Statistical technique that fills in missing values multiple times to account for uncertainty.

Pathway for devices intended to treat conditions affecting fewer than 8,000 patients per year.

FDA document summarizing the basis for PMA approval.

Independent committee adjudicating whether clinical events meet endpoint definitions.

Probability of falsely concluding a treatment works when it doesn't (false positive). Typically controlled at 0.05 (5%) or 0.025 (2.5%, one-sided).

Independent committee that reviews interim data and can recommend stopping a trial for safety, efficacy, or futility.

Documented procedures ensuring consistency and compliance.

Number of participants needed in a clinical trial to reliably detect a treatment effect.

Prior that provides some regularization but is dominated by observed data.

CDISC standard for analysis-ready datasets.

Standardized criteria for assessing tumor response in oncology.

Document summarizing clinical and non-clinical data about an investigational product.

Document explaining trial procedures and risks that participants sign before enrolling.

FDA clearance pathway for devices that are "substantially equivalent" to a legally marketed predicate device. Less rigorous than PMA but still requires careful statistical planning.

Time-to-event analysis accounting for events that preclude the outcome of interest.

Participants who cannot be contacted or assessed during the trial.

Good Clinical Practice is an international ethical and scientific quality standard for the design, conduct, recording, and reporting of clinical trials and other research involving human subjects, ensuring that participants' rights, safety, and welfare are protected and that trial data are credible and reliable.

Adverse event that results in death, is life-threatening, requires inpatient hospitalization or prolongation of existing hospitalization, results in persistent or significant disability/incapacity, is a congenital anomaly/birth defect, or requires intervention to prevent permanent impairment. SAEs must be reported to FDA within specific timeframes (often 24 hours for fatal/life-threatening events). Higher SAE rates in treatment groups can halt trials or block approvals. BSC's safety analysis experience helps clients present safety data in context and identify concerning signals early.

Computing system with documented evidence of reliability and accuracy.

Trial designed to show that a new treatment is not meaningfully worse than an existing treatment (rather than proving superiority).

​

Documented evidence that a process, system, or method consistently produces expected results.

CBER (Center for Biologics Evaluation and Research) regulates biological products, including vaccines, blood and blood components, allergenics, tissues, and cellular and gene therapies, to ensure their safety, purity, potency, and effectiveness. Through this regulation, CBER works to protect and enhance public health, including addressing emerging infectious diseases and bioterrorism threats related to biologic products.

Achievement of pre-defined imaging criteria (e.g., bone fusion on CT).

Compilation of efficacy data across multiple studies.

Frequentist methods are statistical approaches that interpret probability as the long‑run frequency of events under repeated sampling, and they base inference on the sampling distribution of estimators (for example, using p‑values, confidence intervals, and hypothesis tests). In clinical trials, frequentist methods typically control error rates such as type I and type II error using fixed decision rules, without assigning prior probabilities to parameters as in Bayesian methods.

Statistical approach that incorporates prior information and expresses results as probability distributions. A BSC specialty, increasingly accepted by FDA.

Trial designed to show that two treatments have similar effects.

Pre-specified analysis of accumulating data before study completion, typically to assess stopping for efficacy, harm, or futility. Interim analyses must be carefully planned: timing (information fraction), what will be analyzed, decision criteria, and who remains blinded. Key challenge: multiple looks at data inflate Type I error—requires alpha-spending function or Bayesian monitoring boundaries. Number and timing of IAs trade information vs. flexibility. DSMB typically reviews unblinded interim data while the sponsor remains blinded. BSC designs interim analysis strategies, balancing early stopping benefits against statistical and operational complexity.

Questionnaire measuring neck pain-related disability.

Electronic form for collecting clinical trial data. eCRFs are now more common than paper-based CRFs.

Type A meetings are reserved for situations where a development program is stalled or facing a critical regulatory or safety problem, and a meeting is “necessary for an otherwise stalled product development program to proceed or to address an important safety issue.” Examples include resolving clinical holds, major disputes, or other serious roadblocks that prevent further development. These meetings are prioritized and are generally scheduled within about 30 days of FDA’s receipt of a complete meeting request.

