A clearer way
into the evidence.

What makes a measure trustworthy?

A number can look precise without answering the question that matters. Counting how often something happens, for example, may miss whether it affects comfort, communication, or participation.

A useful research measure starts with a clear question. What are we trying to understand? Whose experience matters? Where and how will the information be collected? A person's own report, a caregiver's observation, a clinician's assessment, and a task performed during a visit can each contribute something different.

Researchers then ask whether the measure is reliable—consistent when no meaningful change has occurred—and valid for its intended purpose. They also need to know whether it can detect the kind of change being studied and whether that change matters to the people affected. Evidence from one condition or age group may not carry over to another.

Practicality matters too. Long questionnaires, difficult devices, or frequent reporting can create missing data and extra work. Designing with families helps researchers consider both relevance and burden.

The aim is evidence that is useful and interpretable, with clear limits on what each measure can tell us.

Sources: FDA guidance on fit-for-purpose clinical outcome assessments; NCATS clinical outcome assessment toolkit.

A pattern is a starting point

Imagine that a family's records show better sleep on days with more activity. That is a pattern worth investigating. It does not yet tell us whether activity improved sleep. Better sleep might have made activity easier, or another factor could have influenced both.

This distinction is the difference between an association and a cause. Observational research can identify useful questions and describe how experiences vary. To evaluate an intervention, researchers need a design that addresses other explanations. Depending on the question, that may involve comparison groups, random assignment, repeated measurements, or carefully justified analytical methods.

It also matters who was studied. An average result summarizes a group; it does not describe every individual in it. Age, disease features, other treatments, and circumstances may affect how a result applies to a particular person. Small subgroups can be especially uncertain.

When reading a study, ask: What was compared? What other explanations were considered? How large and uncertain was the difference? Who might the result apply to?

A promising pattern deserves careful follow-up. It is not, by itself, a reason to change treatment.

Sources: NIH: Understanding Clinical Studies; FDA rare-disease development guidance.

Why research needs the story between appointments

A clinic visit offers an important view of health. It cannot show every good day, disrupted night, difficult meal, or moment of participation that happens between appointments.

Natural history research follows how a condition unfolds over time. Researchers may examine symptoms, development, treatments already being used, and other relevant circumstances. This does not mean asking people to stop ordinary care. The purpose is to understand the condition's course and the differences between people, helping future studies ask better questions.

Lived experience helps identify which questions deserve attention. A change that seems small on a standard assessment may matter greatly to someone's daily life. Researchers need to ask people living with a condition—and caregivers when appropriate—what is important, then examine how to measure it meaningfully.

More tracking is not automatically better research. Collection should have a clear purpose, manageable demands, and appropriate consent and privacy protections. Missing entries need understanding rather than blame.

The strongest picture combines relevant observations over time with careful interpretation. It keeps people's priorities in view while being honest about what the data can establish.

Sources: FDA draft guidance on natural history studies (draft, March 2019); FDA patient-focused outcome assessment guidance.

Help useful science travel further.

Have published work that could help the rare-disease community? Share a link to a publicly available paper and a short explanation of its question through the general inquiry form.

Selected papers may be discussed in Science Notes with accurate attribution, links to the original work and a clear explanation of limitations. Interviews, supplied quotations and images require appropriate permission. Preprints will be identified as preprints. Sharing a paper does not guarantee coverage or imply endorsement or partnership.

Please share public links rather than unpublished manuscripts, confidential datasets or identifiable participant information.