03 · What You Need to Know
How to Determine the Number and Timing of Follow-Up Assessments
What Is a Follow-Up Visit in Research?
A follow-up visit is a planned contact or assessment after an initial reference point, such as enrollment, baseline measurement, intervention allocation, or treatment initiation. Its purpose may be to measure outcomes, monitor safety, document exposures, assess adherence, or understand changes in participant circumstances.
The term visit does not necessarily mean attendance at a physical research facility. Depending on the protocol, follow-up can occur through telephone interviews, electronic questionnaires, home assessments, local facilities, or appropriately authorized record review.
Researchers should distinguish the number of participant contacts from the number of substantive outcome assessments. A brief appointment reminder is not equivalent to a full measurement visit, and an administrative contact may not provide analyzable follow-up data.
Why Does the Research Question Determine Follow-Up Frequency?
The number of assessments should follow from the phenomenon being studied rather than from a customary schedule.
If the objective is to estimate an outcome at a particular endpoint, baseline and one follow-up assessment may be sufficient, depending on the design. If the objective is to examine trajectories, fluctuations, delayed effects, or the timing of events, additional observations may be necessary.
For example, a study measuring immediate knowledge acquisition after a training session addresses a different question from one investigating whether learning is retained six months later.
Likewise, a study of symptom variability may require relatively frequent observations, whereas a study of long-term institutional outcomes may use less frequent assessments over a longer period.
Researchers should specify the intended estimand or interpretation before deciding how many measurements to collect.
How Do Different Study Designs Affect Follow-Up Requirements?
| Study design |
Possible follow-up purpose |
Important consideration |
| Pretest-posttest study |
Measure change between baseline and a subsequent assessment |
Two observations do not establish the shape of change or necessarily support causal inference |
| Randomized controlled trial |
Measure prespecified outcomes and monitor relevant harms |
Assessment timing should match the estimand, intervention, and safety requirements |
| Longitudinal cohort study |
Observe outcomes and exposures over time |
Visit spacing should reflect the expected pace of change and analytical objectives |
| Repeated-measures study |
Estimate within-person variation or trajectories |
Measurement frequency must support the planned temporal analysis |
| Qualitative longitudinal study |
Understand how experiences or interpretations develop |
Interview timing should reflect meaningful transitions or events |
| Survival or event-history study |
Identify events and, where needed, their timing |
Event ascertainment and interval uncertainty may determine contact frequency |
| Implementation study |
Examine adoption, delivery, maintenance, or changing context |
Assessments should correspond to implementation stages and outcomes |
These are methodological considerations rather than prescribed visit counts. Two studies using the same general design may require different schedules because their outcomes and intended conclusions differ.
How Does the Expected Timing of an Outcome Affect Visit Scheduling?
Follow-up assessments should occur when the outcome is scientifically meaningful and can be measured with sufficient accuracy.
Some outcomes emerge quickly. Others develop gradually or become observable only after participants have experienced the intervention or exposure for an extended period.
For example, immediate satisfaction with a training program can be assessed shortly after delivery. Changes in workplace practice may require sufficient time for participants to apply what they learned.
Scheduling assessments too early may miss delayed outcomes. Scheduling them too late may overlook transient effects or make it difficult to determine when changes occurred.
Researchers should use relevant theory, prior evidence, outcome definitions, and practical knowledge of the setting to identify plausible measurement points.
When Is One Follow-Up Assessment Enough?
One follow-up may be sufficient when the primary question concerns an outcome at a single prespecified endpoint and the study does not require additional information about trajectories, event timing, or intermediate changes.
For example, a randomized educational trial might compare a validated learning outcome between groups twelve weeks after allocation. If that endpoint directly answers the primary question, repeated intermediate questionnaires may add little value.
However, a single follow-up cannot establish whether an effect emerged immediately, increased gradually, or fluctuated before the endpoint.
It also may not be adequate when participant safety requires additional monitoring or when the research aims to investigate maintenance of effects.
The decision should reflect what the study needs to infer, not an assumption that one visit is always efficient.
When Are Multiple Follow-Up Assessments Necessary?
Multiple assessments may be justified when the study examines trajectories, delayed responses, sustained effects, time-varying exposures, or processes that cannot be understood through a single endpoint.
For example, an intervention may produce an immediate improvement that diminishes over time. Measuring only the immediate outcome would not answer whether the improvement persists.
Repeated assessments may also help distinguish stable differences between participants from within-person changes, provided the design and analysis are appropriate.
Yet additional visits are not automatically informative. Measurements collected at intervals too short to capture meaningful change may produce redundant observations while increasing burden.
The question of whether repeated measurements cause fatigue or alter behavior should also be considered when deciding how frequently to assess participants.
How Should Researchers Choose the Interval Between Visits?
The interval should reflect the timescale of the phenomenon and the precision required for the intended inference.
Daily measurements may be useful for rapidly changing experiences. Monthly assessments may suit some behavioral outcomes. Annual follow-up may be appropriate for certain long-term outcomes.
These are examples, not recommended intervals for all studies in those categories.
Researchers should consider whether relevant changes are likely to occur between observations, whether participants can recall the information accurately, and whether assessment timing itself may influence behavior.
For event outcomes, less frequent observation may produce uncertainty about when an event occurred. Analytical approaches may need to accommodate interval-censored event times rather than treating the date of detection as the exact event date.