Placebo-like control for device or surgical trials where participants undergo a simulated procedure.

Smallest treatment effect that would be meaningful to patients or clinicians.

Approach to missing data that models outcomes separately for different missingness patterns.

System for collecting clinical trial data electronically.

Synonym for pivotal trial; trial designed to support regulatory registration/approval.

Standardized terminology for coding adverse events.

Departure from the approved trial protocol.

Pre-specified criteria for terminating a trial early (for benefit, harm, or futility).

Evaluators unaware of treatment assignment when assessing endpoints.

Comprehensive document detailing all planned statistical analyses. A core BSC deliverable.

Questionnaire measuring low back pain-related disability. Very common in spine trials reviewed by CDRH.

Point when clinical trial database is finalized and no further changes are allowed.

Comprehensive support for FDA Advisory Committee (AdCom) presentations. BSC has extensive panel experience, including preparing statistical presentations, briefing documents, and speaker materials. Our team includes experts who have presented to FDA panels, served on Mock panels, and can provide an insider perspective on what committee members need to see.

Trial where different groups receive different treatments simultaneously.

Evaluation weighing a treatment's benefits against its risks.

Gold-standard design where participants are randomly assigned to treatment or control groups.

Outcomes reported by someone other than the patient or clinician.

​

Additional treatment provided to participants not responding to study treatment.

FDA letter indicating that a submission cannot be approved in its current form. BSC helps clients navigate CRLs with creative re-analysis and redesigned studies.

Centralized laboratory for consistent assessment of images, biomarkers, or other measurements.

Causal inference method for handling post-randomization variables like protocol adherence.

Missing data mechanism where missingness depends on the unobserved values themselves (more problematic).

​

Sensitivity analysis determining how extreme missing data would need to be to change trial conclusions.

CDISC standard for organizing raw clinical trial data.

Probability of achieving statistical significance at the end of a trial, given interim data.

Probability of observing results as extreme as those seen if there were no true treatment effect. Smaller p-values suggest stronger evidence against the null hypothesis.

The ITT analysis set includes all randomized participants, analyzed according to the treatment group to which they were originally assigned, regardless of whether they received treatment, adhered to the protocol, or completed follow‑up. This approach preserves the benefits of randomization and aims to reflect real‑world effectiveness by avoiding bias introduced by excluding non‑adherent or withdrawn participants.

Pre-specified success threshold derived from historical data, used in single-arm studies.

ICH E6 is the International Council for Harmonisation (ICH) guideline on Good Clinical Practice (GCP), setting the ethical and scientific quality standard for the design, conduct, monitoring, recording, and reporting of clinical trials involving human participants to protect their rights, safety, and well-being while ensuring reliable trial data.

FDA approval required before conducting a clinical trial with a significant-risk investigational device in the U.S.

Process of recruiting and registering participants in a clinical trial.

Using data from previous studies as a comparison group (when concurrent controls aren't feasible).

Case report form with annotations mapping fields to SDTM variables.

FDA determination that a device is substantially equivalent to a predicate.

Standardized format for regulatory submissions accepted by FDA, EMA, and other agencies.

A first-in-human study is a type of early clinical study in which a device for a specific indication is evaluated in human subjects for the first time, often as part of an early feasibility study under an IDE. These studies focus on initial clinical safety and device functionality in a very small number of participants before larger feasibility or pivotal trials are undertaken.

Biomarker or measurement used as a substitute for a clinical outcome (e.g., tumor shrinkage as surrogate for survival).

Assumption that there is no difference between treatment and control.

Centralized, independent review of imaging to reduce variability and bias.

​

The Pharmaceuticals and Medical Devices Agency (PMDA) is Japan’s national regulatory authority responsible for ensuring the safety, efficacy, and quality of pharmaceuticals, medical devices, and related medical products.

21 CFR Part 11 is a U.S. FDA regulation that defines the criteria for determining when electronic records and electronic signatures are trustworthy, reliable, and equivalent to paper records and handwritten signatures, and establishes requirements for system validation, security, audit trails, and controlled access in FDA‑regulated environments.

