Evidence standards & calculator methodology
Useful arithmetic. Visible assumptions. No invented certainty.
Hive Prospect separates source facts, explicit assumptions, model inferences and UNKNOWNs. The free tools are designed to expose what controls a decision, not to disguise missing information with an industry-average answer.
Clinical Trial Cost Calculator
The calculator adds only the costs the user enters: patient grants, screen-failure payments, site startup, recurring site/admin cost, CRO fees, central vendors and other pass-throughs. A blank field stays UNKNOWN; it is never silently treated as an industry benchmark.
Core arithmetic: known-cost total = entered patient + screen-failure + site + CRO + vendor + pass-through costs. Cost per enrolled patient = known-cost total ÷ enrollment target.
Enrollment Calculator
The tool makes its assumptions explicit rather than estimating a “typical” recruitment rate.
Screens required: enrollment target ÷ (1 − screen-failure rate).
Recruitment months: enrollment target ÷ (active sites × enrolled patients per site per month).
Total time: recruitment months + entered startup delay.
Expected completers: enrolled patients × (1 − dropout rate).
Protocol Complexity Preflight
The ten-question preflight is a heuristic warning-light screen, not a validated clinical, statistical or regulatory scoring instrument. It asks about visits, countries, sites, procedures, vendors, endpoints, screen failure, participant burden, feasibility evidence and activities that may not clearly support safety or a decision.
Its purpose is to identify areas worth pressure-testing before complexity becomes contracted cost.
Country Comparison
The geography tool compares the same enrollment target across two scenarios using the user’s own site count, startup time, enrollment rate, patient/site costs and expected incentives. It does not assume that an entire country has one price, one startup time or one regulatory value.
A lower arithmetic cost is not automatically the better development decision. Evidence usability, site capability, patient availability and protocol-specific logistics still require review.
Why protocol burden matters
There is published evidence behind the questions.
Tufts Center for the Study of Drug Development research has repeatedly documented growth in protocol complexity. A benchmark analysis of 9,737 protocols and 130,601 investigative-site contracts reported increases in procedures, visits, site work effort and cost measures across Phase I–III studies. A later benchmark paper reports associations between more demanding protocol designs and poorer recruitment, retention, adherence, data quality and cycle-time performance.
CTTI’s quality-by-design framework similarly emphasizes trials that are fit for purpose, with activities essential to participant safety and credible key results, while eliminating nonessential activity where appropriate.
FDA’s final guidance on decentralized clinical-trial elements notes that reducing participant travel and using more convenient local or remote activities may improve engagement, recruitment and retention in suitable trials.
Primary and peer-reviewed references
Sources
- Getz KA, Campo RA. New Benchmarks Characterizing Growth in Protocol Design Complexity. Analysis of 9,737 protocols and 130,601 investigative-site contracts.
- Protocol Design and Performance Benchmarks by Phase and by Oncology and Rare Disease Subgroups. Peer-reviewed review of design burden and trial-performance relationships.
- Improving Protocol Design Feasibility to Drive Drug Development Economics and Performance.
- A Case Study Assessment on the Rationale for, and Relevance of, Non-Core Protocol Data. Tufts CSDD/FDA work examining core versus non-core protocol procedures.
- Clinical Trials Transformation Initiative: Designed with a Quality Approach.
- FDA: Conducting Clinical Trials With Decentralized Elements, Final Guidance.
Evidence discipline
FACT, INFERENCE and UNKNOWN are different things.
FACT means the supporting source or entered value is visible. INFERENCE means Hive Prospect is making a reviewable assessment from those facts. UNKNOWN means the available evidence does not establish the answer. UNKNOWN is not scored as favorable merely to make an opportunity look better.
Calculator output and automated protocol triage are decision-support tools. They do not replace clinical, regulatory, statistical, legal, safety or accounting review.