Prediction of health care expenditure increase: how does pharmacotherapy contribute?

Study on the concentration, distribution, and persistence of health spending for the contributory scheme in Colombia

Colombia is among the countries with the most robust financial protection against personal health spending in the world, with out-of-pocket spending ranking lowest across OECD countries. We investigate the evolution, distribution, and persistence of health spending by age group, sex, health care setting, health condition and geographic region for over 19 million users of Colombia's health system between 2013 and 2021 (contributory scheme). We use average patient-level expenditure data from the Health-Promoting Entities of the Ministry of Health and Social Protection. We applied multivariate statistical techniques such as multiple correspondence analysis, factor maps and correlations. For both sexes, average health expenditure increases gradually with age until 60 years, accelerating thereafter abruptly. Health conditions with the highest percentage of expenditure were those related to neoplasms, blood diseases, circulatory system, pregnancy, puerperium and perinatal period. We found that home-based care in Amazonía-Orinoquía is almost non-existent, and that outpatient care represents a high proportion in all age groups (over 65%) compared to the other regions. There is a strong persistence of expenditure from one year to the next (i.e. they can provide relevant information for prediction), especially in areas with a larger supply of health services such as Bogotá-Cundinamarca. To the authors' knowledge, this is the most comprehensive and detailed micro-analysis of health spending that has been developed for a Latin American country to date.

Identifying Influential Variables on Health Expenditure of the Organisation for Economic Co-Operation and Development (OECD) Countries

Background

Health expenditures of countries have an increasing trend in general and identifying variables affecting health expenditure is an important step toward budget planning for financial sustainability. This study aimed to examine the health expenditure of the Organisation for Economic Co-operation and Development (OECD) countries and identify influential variables.

Methods

The data for the years 2000-2018 of OECD countries' current health expenditure (% of GDP) and economic, demographic, and health variables, considered to affect the health expenditure, to include in the analysis were extracted using the World Bank database (World Bank 2021). Data analys using Chi-Squared Automatic Interaction Detection (CHAID) decision tree technique. Fifteen variables in economic, demographic, and health categories are selected to build the CHAID decision tree.

Results

As a result of CHAID analysis, five variables are identified as influential on current health expenditure, which are gross domestic product per capita, life expectancy at birth, death rate, out-of-pocket expenditure, and fertility rate. Thirty-seven OECD countries are classified into eleven groups by the decision rules in terms of the current health expenditure. The high value of the correlation coefficient between the predicted values and the actual values of health expenditure of countries indicates good prediction performance. Moreover, the regression models built using the identified influential variables as explanatory variables give good forecast accuracy.

Conclusion

As an effective tool, the CHAID decision tree technique provides a rule-based model in the form of a tree with nodes and branches, illustrating the splitting process graphically with identified variables and their cut-off points for classification and prediction.

Machine learning in health financing: benefits, risks and regulatory needs

There is increasing use of machine learning for the health financing functions (revenue raising, pooling and purchasing), yet evidence lacks for its effects on the universal health coverage (UHC) objectives. This paper provides a synopsis of the use cases of machine learning and their potential benefits and risks. The assessment reveals that the various use cases of machine learning for health financing have the potential to affect all the UHC intermediate objectives - the equitable distribution of resources (both positively and negatively); efficiency (primarily positively); and transparency (both positively and negatively). There are also both positive and negative effects on all three UHC final goals, that is, utilization of health services in line with need, financial protection and quality care. When the use of machine learning facilitates or simplifies health financing tasks that are counterproductive to UHC objectives, there are various risks - for instance risk selection, cost reductions at the expense of quality care, reduced financial protection or over-surveillance. Whether the effects of using machine learning are positive or negative depends on how and for which purpose the technology is applied. Therefore, specific health financing guidance and regulations, particularly for (voluntary) health insurance, are needed. To inform the development of specific health financing guidance and regulation, we propose several key policy and research questions. To gain a better understanding of how machine learning affects health financing for UHC objectives, more systematic and rigorous research should accompany the application of machine learning.

The application of artificial intelligence in health financing: a scoping review

INTRODUCTION

Artificial Intelligence (AI) represents a significant advancement in technology, and it is crucial for policymakers to incorporate AI thinking into policies and to fully explore, analyze and utilize massive data and conduct AI-related policies. AI has the potential to optimize healthcare financing systems. This study provides an overview of the AI application domains in healthcare financing.

METHOD

We conducted a scoping review in six steps: formulating research questions, identifying relevant studies by conducting a comprehensive literature search using appropriate keywords, screening titles and abstracts for relevance, reviewing full texts of relevant articles, charting extracted data, and compiling and summarizing findings. Specifically, the research question sought to identify the applications of artificial intelligence in health financing supported by the published literature and explore potential future applications. PubMed, Scopus, and Web of Science databases were searched between 2000 and 2023.

