Search results

Researchers often use data from patients' health records to compare treatments. But many things--not just treatments--affect patients' health. To figure out whether changes in patients' health result from treatment or something else, researchers can use statistical methods called instrumental variables, or IVs. IV methods account for factors that affect health but aren't in patients' health records, such as eating habits. Existing IV methods work well when looking at health outcomes that are measured using certain types of scales, such as blood pressure. But existing methods don't work as well to measure the time until a health event occurs, particularly when an event, like death, has not occurred for many patients in the study.

In this study, the research team created and tested a new IV method to more accurately estimate how a treatment relates to the time until a health event.

The research team based their new method on an existing method called Bayesian Additive Regression Trees, or BART. To test the new method, the team used data created by a computer program to look like real patient data. Then they compared the new method with current methods under different scenarios. Each scenario included three treatments. The team changed the total number of patients, the number of patients who took each treatment, and how alike or different the patients were who took each treatment. Across all scenarios, the team predicted the average treatment effect for all patients and for only patients who received a treatment.

Next, the research team used the new method with real data from patients with lung cancer who were receiving care in New York City hospitals. The team compared three types of surgery: open chest, robotic assisted, and video assisted. The team looked at the effects of each type of surgery on four health outcomes: breathing problems; length of hospital stay after surgery; stay in an intensive care unit, or ICU; and the need to return to the hospital.

Patients, doctors, and researchers helped design the study.

In randomized controlled trials, or RCTs, researchers assign patients by chance to different treatments to compare the benefits and harms. In RCTs, researchers have a high level of control over how patients receive treatment. RCTs often take place in research clinics with staff who monitor how patients follow treatment plans.

Pragmatic RCTs, or pRCTs, take place where patients typically receive treatment, such as a regular clinic. pRCTs can help capture the real-world effects of treatment but determining whether a treatment works can be hard in pRCTs. Also, no clear guidance exists about how to collect and analyze data from pRCTs. Some kinds of analysis are better for helping researchers focus on what's important to patients.

In this study, the research team created guidance for collecting and analyzing data in pRCTs so that results reflect what matters to patients and researchers.

To access the methods and software, please visit the following Github repositories:

• CDP-analysis-2018
• GFORMULA-RCT-SAS
• IV-Bounds
• CHARM_reanalysis
• Adherence_LRCCPPT

The Comprehensive Post-Acute Stroke Services (COMPASS) Study is a pragmatic cluster-randomized clinical trial that evaluated the real-world effectiveness of the COMPASS transitional care (COMPASS-TC) model compared to usual care among adult stroke and transient ischemic attack (TIA) patients discharged home between 2016 and 2018. In Phase 1, 40 North Carolina hospital units were randomized 1:1 to the COMPASS-TC intervention or usual care, stratified by stroke patient volume and stroke center certification. In Phase 2, hospitals randomized to usual care crossed over to implement COMPASS-TC, and hospitals randomized to the intervention sustained COMPASS-TC. The intervention was patient-centered and assessed social and functional determinates of health to inform individualized care plans for secondary prevention, recovery, and referrals to services and community-based resources. COMPASS-TC was consistent with Centers for Medicare and Medicaid Services (CMS) TC management reimbursement requirements.

The primary outcome was functional status (Stroke Impact Scale-16; SIS-16) at 90 days; secondary outcomes were mortality, disability, medication adherence, depression, cognition, self-rated health, fatigue, care satisfaction, home blood pressure monitoring, falls, and caregiver strain. Telephone interviewers, blinded to treatment assignment, assessed these outcomes at 90 days.

Research that focuses on what's most important to patients can inform health decisions. Researchers use different methods to identify what's most important to patients.

In this study, the research team compared two methods for identifying what's most important to patients: one-on-one interviews and group concept mapping, or GCM. GCM is a three-round process that helps researchers get input from a group. In the first round, people brainstorm topics that are important to them. Next, people sort the topics into clusters based on similar ideas. Finally, researchers create a map to display and discuss the topics. Researchers can use the complete GCM process or the brainstorming round only.