Magnitude of the difference between treatment and control that the trial is designed to detect.

Main outcome measure that the trial is designed and powered to evaluate. Drives the primary efficacy conclusion.

Public meeting where independent experts advise FDA on approval decisions. High-stakes; BSC has extensive panel preparation experience.

Specific adverse event term in MedDRA.

A Modified Intent‑to‑Treat (mITT) analysis set is a prespecified variation of the ITT principle in which some randomized participants are excluded based on additional criteria, such as not receiving any study medication or lacking key baseline or post‑baseline measurements. mITT sets are widely used but heterogeneous; they can introduce bias if exclusions are subjective or not clearly justified, so regulatory and methodological guidance recommends that any modifications be minimal, explicit, and scientifically defensible.

Prior distribution expressing minimal prior knowledge, letting data drive conclusions.

The Full Analysis Set is the primary analysis population defined in ICH E9 as being as complete and as close as possible to the intention‑to‑treat (ITT) ideal, typically including all randomized participants with minimal exclusions. Limited exclusions may be allowed for clear, pre‑specified reasons such as failure to meet major entry criteria, never receiving any study treatment, or having no post‑randomization data.

Follow-up is the period after a participant is enrolled or completes an intervention during which investigators continue to collect outcome and safety data, often at predefined visits or time points, to assess both short‑term and long‑term effects. It can also refer to the set of activities (such as visits, tests, and contacts) used to maintain contact with participants and obtain these data over time.

The most rigorous FDA pathway for Class III (highest risk) medical devices. Requires clinical trial data demonstrating safety and effectiveness. BSC frequently supports PMA submissions.

Statistical analysis of sufficient quality and documentation for regulatory submission.

Range of values within which the true treatment effect likely falls (typically 95% CI).

FDA program providing intensive interaction and guidance for devices that offer significant advantages over existing treatments.

Method for controlling Type I error across multiple interim analyses (e.g., O'Brien-Fleming, Pocock).

Visual acuity testing protocol used in ophthalmology trials.

Comprehensive document describing a clinical trial's design, methods, results, and conclusions. Required for regulatory submissions.

CDER (Center for Drug Evaluation and Research) ensures that safe and effective drugs are available to improve the health of people in the United States, overseeing both prescription and over-the-counter drugs, including many biologic therapeutics and generics. It evaluates drugs before marketing to determine whether their health benefits outweigh known risks and monitors drug safety, quality, and labeling throughout the product lifecycle.

A futility analysis is an interim evaluation in an ongoing clinical trial that assesses whether it is unlikely the study will achieve its primary objective if it continues as planned. If the accumulating data show a very low probability of ultimately demonstrating the targeted treatment effect (for example, low conditional or predictive power), the trial may be stopped early for futility to avoid exposing participants to an ineffectual intervention and to conserve resources.

Prior distribution incorporating meaningful prior knowledge (from historical data, literature, etc.).

Single endpoint combining multiple outcomes (e.g., "death, MI, or stroke"). BSC has deep expertise in composite endpoint design.

Biostatistical consulting focused specifically on regulatory approval pathways. Unlike academic or general biostatistics, regulatory biostatistics requires deep knowledge of FDA expectations, submission standards, and proven methodologies. BSC's approach ensures statistical strategies align with regulatory requirements from study design through approval.

Computationally constructed control group using external/historical patient data.

The probability that a trial will achieve statistical significance if there truly is a treatment effect of a specified size. Power = 1 - β (Type II error). 80% power means an 80% chance of detecting the effect if it exists (but 20% risk of false negative). Higher power requires larger sample sizes. FDA typically expects ≥80% power, often 90% for pivotal trials. Inadequate power wastes resources and subjects—studies may fail to detect real benefits. BSC's sample size calculations balance adequate power with feasibility, explicitly considering effect size uncertainty and dropout rates.

Analyses testing whether conclusions are robust to different assumptions or methods.

​