Study analyzes more than 100,000 health insurance-related lawsuits

“Patient forced to pay for surgery to be compensated by health insurer,” “Court orders health insurer to pay for woman’s egg freezing,” “Superior Court of Appeals upholds age group-based increases in health insurance premiums.” The main thing these headlines have in common is disagreements between health insurers and their customers, which often lead to lawsuits. This type of legal conflict has become increasingly common in Brazil, especially because of some changes in regulations made by the National Health Insurance Regulatory Agency (ANS). However, researchers at Fundação Getulio Vargas’ Sao Paulo Law School can make a strong contribution to this debate by using innovative tools to help understand this kind of conflict.

An applied scientific research project called “Litigation in the Private Health Insurance System: An Empirical Analysis Using Big Data and Artificial Intelligence” will use these two technological tools to examine decisions by appeal courts over the last four years in the states of Rio de Janeiro, Sao Paulo and Minas Gerais, which account for the bulk of lawsuits against health insurers in Brazil. In all, more than 100,000 court sentences will be compiled in a database and then analyzed. This may improve our understanding of how these court decisions are made. The three states in question account for 55% of Brazil’s private health insurance customers.

Researcher Daniel Wang, who is leading the study, explains the most common types of legal conflicts related to health insurance in the state of Sao Paulo: “The data indicates that claim denials are the main cause of litigation involving health insurers, making up more than half of all decisions at trial court and appeal court levels. Disputes related to increases in premiums and maintenance of contracts represent a significantly lower proportion of cases, although the absolute numbers are still large,” he says. 

Impact

According to Wang, the research project is important because approximately 50 million people depend on the private health insurance system in Brazil and the number of lawsuits has been growing steadily. So, it is necessary to understand how the courts tend to make decisions in response to different demands.

“There are relatively few academic studies about the private health insurance system, especially if we consider the sector’s size in Brazil and the huge amount of research focused on lawsuits against the public health system. This study will use big data and artificial intelligence in an attempt to overcome these difficulties and present the first ever comprehensive and detailed panorama of litigation in Brazil’s private health insurance system. We will also complement the literature on the relationship between the judiciary and regulatory agencies. Furthermore, we will look at experience in other countries, especially those where there are similar demands against health insurers, such as the Netherlands and Switzerland,” says Wang.

According to him, compiling and highlighting this information may help prevent future legal conflicts between health insurers and customers. He notes that disagreements often happen due to a lack of knowledge about litigation in this area, giving rise to the following questions: What are the most debated topics in these legal proceedings? How does the justice system understand customers’ demands? And how do these court sentences impact the regulation of private health insurance services?

Existing research on these subjects has various limitations, largely due to difficulty in gathering data on litigation in the sector. “On the one hand, the private health insurance system is very important to the population, and on the other hand, there are many lawsuits filed against it,” explains Wang.

Database

Through deep quantitative analysis, data can help us understand not just the rights and duties of health insurers, but also those of customers. In order to achieve this objective, the research project is making pioneering use of two tools with strong innovative potential: artificial intelligence, to help detect court decisions regarding the private health insurance system and categorize them; and big data, allowing large databases to be analyzed.

Wang explains that his research project is managing to go into detail in this area in an unprecedented way. “Through AI and big data, we managed to create a robot capable of searching for and correctly collecting the court decisions that interest us for this study. As well as collecting them, this robot is also able to perform certain classifications about these rulings. This is because in addition to collecting all decisions at trial court and appeal court levels, it will also group them by similarity. This will give us greater knowledge about different categories of decisions,” he says.

Court decisions are being put into three main categories. First, there are “claim denials,” referring to refusals by health insurers to pay for a particular medical service or to pay a particular clinic. The second category is “increases in monthly premiums,” covering annual increases and also increases when customers move into an older age group, for example. Finally, there is the “contract maintenance” category, which includes cases of nonpayment and contract termination, which may happen when a customer leaves a certain employer, for example. According to Wang, there is also a fourth category, which accounts for many lawsuits, called “other.” However, the first three categories give a good general picture of the types of conflicts that tend to exist between health insurers and their customers.

Innovation

The project, which is due for completion in the second half of 2024, employs cutting-edge methodologies, such as machine learning, big data and network theory. As well as contributing to knowledge about this type of litigation focused on the health insurance market, it is expected that the study will contribute to the development of research methodologies through the creation of technological assets and quantitative research techniques. This may help the research community use these techniques to answer numerous other questions that affect society, not only at FGV’s schools but at many other institutions.

Wang emphasizes that the idea of ​​using these computing tools is aimed at not just boosting the capacity for analysis but also generating rapid replicability. Following the development of the robot and its “training” to research and collect data correctly, it will be easier to apply it in the same way to other states, with minor adjustments related to regional characteristics. This replicability will save time and money, by taking advantage of the same programming when researching court decisions related to the private health insurance system in different periods and regions.

Products

To ensure that the information found has a real impact, reports featuring descriptive and inferential statistics will be produced and presented to stakeholders, including regulators, health insurers, members of the judiciary, civil society, consumer protection agencies and the general public. These reports will be written as the study progresses and they will be delivered along with other products that aim to increase the applicability and impact of this research, such as events with key sector stakeholders, press releases for the mainstream media and discussions with researchers and professionals in the field to present and analyze the research findings. Scientific articles and seminars are also planned, in order to share the methods and results of this research with other researchers and institutions in Brazil and worldwide.

Support and funding

This study is financed by Fundação Getulio Vargas’ Applied Research Fund (FPA FGV), which since 2014 has supported more than 251 projects selected by FGV’s Research and Innovation Commission. In addition to FGV, the Brazilian Private Health Insurance Association (FenaSaúde), which represents 14 Brazilian health insurers, is also funding this research, allowing the project to be expanded and to research even more databases. As well as the FGV Sao Paulo Law School, the FGV Center for Health Planning and Management Studies (FGVsaúde) at the Sao Paulo School of Business Administration (FGV EAESP) is also supporting the study.

Applied scientific research

Applied scientific research is the main focus of the FGV Applied Research and Knowledge Network, created in 2016 with the aim of promoting excellent, high-impact and innovative research in different areas of society. By proposing different processes and solutions to tackle the challenges faced by Brazil’s people, this research contributes to FGV’s mission to promote the country’s socioeconomic growth.

The research project produced a report about the state of Sao Paulo, titled “Litigation in the Private Health Insurance System: An Empirical Analysis of Jurisprudence at the First and Second Levels of the Sao Paulo State Court of Appeals,” which you can access here.