The research team looked at one-on-one interviews versus GCM and compared the number of topics patients named and the amount of time and money required.

Research tools available to help advance knowledge of effective community-based elder mistreatment (EM) interventions are limited. The field lacks an understanding of what success means in EM response program (EMRP) interventions, which work directly with victims to reduce the risk of re-victimization. Without establishing indicators of EMRP success, it is not possible to develop valid intervention outcome measures to compare different EMRP models toward the development of evidence-based practice. Informed by the EMRP practice principle of older adult self-determination, this study developed a victim-centric taxonomy of case outcomes that indicate EMRP success.

This study drew on two sources of data, including interviews with EM victims and a scoping review to inform taxonomy development. Prioritizing the perspective of victims, this study conducted interviews with 27 victims involved in EMRP services who vary in EM subtype, gender, and race/ethnicity.

The taxonomy of successful EMRP outcomes will serve as important research infrastructure to support the development of EMRP intervention outcome measurement in future research.

Researchers usually design studies to find out if a new treatment works better than no treatment or better than a treatment that is currently used. But sometimes researchers may want to know that a new treatment is not worse than one that's in use. These studies are called non-inferiority, or NI, studies. Researchers conduct NI studies when a new treatment has other benefits such as fewer side effects, even though it may not work better than the one in use. NI studies can provide useful information, but they are hard to design and conduct.

In this study, the research team tested different statistical methods for NI studies that compare treatments.

To access the methods and software, please visit the following Github repositories:

• Poisson3armNI
• Binary3armNI
• bayesianSWcontinuous
• SMART3armNI

Drug-Abuse Treatment Outcomes Study (DATOS) is a prospective study designed to determine the outcomes of adult drug abuse treatment delivered in typical, stable, community-based programs and to provide comprehensive information on continuing and new questions about the effectiveness of drug abuse treatment for adults currently available in a variety of publicly funded and private programs. The study examined the role of treatment outcomes and program type, client characteristics (including dependence, treatment history, and physical and mental health comorbidities), treatment received (e.g., length and intensity of services provided), therapeutic approaches, provision of aftercare, and research on the components of effective treatment, including factors that engage and retain clients in programs. Four types of programs were included: outpatient methadone (OPM), short-term inpatient (STI), long-term residential (LTR), and outpatient drug-free (ODF). Respondents were sampled from among adults admitted to drug abuse treatment programs in 11 representative U.S. cities during 1991-1993.

Clients entering treatment completed two comprehensive intake interviews (Intake 1 and Intake 2), approximately one week apart. This information is provided in Parts 1 and 2 of the data collection. These interviews were designed to obtain baseline data on drug use and other behaviors, as well as information on background and demographic characteristics, patterns of dependence, living situation and child custody status, education and training, income and expenditures, and HIV risk behaviors, along with assessments of dependence, mental health, physical health, and social functioning. Data on criminal justice status and criminal behavior are reported in Part 5, Illegal Activities Data, and are drawn from the Intake 1 interview. Data reflecting during-treatment progress, including service delivery and client satisfaction, were collected in the one-, three-, and six-month in-treatment interviews (Parts 3, 4, and 8). The 12-Month Post-Treatment Follow-Up Interview (Part 6) replicated many of the intake questions and focused on key behaviors in the year following treatment. Part 7 includes variables for time in treatment and interview availability indicators. The 12-Month Follow-Up Urine Result data (Part 9) provide the results from urine sample tests that were given to a sample of subjects at the time of the 12-Month Follow-Up Interview. Urine specimens were tested for eight categories of drugs (amphetamines, barbiturates, benzodiazepines, cannabinoids, cocaine metabolite, methaqualone, opiates, and phencyclidine). The drugs covered in the study were alcohol, tobacco, marijuana (hashish, THC), hallucinogens or psychedelics such as LSD, mescaline, and PCP, cocaine (including crack), heroin, narcotics or opiates such as morphine, codeine, Demerol, Dilaudid, and Talwin, downers or depressants such as sedatives, barbiturates, and tranquilizers, amphetamines or other stimulants such as speed or diet pills, and other drugs. Part 10 contains data for 1393 clients who were interviewed 5 years post treatment. This part contains many of the same types of questions asked during previous interviews.

