Frequently Asked Questions

The process by which a change in one factor (such as an exposure) directly influences the likelihood of a specific outcome occurring. Because we usually can’t observe this influence perfectly, researchers rely on a set of guiding principles to judge whether a relationship is likely to be truly causal rather than just a coincidence or correlation.

A research study that tests the safety, efficacy, or both of a medical intervention - such as a drug, device, or procedure - in human participants. Clinical trials are conducted prospectively and follow a pre-specified protocol, but vary in study design.

A defined group of people who share a common characteristic or experience and are followed together over time for the purposes of research.

A factor that influences both an exposure and an outcome, making it appear that one causes the other when the true relationship is more complex. Identifying and accounting for confounding variables is one of the most important challenges in observational research. For example, a study might find that women who take a certain medication have worse pregnancy outcomes, but if those women also had more severe underlying health conditions to begin with, the condition itself may be driving the outcome, not the medication.

A group of participants in a study who do not receive the main intervention or exposure being studied (or who receive a standard or comparison condition), and are used as a point of comparison. By comparing outcomes between the control group and the group that received the intervention or exposure, researchers can better isolate the effect of the factor being studied and assess whether a causal relationship exists.

A statistical relationship between two variables (like an exposure and outcome) - meaning they tend to change together. Correlation does not mean that one thing caused the other; a third factor may be responsible for both, or the relationship may be coincidental. Distinguishing correlation from causation is one of the fundamental challenges in research.

Data from which all protected health information (PHI) has been removed or altered so that individuals cannot reasonably be identified. De-identification follows established standards to protect participant privacy while preserving the scientific value of the data.

How well a treatment or intervention produces the intended effect under controlled, ideal conditions (such as in a clinical trial). It answers the question: does this work when tested in a carefully managed setting, often compared to a control or placebo?

How well a treatment or intervention produces the intended effect in real-world conditions. It answers the question: does this work in everyday practice, where people, settings, and use may differ from a controlled setting (like a clinical trial)?

Digital records maintained by healthcare providers that document a patient's medical history - including events like diagnoses, medications, lab results, clinical visits, and physician notes - and when they happened in time.

The specific characteristics a person must have to be able to participate in a study. Studies may have inclusion criteria (characteristics someone should have to be included) and/or exclusion criteria (characteristics that would exclude someone from being able to participate).

Any factor that a participant has come into contact with that may influence a health outcome - such as a medication, a health condition, or a lifestyle factor.

A federal law that establishes national standards for protecting the privacy and security of individuals' health information. In the context of research, HIPAA governs how health data from health systems or other covered entities (like your electronic health record) can be collected, stored, used, and shared, and requires that organizations handling covered health information put appropriate safeguards in place to protect it.

When you sign a HIPAA Authorization form as part of participation in the Pregnancy Evidence Project, you are providing permission to request your health records from your care providers for research purposes.

The process by which a potential participant in a research study is fully informed about a study's purpose, procedures, risks, benefits, and their rights - including the right to withdraw at any time - before agreeing to participate. Informed consent is both an ethical obligation and a regulatory requirement for all human research.

An exposure, treatment, or change that may affect an outcome. In some studies, researchers assign the intervention (such as in a clinical trial), while in others (like observational studies) they observe interventions or exposures that occur naturally.

An independent committee of experts - including scientists, clinicians, and non-scientific community members - required by law to review and oversee research involving human participants. An IRB ensures a study's design is ethical, that risks to participants are minimized, and that proper informed consent procedures are in place before any research can begin.

Data collected from the same individuals over time. It allows researchers to track how health, symptoms, or outcomes change across a defined period - in the case of the Pregnancy Evidence Project, throughout the perinatal period - rather than capturing a single snapshot. Longitudinal data is particularly valuable for understanding how multiple variables interact and relate to outcomes.

A type of research that collects data from people going about their normal lives, without assigning treatments to different sub-groups or requiring participants to change their behavior. Rather than testing an intervention, observational studies analyze patterns that are already occurring across real-world experiences to identify associations, risk factors, and outcomes. The Pregnancy Evidence Project is an observational study.

Read the blog here for more context on observational studies in pregnancy, and why they’re so important for maternal health research.

A measurable health event or change in health status that researchers are trying to understand. Understanding how exposures relate to outcomes is central to the goals of observational research.

A patient registry is a data set created to enable observational research on a patient cohort or therapeutic area of interest to researchers - in the Pregnancy Evidence Project, maternal and neonatal health.

Data collected by a patient registry can help describe what happens in real life, compare different treatments, or look at safety. Observational studies can use patient registry data to establish evidence about how different factors (like medications or health conditions) might affect health outcomes.

A substance or treatment that has no active effect on the outcome being studied but is used as a comparison in research. It is designed to resemble the real intervention so that differences in outcomes can be more confidently attributed to the intervention itself rather than expectations or other factors.

A study that follows participants forward in time, collecting data in real time as events unfold rather than looking back at what already happened. This means that researchers define the study population, exposures, and outcomes before anyone experiences the outcome - and importantly, can allow researchers to assess causation, not just correlation. Prospective study design helps reduce the risk of certain biases that can affect research findings, like recall bias.

Any information that can be used to identify you and relates to your health, healthcare, or payment for healthcare. This includes obvious identifiers like your name and date of birth, but also less obvious ones like geographic data, dates of clinical visits, health insurance details, and device identifiers. In research, PHI is handled with the highest privacy and security standards to protect participant privacy.

A specific and rigorous type of clinical trial that randomly assigns participants to an intervention group (such as a drug, device, or procedure) or a control group (a similar group at baseline that does not receive the treatment or intervention being studied - such as standard of care or placebo) and compares their outcomes to directly test safety, efficacy, or both.

It’s considered the gold standard in research because of the controlled conditions that can establish a causal relationship between the intervention and outcome.

Information collected outside of controlled clinical trials from sources like electronic health records, health insurance claims, surveys, patient-reported experiences, wearable technologies (like smartwatches or sensor devices), lifestyle data, and beyond.

Conclusions about health outcomes, treatment patterns, or disease progression from analysis of real-world data.

The Pregnancy Evidence Project collects real-world data with the goal of generating real-world evidence that can inform maternal health research and care.

The natural tendency to misremember or forget details about past experiences, which can affect the accuracy of research findings. This is one reason prospective studies are often considered more reliable than studies that ask participants to look back on experiences that happened a long time ago.

A systematic scientific investigation designed to test a hypothesis, answer a specific question, or advance knowledge in a field. In health research, studies collect and analyze data in ways that can improve our understanding of disease, treatments, health outcomes, or, in the case of the Pregnancy Evidence Project, pregnancy.

A study that looks backward in time to answer a research question, analyzing data that has already been collected (rather than a prospective study, where data is collected in real time before the outcome is known). For example, researchers might look back at real-world data to study whether a medication taken in pregnancy was linked to a particular birth outcome, but they can only work with what was documented at the time, since the outcome being studied has already happened.

The overall structure and strategy of a research study. It defines how the study is set up to answer a research question - such as the type of study, how participants are grouped or compared, and how data will be used to evaluate relationships. Study design shapes what kinds of conclusions can be drawn from the research.

A detailed, written plan that specifies exactly how a study will be carried out in practice. It includes the study’s goals, who will participate, what data will be collected, how it will be analyzed, and the procedures used to protect participant safety and ensure data quality. While the study design describes the overall structure, the protocol provides the step-by-step details needed to implement it and is reviewed to assess the study’s ethics and safety.