 78 and 86 years, with values reaching US$ 466.5 PPP. In addition, spending on medical imaging is generally higher as years go by for all ages, until about 80 years of age, when it starts to decrease again. (A) Annual per capita expenditure on medical imaging by age, gender and geographic area (PPP US dollars), (B) Distribution of total expenditure on medical imaging by gender and geographic area Figures S1 to S7 of the Supplement illustrate the per capita expenditure for the types of radiological imaging, showing that the highest monetary value corresponded to the categories of special and interventional radiology, and magnetic resonance imaging. On the other hand, when we analyze the total expenditure on medical imaging, we find a higher share of urban areas and of the female gender, which means that there is a greater use of these services by women (even though their per capita expenditure is less than that of men). Expenditure on medical imaging as a percentage of total health care expenditure by age is shown in Fig. 2. We observe that the lowest proportion is found in newborns and people older than 90 years of age, both in rural and urban areas. The highest percentage of total expenditures by age on imaging was between the ages of 5 and 14 in women, and between the ages of 7 and 18 in men, regardless of the geographic area of health care delivery. In addition, the proportion of total expenditure on radiological imaging was higher in women aged 3 to 14 years than in men of the same age. Also, for all ages the proportion is higher in rural regions reaching an average of 53.4% spent on medical imaging. Figures S8 to S11 of the Supplement presents the proportion of the expenditure of each of the predefined groups of radiological imaging over the per capita expenditure of imaging, showing that the largest share among the years refers to special and interventional radiology for urban males, followed by magnetic resonance imaging for rural females. Percentage of total per capita expenditure on medical imaging services by age, gender and geographic area The evolution of the frequency of use on radiological imaging per capita is shown in Fig. 3, indicating a higher utilization per capita with increasing age. On the other hand, no increase in per capita use of medical imaging is observed over the years. Throughout the analysis period, urban females had the highest per capita health service use. Figures S12 to S18 present the frequency of use by type of medical imaging, highlighting general radiology and computed tomography as the most frequent and other imaging techniques and interventional radiology as the least frequent. The types of medical imaging in which there is no marked increase in use according to age are special and interventional radiology, magnetic resonance imaging and other imaging techniques and studies. In the case of ultrasound, as expected, there is a significant increase in the female reproductive period. Annual frequency of use on medical imaging services by age, gender and geographic area Discussion In our analysis, we describe the per capita expenditure on medical imaging by age, gender, geographic area and type of imaging in the population enrolled to the contributory scheme in Colombia. Per capita expenditures on radiological imaging were positively correlated with age (Pearson correlation of 0.91), with people between 75 and 89 years old having the highest monetary value. Men had a higher per capita expenditure on medical imaging than women, both in rural and urban areas, but a lower share of total spending, due to the higher utilization of these medical services by women. Women contribute substantially to the overall spending; this could be explained by the higher number of diagnostic screenings used during the different life cycles (for example, in the perinatal period). This underscores the importance of sustaining and expanding availability of diagnostic imaging for screening programs (such as mammograms), emphasizing their pivotal role in reinforcing primary prevention efforts. Our results regarding the distribution of total expenditure are similar to those found for the U.S. patient population4 [6, 21]. When examining procedural spending, the significant increase in investment in interventional radiology among urban males and in magnetic resonance imaging among rural females is striking. These results highlight the importance of health professionals and policy makers having a good understanding of the scope and indications for interventional radiology. It is crucial that they are familiar with this subspecialty to understand its indications, its potential scope and the positive impact it can have on patients. In rural areas, despite experiencing higher per capita spending, challenges persist due to the scarcity of high-complexity medical procedures, coupled with a shortage of specialists and supra-specialists (which increases their price). This creates a scenario where rural regions grapple with high-cost health technology and underscores the need for strategic healthcare resource allocation to address these disparities effectively. This apparent inconsistency -higher per capita spending in rural areas despite fewer available resources- can be explained by the way Colombia’s health financing system is structured. Specifically, the Capitation Payment Unit (CPU), which defines the health insurance premium, is higher for rural regions. This is because delivering healthcare in these areas often entails greater logistical challenges and limited availability of high-complexity services. As a result, when such services are required, their cost tends to be significantly higher due to factors like patient transportation, outsourcing, or the lack of economies of scale. This structural adjustment in the CPU is intended to compensate for the higher cost of service delivery in rural contexts but also contributes to elevated per capita expenditures. These findings should be interpreted within the context of the contributory population, acknowledging that the majority of rural residents (who are enrolled in the subsidized scheme) may face substantially greater barriers to accessing diagnostic imaging, especially advanced modalities. Our data highlights the importance of increasing investment to improve hospital infrastructure and the acquisition of diagnostic imaging equipment, expenses that, although they may seem high in the short term, according to previously expected projections of population growth, could be cost-effective in the long term [22], and could substantially improve the health of populations with a positive impact on both primary and secondary prevention. Moreover, although our study does not show a significant increase in the frequency of use of medical imaging, it is also true that its use during the COVID-19 pandemic period was not negatively affected. This demonstrates, once again, the importance of radiology departments as key elements in the detection and treatment of pathologies for the entire population, regardless of age, gender or area of residence. Its constant technical, workforce and financial strengthening is a must in any public or private clinic, as well as in any health system that wants to advance in the promotion of health equity and efficiency. Our research work has some limitations. We were unable to study segmentations by ethnic-racial group, income level, among other variables, since the database analyzed did not include such sociodemographic details. In addition, our studies are not generalizable to people enrolled to the subsidized and special and/or exception schemes, given that their prevalence, incidences and uses of health services are different [23]. The databases for these health insurance schemes are not publicly available due to the poor quality of their information, which further limits the possibility of conducting robust analyses on these population groups. Finally, our findings suggest that due to the future aging of the Colombian population (irreversible change in the population pyramid), healthcare investment in medical imaging is destined to increase significantly. Radiology being one of the medical specialties where artificial intelligence has been implemented the most, the future impact of this technological tool and its effects on the cost and complexity of radiological services is yet to evolve especially in developing countries with limited public health budgets [22]. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements None. Author contributions OE, GP: conception and design of the work, data collection, data analysis and interpretation, drafting the article, critical revision of the article, final approval of the version to be submitted. VB: data collection, data analysis and interpretation, drafting the article, critical revision of the article, final approval of the version to be submitted. JR, MA, SH, DD, SM, RD: drafting the article, critical revision of the article, final approval of the version to be submitted. Funding None. Data availability The authors declare that they had full access to all of the data in this study and the authors take complete responsibility for the integrity of the data and the accuracy of the data analysis. The data and programming code can be found at https://github.com/vbejaranos/Radiology_Exp_Col. Declarations Ethics approval and consent to participate Not applicable. Consent for publication All authors agree with the content of this manuscript. Competing interests The authors declare no competing interests. Footnotes In Colombia, health insurers receive an annual premium from the government for each member, depending on their geographic region, sex and age. In the field of economic measurement, PPP is defined as currency conversion rates that equalize the purchasing power of different currencies (eliminating differences in price levels between nations) [19]. Given that individuals enrolled in the contributory scheme comprise only a subset of Colombia’s rural population, the rural estimates reported here pertain exclusively to this group and should not be generalized to the country’s overall rural population. No such studies have been conducted for other countries. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The authors declare that they had full access to all of the data in this study and the authors take complete responsibility for the integrity of the data and the accuracy of the data analysis. The data and programming code can be found at https://github.com/vbejaranos/Radiology_Exp_Col.