This study evaluated program design, quality of treatment delivery, and program effectiveness of three separate state sponsored alcohol specific treatment programs in prisons located in Montana, Ohio, and Texas from 2006 to 2012.

Comparative effectiveness research compares two or more treatments to see which one works best for which patients. But patient traits, such as age or income, may affect patients' treatment choices. These traits may also affect patients' responses to treatments. As a result, researchers may have trouble telling whether a patient's traits, the treatment, or a mix of the two affected how well a treatment worked.

Statistical methods called matching methods can help address this problem when researchers use patient data to compare the effects of treatments. Matching methods help researchers find data from patients who had similar traits such as age or race and received different treatments. Because the patients are similar except for the treatment they receive, the differences in patients' health can more likely be credited to the treatment. Existing methods work well for comparing up to two treatments. But they may not work with three or more treatments.

In this study, the research team created two new matching methods to compare the effects of three or more treatments. The team then analyzed the new methods under different conditions to see how well each worked."

A randomized trial is one of the best ways to learn if one treatment works better than another. Randomized trials assign patients to different treatments by chance. But they are not always affordable, and they take a long time to complete.

When randomized trials aren't possible, researchers can use observational studies to learn how treatments work. In observational studies, researchers look at what happens when patients and their doctors choose the treatments. Traits such as age or health may affect treatment choices. These traits may also affect patients' responses to treatment, making it hard to know if the treatment or the traits affected the patients' responses.

Some study designs and statistical methods may help address this problem and make results from observational studies more useful. These methods can give researchers more data about whether treatments work and how the same treatment can affect groups of patients differently.

The research team conducted three studies to test different methods of designing and analyzing observational studies. They wanted to know if observational studies that used these methods produced results similar to randomized trials.

The purpose of the project was to (1) determine whether the combined dynamic (SOTIPS) and static risk assessment (Static-99R) tools better predicted sexual recidivism than either alone, and (2) determine whether the tools could be implemented successfully in more representative populations. Previous research has established a "status quo" for risk assessments.

This study was set within the context of the developing sexual offender risk prediction field, where investigators explored reliable and valid means to assess what have been termed "dynamic risk factors." Instruments that identify the specific psychological risk factors present in the individual offender ought to allow treatment for that individual to be tailored to these specific needs, thus increasing its effectiveness. Thus, instruments have been designed to:

• Assess psychological factors that are empirically related to sexual recidivism, thus creating a basis for selecting treatment targets
• Show robust incremental predictive validity relative to Static-99R or other measures of static risk factors
• Measure change in a way that is convincingly related to sexual recidivism
• Incorporate and point risk managers towards some of the factors identified in the desistance literature
• Improve the effectiveness of treatment in reducing sexual recidivism

Enrollment of sex offenders in the evaluation study began in April 2013. To be included, offenders needed to be Static-99R eligible (an adult male convicted of a contact or non-contact sex offense with an identifiable victim), mentally cognizant, released to community supervision, and at least 18 years old in January 2013 in Maricopa County and April 2013 in New York City.

Overview

The Florida Department of Health and the Florida Cancer Data System (FCDS) collaborated with a hospital network composed of nine clinical facilities, to capture electronic medical records (EMR) data of patients who were diagnosed with or treated for invasive breast cancer from 2007 to 2010. Certain hospital data elements were available throughout 2006. An additional year of 2011 follow-up data was also available for a subset of patients receiving medication treatment. The purpose of the data capture was to advance patient-centered outcomes research to reduce the morbidity and mortality of cancer and other comorbidities.

