Research Summaries

In this RECOVER study, researchers wanted to find out if natural language processing (NLP) could be used to identify Long COVID symptoms in children. NLP is a tool that can help find details in electronic health records (EHRs) beyond what is usually looked at in EHRs, such as diagnosis or billing codes (known as standard EHR data). Researchers used an NLP tool to look for 25 signs of Long COVID in children: 21 symptoms (like pain or extreme tiredness) and 4 types of daily life challenges (such as trouble with school). They compared children who had been diagnosed with Long COVID to those who had COVID-19 but did not develop Long COVID. The NLP tool analyzed more than 48,000 doctors’ notes within the EHRs of more than 10,000 children across 12 hospitals. Researchers found that the NLP tool identified almost all 25 symptoms much more often in the children who had Long COVID. The NLP tool also often identified patients’ symptoms that were not recognized when researchers only looked at standard EHR data. The study shows that using NLP to read EHR notes can help researchers better understand the symptoms and daily challenges that children with Long COVID experience when compared to looking only at codes and medication lists. This supports the idea that NLP should be used when doing scientific studies that need to identify children with Long COVID.

In this RECOVER study, researchers wanted to understand the connection between Long COVID and chronic kidney disease (CKD) in adults who had COVID-19. CKD is long-term, permanent damage to the kidneys. Researchers looked at the electronic health records (EHRs) of more than 2 million adults across 59 health systems in the US to see if people with CKD were more likely to develop Long COVID. The study found that, compared to people without CKD, those who had even mild cases of CKD were more likely to get Long COVID after COVID-19. Researchers also found that, compared to adults who did not get Long COVID, those who developed Long COVID and had healthy kidneys before getting COVID-19 were more likely to develop new CKD or have worsened kidney function within a year. These findings show how Long COVID and kidney disease can affect each other and highlight how important it is for doctors to monitor people’s kidney health after they get COVID-19.

This RECOVER study looked at whether people with Human Immunodeficiency Virus (HIV) are more likely to get Long COVID after having COVID-19. HIV is a virus that attacks the body’s immune system. Researchers used electronic health records (EHRs) from 2 nationwide research networks of people ages 21 and older who had COVID-19. They divided them into 2 groups: people with HIV and people without HIV. Researchers used 2 methods to find people in the groups who might have Long COVID. One used a computer system that looked for Long COVID symptoms in medical records. The other looked for official diagnosis codes that doctors enter when they think someone has Long COVID. They found that people with HIV may have a slightly higher risk of being diagnosed with Long COVID when looking for Long COVID symptoms in medical records. However, they found that there was no clear link between having HIV and receiving the official diagnosis code for Long COVID. This suggests that Long COVID may be missed in people with long-term health problems like HIV.

This RECOVER study tested a new way to identify Long COVID in children. Researchers created a special computer-based tool, called a computable phenotype (CP), that scans electronic health records (EHRs) for specific codes and symptoms linked to Long COVID. To test the CP, researchers scanned records for nearly 340,000 children who had COVID-19. Doctors then checked a smaller group of 651 children’s records to see how well it worked. The study showed that the CP was able to successfully find that a child had Long COVID. But it sometimes marked symptoms that fit in with health problems someone had before getting COVID-19 (pre-existing conditions), so it wasn’t clear if those symptoms were due to COVID-19. When the researchers updated the CP to pay closer attention to pre-existing conditions, it became more accurate at identifying Long COVID from EHRs. This study is important because it provides a faster and more consistent way for researchers to identify Long COVID in large groups of children. This tool can help doctors give better care to children with Long COVID.

In this RECOVER study, researchers looked at whether getting COVID-19 a second time increases a child’s or teenager’s risk of developing Long COVID. Researchers compared the electronic health records (EHRs) of more than 400,000 children and teens who had COVID-19 once with the EHRs of about 58,000 children and teens who had COVID-19 twice during the Omicron period. They found that children and teens who had COVID-19 twice were two times more likely to be diagnosed with Long COVID compared to those who only had it once. Children and teens who got COVID-19 twice also experienced other health problems more often, like being very tired, headaches, stomach pain, and heart or kidney problems. This study is important because it shows that children and teens are much more likely to develop Long COVID after getting COVID-19 more than once. These findings show why it is important to continue learning about ways to protect children and teens from getting COVID-19.

In this RECOVER study, researchers looked at whether Paxlovid could help prevent Long COVID. Paxlovid is a medicine commonly used to treat COVID-19 before it makes a person very sick. Researchers looked at the electronic health records (EHRs) of more than 445,000 people who had a high chance of getting very sick from COVID-19. People at higher risk included older adults (aged 50 and up) and people with certain medical conditions. To see if Paxlovid prevented people from developing Long COVID, researchers compared people who took Paxlovid right away once they got COVID-19 to people who did not. To make this a fair comparison, the researchers used a statistical method called a target trial emulation (TTE). A TTE uses EHRs to make two study groups as similar as possible, like they would be in a clinical trial. This makes the comparison fair because positive TTE results are more likely due to taking the treatment (in this study, Paxlovid) and not other differences, like a person’s age or health history. 

Researchers found that Paxlovid did not stop most people who were at high risk of getting very sick from COVID-19 from getting Long COVID. However, people who took Paxlovid were slightly less likely to develop some symptoms linked to Long COVID like fatigue (feeling very tired) and having trouble thinking. Researchers also found that Paxlovid seemed to improve the symptoms of people aged 65 and older a little bit more than it did for younger people. This study is important because it shows that taking Paxlovid is not a way to prevent Long COVID in most people, but it may help reduce symptoms in some higher-risk groups.

In this RECOVER study, researchers created a new computer tool to help doctors identify Long COVID more quickly and accurately. The tool, called a hybrid natural language processing (NLP) pipeline, quickly scans patients’ electronic health records (EHRs) to find descriptions of symptoms and figure out if a patient has them. Researchers tested this tool across 11 US health systems and found it was very accurate at identifying the right Long COVID symptoms. Patient EHRs contain a lot of important information, but it can take a long time to find specific symptoms across thousands of records. Researchers can now look at large amounts of health records from thousands of patients at once using the hybrid NLP pipeline technology. This study is important because it gives researchers a faster, more accurate way to identify Long COVID.

In this RECOVER study, researchers wanted to update a smart computer program, called a machine learning pipeline, to better identify people with Long COVID. In 2021, the first version of the program, called LCM 1, was created to identify people with or likely to have Long COVID. LCM 1 depended on people having a COVID-19 diagnosis date in their electronic health records (EHR). This meant LCM 1 could miss people who may have taken a COVID-19 test at home. LCM 1 also did not look at information about whether people got COVID-19 more than once. To improve the program and create a new version called LCM 2, researchers used more than 5 million EHRs from a large set of data called the National COVID Cohort Collaborative (N3C). They taught the program to look at a person’s health information over many years, not just starting from their first recorded COVID-19 diagnosis. Researchers found that LCM 2 was very accurate. They used it to estimate that about 1 in 10 people in the database who had COVID-19 went on to develop Long COVID. This study is important because it shows that older machine learning models, like LCM 1, can be updated to keep up with the way an illness is tracked and diagnosed over time. This can help other researchers improve their machine learning models to produce more accurate findings.