The Hidden Costs Of Clinical Trial Technology Nobody Budgets For
By John Oncea, Chief Editor, Clinical Tech Leader
When a site decides to buy an EDC system, the number that gets scrutinized is the purchase price. Liam Devenish thinks that’s the least useful number in the whole decision. Devenish spent nine months as a sub-investigator on clinical trials at Wits RHI in Johannesburg before moving into healthcare financing and economics research at the London School of Economics, a combination that gives him a genuinely different vantage point on what “the cost of technology” actually means in a clinical trial setting.
Integration Is The First Question, Not An Afterthought
Asked how organizations should think about the value of clinical trial technology beyond simply purchasing a tool, Devenish didn’t start with cost; he started with fit. “Does the technology that’s deployed integrate well into the existing and broader healthcare system?” he said. “Does the technology feed into, grow, or improve the existing system capacity that is there?”
That’s a deliberately broader question than whether a tool works for the trial it’s bought for. A platform that performs well in isolation but has no life outside the trial – no institutional utility once the study ends – is a different kind of investment than one that strengthens infrastructure the site will use again. “Outside of a clinical trial, you’re not going to have that data available anyway, so it’s not really a data repository,” Devenish noted of tools built purely for one study’s needs. The distinction matters for anyone deciding whether a purchase is a one-time expense or a durable asset.
What Gets Left Off The Line Item
Pressed on which costs organizations most commonly underestimate, Devenish went straight to training and was clear that it isn’t a one-time expense. “There’s an extremely high requirement for training, ongoing training. It’s not a one-off,” he said. “Anytime you have any changes to the platform, you need refresher training, and you need training in the tool itself as well as the meta level of GCP or whatever the external regulatory requirements are.”
Beyond training, he pointed to costs that read as almost too basic to need stating, which is exactly why they get missed. Reliable infrastructure – computers, internet access, continuous electricity – is a real line item, not a given, and in South Africa specifically, uninterrupted power “is actually a problem.” Security requirements for housing the technology add further overhead. None of these show up on a vendor’s price sheet, but all of them determine whether a technology purchase actually functions as intended once it’s live.
The payoff for absorbing those costs, in Devenish’s framing, is that they’re front-loaded rather than recurring. “Once you have people who are trained and comfortable with a tool, you can then repeat the process,” he said. “Once you’ve made that initial investment, both in the infrastructure and the processes and the people, then it’s much easier” – a lower barrier for the next research effort in that same environment.
Two Kinds Of Efficiency, And A Third That Gets Skipped
When it comes to measuring whether a technology investment actually delivered value, Devenish separates the question into two distinct economic concepts that rarely get named explicitly in clinical operations conversations.
The first is allocative efficiency. “For a certain amount of money, are we putting this money in the right place to get the correct results?” said Devenish. That requires a genuine economic assessment, not just the sticker price of a platform, but the opportunity cost of the money and time spent, including the value of training and the value of staff hours.
The second is technical efficiency: once you already have a technology, “how do you ensure that you get the biggest bang for buck” from it? Essentially, are you extracting the maximum output the tool is capable of, relative to its inputs?
The piece organizations most often skip, Devenish argues, is quality. “An important element of efficiency is that something is efficient. A technology is efficient in so far that it achieves the correct aim at a reasonable price,” he said. “If the unit output that you have is low quality, you haven’t increased the efficiency.” He pointed to a familiar pattern in healthcare broadly: cutting the quality of a service in pursuit of financial savings, which nets out to no efficiency gain at all: “probably net neutral.” Technology that’s cheaper and faster but produces worse data isn’t efficient. It’s just cheaper.
Who Should Actually Own The Evaluation
Devenish is direct that this isn’t a job for a single stakeholder. “Whoever’s involved in the design of a tech implementation needs to consider the efficiency assessment from the outset,” he said, and the evaluation has to run the full length of the chain: the people using the tool day to day, the people who built it, the payer funding it, and a governance layer above all of them assessing whether the tool is still serving the purpose it was procured for. “In a clinical research setting, it would be every member of the team that’s interacting with the tool” who has a stake in that assessment plus, at the far end, “the clinical leader or whoever it is” evaluating outcomes.
The Real Question Before Any Purchase
Strip away the specific line items, and Devenish’s framework reduces to a sequence most sites skip entirely: does this integrate with what we already have, what will it actually cost to run once training and infrastructure and maintenance are counted, and are we measuring quality alongside speed and price and not instead of them. None of that shows up in a vendor demo. All of it determines whether the technology was worth buying.
This is a companion piece to What South Africa’s Trial Sites Teach Global Clinical Research, which explores how infrastructure constraints reshape the same cost calculus.