A breast cancer pilot study was also conducted from a subset of all transmitted EMR records, consisting of admission records with a principal and/or secondary ICD-9-CM diagnosis between 174.0 and 174.9. The subset dataset was then linked to the central cancer registry using patient social security number, first and last name, and date of birth. Using a deterministic matching algorithm a total of 11,506 unique patients were matched to a patient in the FCDS database, resulting in 12,804 primary tumors and 53,940 unique hospital admission records. While the hospital EMR defined the patient dataset, all registry records for that patient were included in the final breast cancer pilot database, regardless of the reporting hospital or the date of diagnosis. This was to ensure capture of the entire diagnostic and treatment profile for each breast cancer patient.

Data Access

These data are not available from ICPSR. The data contain confidential information that can directly identify a patient. There are also reporting facility data. Therefore, to obtain these data, researchers will need to follow the Florida Cancer Data System data-sharing agreement process, as outlined on the FCDS data sharing request.

Patient-reported outcome measures are surveys that ask patients how they feel and what activities they can do. These surveys ask about things such as how well people sleep and how much their pain interferes with daily life.

In this study, the research team wanted to learn if two clinics could gather patient-reported outcome measures during routine care visits, and if patients with type 2 diabetes could use the results to set goals for improving their health. The research team also wanted to learn if patients and clinic staff saw value in using these measures.

Comparative effectiveness research compares two or more treatments to see which one works better for which patients. Sometimes, groups of people respond differently to the same treatment. For example, women might, on average, receive more benefit from a treatment than men do. If researchers group women and men together when they analyze study data, they may miss this difference and overlook some of the benefits of a treatment.

Researchers can analyze data on the effects of a treatment in many ways. Each way has strengths and weaknesses. Bayesian regression is one method that allows researchers to consider various factors in their analysis, such as patients' ages, sex, or health problems. This method can help researchers understand how different groups of people respond to a treatment. But it requires advanced computer programs that are not readily available to all researchers.

In this study, the research team wanted to make it easier for researchers to use Bayesian regression.

To access the R package, please visit the Beanz CRAN webpage.

A randomized controlled trial, or RCT, is often the best way to learn if one treatment works better than another. RCTs assign patients to different treatments by chance. But RCTs are not always feasible. In such cases, researchers can use observational studies. In observational studies, researchers look at what happens when patients and their doctors choose the treatments. Traits such as age, gender, or health status may affect treatment choices. These traits may also affect patients' health, making it hard to know if changes in patients' health are due to treatment or to patient traits.

To figure out whether changes in patients' health result from treatment or something else, researchers use statistical methods. Two of these methods are:

• Propensity score, or PS. PS methods compare the health of patients who have similar measured traits but received different treatments. These traits are in patient health records.
• Instrumental variable, or IV. IV methods account for things that may affect treatment choice and patients' health but aren't in the patients' health records, such as personal preference about treatment.

But existing PS and IV methods don't work well when data sets include a lot of traits and health conditions for each patient. Such data sets are called high-dimensional data. In this study, the research team created and tested one PS method and one IV method for use with high-dimensional data.

A rare disease is one that affects fewer than 200,000 people in the United States. Because few people have these diseases, clinical studies on treatments can be hard to conduct. One way to study rare disease treatments is with an small n sequential multiple assignment randomized trial (snSMART) study.

snSMART studies have two stages. In the first stage, researchers assign patients to a treatment by chance. In the second stage, patients may stay with the same treatment or switch treatments. Patients stay on the same treatment if it's working well. If the treatment isn't working, researchers assign patients by chance to a new treatment.

snSMARTs can help researchers learn more from a smaller number of patients than a standard clinical study. But most current methods for analyzing snSMARTs use data only from the first stage, which can lead to inefficient results.