Do Follow-Up Visits Need to Occur at Exactly the Same Time for Everyone?
Consistent assessment timing can support comparability, particularly when outcomes change with time since intervention or exposure.
However, rigid scheduling may increase missed appointments. Prespecified assessment windows can provide flexibility without sacrificing the temporal definition of an outcome.
For example, a protocol might define a target assessment at twelve weeks with an allowable window of several days, when that tolerance is scientifically appropriate.
SPIRIT 2025 recommends clearly describing the timing of enrollment, interventions, assessments, and visits in randomized trial protocols, including target time points and acceptable windows where relevant. Its protocol schedule provides a useful structure for documenting these decisions.
Researchers should distinguish the planned visit date from the actual assessment date and account for meaningful timing variation in the analysis when necessary.
How Do Safety Monitoring Requirements Affect the Number of Visits?
Some studies require follow-up contacts primarily to detect, assess, or manage potential harms rather than to collect efficacy outcomes.
For example, an investigational intervention may require monitoring at specific intervals because of known or plausible safety concerns. Such assessments cannot be removed simply because the primary outcome could be measured less frequently.
Safety schedules should be based on the intervention, available evidence, risk assessment, and applicable clinical, ethical, or regulatory requirements.
Researchers should distinguish essential safety contacts from assessments introduced mainly for administrative convenience. A remote safety contact may sometimes be appropriate, but not when a necessary examination or procedure requires in-person assessment.
How Can Follow-Up Frequency Affect Participant Retention?
Frequent follow-up can provide opportunities to maintain contact and identify participation difficulties. It can also increase demands on participants, especially when appointments require travel, lost income, or repeated lengthy assessments.
These effects may vary by population and study context. More contact does not automatically improve retention, and fewer visits do not guarantee lower attrition.
Researchers should examine whether the cumulative schedule is manageable and whether participation burden could contribute to attrition and missing outcome data.
Retention planning should include clear communication, realistic scheduling, and appropriate procedures for obtaining essential outcome information when participants discontinue an intervention but remain willing to contribute follow-up data.
Can Remote Follow-Up Replace In-Person Visits?
Sometimes. Telephone interviews, online questionnaires, home-based assessments, and authorized record review may reduce travel and scheduling burdens.
However, alternative modes must be suitable for the outcome and participants. A laboratory measurement may require specialized equipment, while a self-reported experience may be collected remotely if the instrument and administration procedures support valid interpretation.
Researchers should examine whether internet-dependent participation excludes relevant people before assuming remote follow-up is universally accessible.
Changes in assessment mode may introduce measurement differences, privacy concerns, or additional training requirements.
Should Researchers Combine Several Assessments Into One Visit?
Combining procedures may reduce transportation costs and the number of separate appointments. This can be valuable when participants live far from the research site or have limited scheduling flexibility.
However, longer combined visits may increase fatigue or discomfort. Some assessments also require particular timing, fasting conditions, rest periods, or separation from other procedures.
Researchers should evaluate total burden and measurement quality rather than simply minimizing the number of calendar appointments.
The appropriate duration of a combined visit should be assessed using realistic estimates of research visit and assessment duration.
How Should Follow-Up Schedules Be Reflected in the Analysis Plan?
The analysis should correspond to the timing and number of observations collected.
For a single endpoint, the primary analysis may focus on the outcome measured at that time, potentially adjusting for baseline characteristics as prespecified.
For repeated observations, mixed-effects models, generalized estimating equations, survival methods, or other longitudinal approaches may be appropriate depending on the outcome, estimand, assumptions, and study design.
Repeated observations from the same participant are generally correlated. They should not be treated as independent observations without justification.
Researchers should also consider unequal observation intervals, missing visits, and whether measurement frequency is sufficient to identify the intended temporal pattern.
More measurement points do not automatically increase the effective sample size in proportion to the number of observations.
How Can Researchers Decide Whether an Additional Visit Is Worthwhile?
For each proposed visit, identify the information that would be unavailable without it. Determine whether that information is essential to the primary outcome, a justified secondary objective, participant safety, or a meaningful longitudinal analysis.
Then compare its incremental value with the time, travel, cost, fatigue, and operational demands it creates.
If an additional visit is intended to improve precision, its contribution should be evaluated through an appropriate design or simulation analysis rather than assumed.
Where the visit provides little distinct information, researchers should consider removing nonessential measurements or using less burdensome collection methods.
The aim is not to obtain the maximum possible number of observations. It is to obtain enough information to answer the research question reliably.
What Should a Follow-Up Schedule Include in the Research Protocol?
A protocol should identify the timing and purpose of each assessment, the procedures involved, expected duration, acceptable windows, data collection mode, and relevant safety requirements.
For randomized trials, SPIRIT 2025 provides a structured approach to documenting participant timelines and data collection procedures. Its recommendations concern transparent protocol content rather than prescribing a universal visit frequency.
Researchers should also describe plans for participant retention, handling missed visits, and collecting relevant outcomes after intervention discontinuation where appropriate.
Material changes to an approved follow-up schedule may require ethics, institutional, sponsor, or regulatory review before implementation.
Watch Out
Do not add follow-up visits merely because more data seem desirable. Every assessment should have a defined scientific or safety purpose. Equally, do not eliminate necessary visits solely to reduce costs when doing so would compromise outcome ascertainment, longitudinal inference, or participant protection.