RESULTS

We discovered that AI has a significant impact on various aspects of health financing, such as governance, revenue raising, pooling, and strategic purchasing. We provide evidence-based recommendations for establishing and improving the health financing system based on AI.

CONCLUSIONS

To ensure that vulnerable groups face minimum challenges and benefit from improved health financing, we urge national and international institutions worldwide to use and adopt AI tools and applications.

Predicting the Annual Funding for Public Hospitals with Regression Analysis on Hospital’s Operating Costs: Evidence from the Greek Public Sector

The funding of public hospitals is an issue that has been of great concern to health systems in the past decades. Public hospitals are owned and fully funded by the government, providing in most countries medical care to patients free of charge, covering expenses and wages by government reimbursement. Several studies in different countries have attempted to investigate the potential role and contribution of hospital and clinical data to their overall financial requirements. Many of them have suggested the necessity of implementing DRGs (Diagnosis Related Groups) and activity-based funding, whereas others identify flaws and difficulties with these methods. What was attempted in this study is to find an alternative way of estimating the necessary fundings for public hospitals, regardless the case mix managed by each of them, based on their characteristics (size, specialty, location, intensive care units, number of employees, etc.) and its annual output (patients, days of hospitalization, number of surgeries, laboratory tests, etc.). We used financial and operational data from 121 public hospitals in Greece for a 2-years period (2018–2019) and evaluated with regression analysis the contribution of descriptive and operational data in the total operational cost. Since we had repeated measures from the same hospitals over the years, we used methods suitable for longitudinal data analysis and developed a model for calculating annual operational costs with an R²≈0.95. The main conclusion is that the type of hospital in combination with the number of beds, the existence of an intensive care unit, the number of employees, the total number of inpatients, their days of hospitalization and the total number of laboratory tests are the key factors that determine the hospital’s operating costs. The significant implication of this model that emerged from this study is its potential to form the basis for a national system of economic evaluation of public hospitals and allocation of national resources for public health.

Potentially inappropriate medications and medication combinations before, during and after hospitalizations: an analysis of pathways and determinants in the Swiss healthcare setting

Background

A hospitalization phase represents a challenge to medication safety especially for multimorbid patients as acute medical needs might interact with pre-existing medications or evoke adverse drug effects. This project aimed to examine the prevalence and risk factors of potentially inappropriate medications (PIMs) and medication combinations (PIMCs) in the context of hospitalizations.

Methods

Analyses are based on claims data of patients (≥65 years) with basic mandatory health insurance at the Helsana Group, and on data from the Hirslanden Swiss Hospital Group. We assessed PIMs and PIMCs of patients who were hospitalized in 2013 at three different time points (quarter prior, during, after hospitalization). PIMs were identified using the PRISCUS list, whereas PIMCs were derived from compendium.ch. Zero-inflated Poisson regression models were applied to determine risk factors of PIMs and PIMCs.

Results

Throughout the observation period, more than 80% of patients had at least one PIM, ranging from 49.7% in the pre-hospitalization, 53.6% in the hospitalization to 48.2% in the post-hospitalization period. PIMCs were found in 46.6% of patients prior to hospitalization, in 21.3% during hospitalization, and in 25.0% of patients after discharge. Additional medication prescriptions compared to the preceding period and increasing age were the main risk factors, whereas managed care was associated with a decrease in PIMs and PIMCs.

Conclusion

We conclude that a patient’s hospitalization offers the possibility to increase medication safety. Nevertheless, the prevalence of PIMs and PIMCs is relatively high in the study population. Therefore, our results indicate a need for interventions to increase medication safety in the Swiss healthcare setting.

Total Health Expenditure and Its Driving Factors in China: A Gray Theory Analysis

The continuous growth in total health expenditure (THE) has become a social issue of common concern in most countries. In China, the total health expenditure (THE) is maintaining a rapid growth trend that is higher than that of the economy, which has become increasingly obvious in the 21st century and has brought a heavy burden to the government and residents. To analyze the main driving factors of THE in China in the 21st century and establish a predictive model, gray system theory was employed to explore the correlation degree between THE and nine hot topics in the areas of the economy, population, health service utilization, and policy using national data from 2000 to 2018. Additionally, a New Structure of the Multivariate Gray Prediction Model of THE was established and compared with the traditional grey model and widely used BP neural network to evaluate the prediction effectiveness of the model. We concluded that the Chinese government and society have played a crucial role in reducing residents’ medical burden. Besides this, the improved economy and aging population have increased the demand for health services, leading to the continual increase in THE. Lastly, the improved NSGM(1,N) model achieved good prediction accuracy and has unique advantages in simulating and predicting THE, which can provide a basis for policy formulation.