In this project, the research team developed and tested new methods that use data from both stages to analyze snSMARTs. The team compared results from the new methods to actual treatment effectiveness to see:

• Bias, or whether results are too high or too low
• Efficiency, or how big the difference is between the results and actual treatment effectiveness

This study contains two supplementary documentation files. There is no data included in this release.

In a stepped-wedge cluster randomized trial, or SW-CRT, researchers compare new treatments to standard treatments in groups of patients, such as patients at different clinics, to look at the treatments' effectiveness. They assign groups by chance to switch from the standard to new treatment at different time points until all groups have received the new treatment. The different time points to switch treatments are called steps.

SW-CRTs take time and resources. If researchers know they can't collect data on all groups and all steps in a SW-CRT, they can plan to use an incomplete SW-CRT design. In incomplete SW-CRTs, researchers plan the study knowing that some clinics or steps will have missing data. But researchers need better guidance for planning incomplete SW-CRTs that still get accurate results.

Also, current methods for planning how many patients and groups should take part in SW-CRTs don't work well for large studies. They also don't work well with certain types of outcomes, like yes or no outcomes; outcomes that have counts, like number of hospital visits; or continuous outcomes, like a score from 0 to 100.

In this study, the research team developed and tested new methods to design and analyze SW-CRTs with different patterns of planned missing data, large data sets, and different types of outcomes.

Research studies called clinical trials test treatments to see if they are safe and effective for patients. When designing clinical trials, researchers must plan to include enough patients with different traits for the study to have accurate results. Once the study starts, researchers must follow the plan. Sometimes, early results from a trial show that a group of patients with a certain trait may have more benefits or harms from the treatment than other groups. For example, the treatment may not work for patients with a history of heart disease. In the standard trial design, researchers can't change the plan to stop enrolling these patients once the trial starts.

In this study, the research team compared the standard trial design with more flexible approaches known as adaptive enrichment designs. These designs set up rules that allow researchers to change the study plan. For example, if early results show a treatment doesn't work for patients with heart disease, researchers can stop enrolling these patients in the trial. The team compared the trial designs using data from four completed trials.

To access the methods and software, please visit the AdaptiveDesignStreamlinedOptimizer GitHub.

These data are part of NACJD's Fast Track Release and are distributed as they were received from the data depositor. The files have been zipped by NACJD for release but not checked or processed except for the removal of direct identifiers. Users should refer to the accompanying readme file for a brief description of the files available with this collection and consult the investigator(s) if further information is needed.

This study identified risk and protective factors for dating violence (DV) among young adults (ages 18-22) with a history of maltreatment and placement in foster care, and who had enrolled in Fostering Healthy Futures (FHF) during 2002-2009. FHF is a Colorado-based positive youth program for maltreated youth. This study focused on factors that ameliorated the effects of risk to reduce DV perpetration and victimization in young adulthood. The participants were interviewed at three different points during the FHF time frame. That data provided a basis for determining risk and mediating factors which in turn were compared to the current study's DV outcomes.

The risk and protective factors included:

• Mental health
• Substance abuse
• Social support
• Gender Stereotypes
• Attitudes about Teen DV
• Communication Skills

Perpetration and victimization outcomes were then examined in relation to the risk and protective factors.

The collection includes 1 SPSS file: NIJ-2013-VA-CX-0002---2nd-revision---5-17-18.sav (215 cases / 2023 variables).

Randomized controlled trials, or RCTs, look at how well treatments work. But people who take part in RCTs may differ from patients who receive care in clinics. For instance, patients who take part in RCTs may be less likely to smoke or may have fewer health problems. These differences can affect how well a treatment works. As a result, a treatment may work differently for a patient receiving care in a clinic than it did for patients who took part in the RCT.

Researchers can use statistical methods to account for differences in patient traits and behaviors. In this project, the research team developed and tested new methods to account for these differences. They used the methods to apply RCT results to patients receiving care in clinics.

To access the methods and software, please visit the generalizability_g_form_IPW and ExtendingInferences GitHub repositories.

