Clinical trial cost per patient, calculated from your own budget

Cost per patient looks like a simple division: total cost divided by number of participants. In practice the answer depends entirely on which participants you divide by, and which costs you include in the top line. A budget of 1,200,000 divided over 60 enrolled participants is 20,000 each. The same budget divided over the 48 who completed the study and have evaluable data is 25,000 each. Both are correct; they answer different questions.

The calculator works in three steps. First it adds your fixed costs: things you pay whether you enrol 10 participants or 100, such as ethics and regulatory submissions, site start-up, project management, monitoring set-up, database build and study software. Second it adds your variable costs: visit payments, procedures, laboratory and imaging fees, participant reimbursements and recruitment spend, each multiplied by the number of participants or visits they apply to. Third it divides the total by each denominator in turn and shows the results side by side.

The three denominators matter because trials lose people at different points. Of everyone you screen, some fail eligibility. Of those who are enrolled, a few withdraw before randomization. Of those randomized, some drop out or are lost before the primary endpoint. Only the rest are evaluable. Every drop adds cost without adding a data point, so the cost per evaluable participant is always the highest of the figures and is the one a funder or sponsor should plan around.

Use the result as a sense check, not as a price. If the cost per evaluable participant looks very different from what comparable studies in your area report, find out why before you sign anything: it may be a long follow-up, an expensive procedure, or a dropout assumption that is too optimistic. Compare like with like: the same denominator, the same cost categories, and the same follow-up length.

Pair this tool with the other planning aids on this site. The sample size calculator gives you the number of participants you need; the dropout-adjusted sample size calculator turns that into the number you must enrol; the timeline calculator shows how long enrolment will take, and time is itself a cost because monitoring and software run per month. For the budget itself, the clinical trial budget template gives you the line items to collect quotes against.

The software line is easy to under-think. A common pattern in small studies is to build the data capture in spreadsheets or a general survey tool to save money, then pay far more in cleaning, query chasing and rework. Capture has a free sandbox with every feature and you pay only once you go live with real participants, so the data-capture line can be tested before it is committed. See low-cost clinical trial software and the EDC cost calculator for that part of the budget.

The calculator is arithmetic on your inputs. It does not know your country, therapeutic area, site mix or contract terms, and it makes no claim about what a trial "should" cost. It does not model inflation, currency changes, protocol amendments, pass-through costs charged at cost plus a fee, milestone-based payment schedules or patient-level cost variation (for example, a few participants needing many unscheduled visits). Real budgets often have a contingency line; the calculator treats contingency only if you add it as a cost.

It also does not replace a feasibility assessment or a quote from sites and vendors. If the study is going to a funder, a regulator or an investor, the budget should be built bottom-up with real quotes and reviewed by someone who has run a comparable study. Where the cost model drives a go/no-go decision, involve a statistician for the sample size and dropout assumptions and a finance or clinical operations lead for the cost side.

Finally, this is a demonstration and planning aid, not validated software. Do not use its output as an audited cost figure in a regulatory submission or grant report without checking the arithmetic yourself.