Patients may respond differently to the same treatment due to differences in personal traits such as age, gender, or the number and type of health problems they have. Researchers use statistical methods to predict how well a treatment may work for patients based on their personal traits. But current methods may not work well if patients have many health problems or are taking other medicines.

In this project, the research team created new methods to figure out which patient traits are related to treatment benefits to help doctors and patients understand the likely treatment benefits for individual patients.

To access the methods, software, and R package, please visit the personalized CRAN webpage and personalized GitHub

Patients' beliefs and expectations may affect how they respond to treatment. But these feelings are hard to measure.

In this study, the research team created a set of surveys called Healing Encounters and Attitudes Lists, or HEAL. HEAL helps researchers understand patients' beliefs and expectations about treatment. HEAL measures patients'

• Connections with their doctors and nurses
• Feelings about their doctor's office and staff
• Expectations about treatment
• Outlook on life
• Strength of spiritual beliefs
• Comfort with complementary and alternative medicine, or CAM
• The team also used the Patient-Reported Outcomes Measurement Information System, or PROMIS, to measure patients' pain, health, and function. PROMIS is a set of surveys researchers and doctors use for many diseases and treatments.

The team wanted to learn if HEAL could predict how patients respond to treatment for chronic pain. Chronic pain is pain that lasts for months or years. The team used HEAL and PROMIS to look at why some groups of patients respond differently to treatment for chronic pain. Patients got either conventional treatment, such as physical therapy or medicine, or CAM, such as acupuncture, chiropractic treatment, or massage.

One goal of comparative effectiveness research is to find out which treatments work best for different groups of patients. For example, treatments may work differently for patients with only one health problem than for those with more than one health problem.

In observational studies, researchers look at health outcomes when patients and their doctors choose the treatments. These studies often use data from electronic health records, or EHRs. Researchers can apply propensity score, or PS, methods to look at different groups of patients. With PS methods, researchers create groups of patients with similar traits who had different treatments. But PS methods require researchers to have data on all patient traits that could affect how well the treatment works. With EHR data, data on some patient traits, like health problems, may be missing. Using current PS methods in observational studies may lead to biased results.

In this study, the research team created new guidance for using PS methods with EHR data to look at the effects of treatment in different groups of patients. The team also created and tested new PS methods to make groups of patients with similar traits.

Comparative effectiveness research compares two or more treatments to see which treatment works better for which patients. Such research may include

• Randomized controlled trials, or RCTs. Researchers assign patients to a treatment by chance. Researchers consider RCTs to be the best way to figure out when changes in patients' health result from the treatment.
• Observational studies. Researchers study what happens when patients and their doctors choose treatments. Patient traits, such as age or health, may affect treatment choices. These traits may also affect patients' responses to treatments. Determining whether a patient's traits, the treatment, or a mix of the two affected how well the treatment worked may be difficult.

In observational studies, researchers use statistical methods to help find out whether changes in patients' health result from treatment or something else. Existing methods work well when studies look at whether treatment affects the risk of a health event, such as a heart attack. In these cases, researchers can compare how often patients had heart attacks before and after patients receive treatment. But existing methods don't work well when studies look at the risk of a one-time event, such as death.

In this study, the research team tested a new statistical method for observational studies called posttreatment event rate ratio, or PTERR, that helps figure out whether a treatment reduces the risk of death.

Cluster randomized trials, or CRTs, are studies that compare treatments across different groups of patients, or clusters. An example of a cluster is people who receive care at one clinic.

To reduce bias in CRT results, researchers assign clusters by chance to different treatments. But what happens after they assign treatment can lead to differences across clusters and bias the results. For example, patients who visit clinics assigned to a treatment may be older than patients who visit clinics not assigned to that treatment. Current statistical methods for analyzing data from CRTs don't work well to account for these differences.

In this study, the research team developed new methods to account for differences across clusters after treatment assignment.

Overview

The North Carolina Integrated Data for Researchers (NCIDR, pronounced "Insider") was funded to develop a robust research data warehouse for storing merged data from four different publicly-funded sources in North Carolina. Community Care of North Carolina maintains this unique database on behalf of the North Carolina Department of Health and Human Services, and facilitates requests for access to integrated behavioral health services data for research purposes. This expanded data set has great value to researchers in North Carolina and elsewhere. The NCIDR warehouse is a unique resource for obtaining the most complete picture of the health services delivered to people with severe mental illness in North Carolina. Few examples of such an integrated warehouse exist anywhere else, and NCIDR makes it possible for researchers and epidemiologists to conduct comparative effectiveness research related to people with these conditions.

The merged data sources include:

• Medicaid claims and enrollment data for nearly 1 million individuals with MH, DD and SA diagnoses.
• IPRS (Integrated Payment and Reporting System) -- covers primarily outpatient mental health services for people that do not qualify for Medicaid (approximately 250,000 individuals).
• HEARTS (Healthcare Enterprise Accounts Receivable Tracking System) -- documents services delivered by inpatient State Mental Health facilities (approximately 25,000 individuals).
• Piedmont Behavioral Health (Medicaid waiver) -- behavioral health encounter data from Medicaid's capitated arrangement in five counties (approximately 25,000 individuals).

Data are available for four state fiscal years, 2008 through 2011 (7/1/2007--6/30/2011). Each year has three data sets (claims, client, provider) in addition to multiple lookup tables with definitions. Population includes any Medicaid client with a claim that contains any MH, DD or SA (290xx through 319xx) diagnosis at least once in the four year time period, plus all clients appearing in the other 3 data sources. Requests will need to specify required time periods and clearly define the population being studied. Note that individuals dually enrolled with Medicare during months in which they are dually enrolled are excluded.

The data available for future use will include a claims file, a client file, a provider file and multiple lookup files. The claims file contains approximately 83 columns including 30 columns for diagnosis codes. The client file is approximately 78 columns which displays 12 columns each (one for each month in the SFY) for eligibility, enrollment, assigned network, primary care physician and dual status indicator. There are 5 columns in the Provider file. The lookup file will contain tables for every code that requires a description. The data will be parsed into individual state fiscal years.

Data Access

These data are not available from ICPSR. The process for requesting access to the integrated data is detailed on the NCIDR Web site, specifically the Request Process Overview page. Researchers interested in requesting access are strongly encouraged to contact the Director of Evaluation at [email protected] to discuss his/her intent to submit a Request Form. Some may also need to complete the Data Use Agreement if requesting data that are not completely de-identified.

Although IRB approval must be documented prior to release of data, NCIDR will accept applications with conditional IRB approval and researchers may discuss projects with the Director of Evaluation at any stage of development. A Research Oversight Committee (ROC) that includes stake holders from the NC Department of Health and Human Services (DHHS), the NC Division of Medical Assistance (DMA), the NC Division of State Operated Healthcare Facilities (DSOHF), the NC Division of Mental Health, Developmental Disabilities and Substance Abuse Services (DMHDDSAS), the NC Office of Rural Health and Community Care (ORHCC), the Community Care of North Carolina (CCNC) and other community partners will review research requests and grant approval when applicable. Once approved, please note that CCNC must charge a nominal fee of $3,000 to cover costs related to the preparation and transmission of files to the researcher (additional charges may apply depending on the specific programming needs).

This project aims to improve the methods that researchers use to compare how treatments affect different patients. When researchers use data from patients' health records to compare treatments, it's often hard to know whether changes in a patient's health are from the treatment or something else. Factors other than the treatment may affect the patient's health, including

• A patient's traits, such as age, gender, or other health problems
• Group-level factors, such as where patients get care or where they live

To address this problem, researchers rely on statistical methods. Existing methods use data from patients who have similar traits but received different treatments. But they may not work well if some group-level factors affect both the treatment and patients' health. In this study, the research team created two new ways of including group-level factors in the methods they use to find similar patients.