Erectile Dysfunction Treatment Selection: Daily Dosing Against an Unpredictable Schedule
A single patient with straightforward erectile dysfunction and a job that makes planning around a pill genuinely difficult. The disagreement has nothing to do with whether a PDE5 inhibitor will work — it's about which dosing strategy a schedule like his can actually sustain, a question no trial was built to answer.
Ray H., a 52-year-old self-employed long-haul truck driver, has never taken anything for erectile dysfunction before today — an IIEF-5 score of 15 places him in the mild-moderate range, with no cardiac history, no nitrate use, and nothing on his labs to suggest a cause beyond ordinary age-related change. The pharmacology of his options isn’t in question; every PDE5 inhibitor on the market would likely help him. What actually complicates the visit is his schedule: home for one day, then gone for four, then home for three, with no pattern regular enough to plan a pill around the way a man with a predictable week could.
That constraint is doing more work in this decision than any trial data can, because the trial data that would resolve it doesn’t exist — daily tadalafil, on-demand tadalafil, and the shorter-acting agents have never been compared head-to-head on the outcomes a patient like Ray actually weighs: spontaneity, cost, and how many doses go unused because the moment never lined up with the plan. Daily low-dose tadalafil maintains a steady state that removes timing from the equation entirely, at the cost of continuous exposure and, historically, a higher price tag than on-demand generic sildenafil — a gap that has narrowed considerably since tadalafil itself went generic. On-demand sildenafil, by contrast, asks him to plan around its own labeled onset window of roughly thirty to sixty minutes and the label’s caution that a high-fat meal delays absorption, a workable ask for a man with a fixed schedule and a harder one for a man who doesn’t know which nights he’ll be home until the day arrives. Vardenafil sits between the two on half-life without solving Ray’s actual problem, which was never about potency — every agent under discussion would likely restore function adequately; what differs is how much advance planning each one silently demands of a man whose week refuses to hold still long enough to plan around.
In clinic, choosing a first agent
Start daily tadalafil, 5 mg, low dose. His whole problem with on-demand dosing is that his schedule doesn't give him thirty minutes of advance notice most weeks — daily dosing removes that requirement completely, which is exactly the lifestyle fit this decision should be optimizing for.
Mechanistically I don't disagree, but he's treatment-naive, and the more conservative sequence is usually to start on-demand and step up to daily only if the timing problem proves as disruptive in practice as it sounds in the room. There's also a real cost question — daily dosing used to run meaningfully more expensive than generic on-demand sildenafil on a high-deductible plan like his.
The cost gap you're describing has narrowed a great deal since tadalafil itself went generic — it's not the same math it was when daily dosing meant paying brand-name prices every month.
I'd weigh what he's already told us more than a general conservative-first default that has no trial behind it either. He's not choosing between two similarly-easy options — he's told us plainly that timing a dose to his schedule has been the actual problem, and there's no evidence that starting conservative and stepping up serves him better than matching the lifestyle need from the first prescription.
Agreed: start daily generic tadalafil 5 mg given his schedule, with on-demand generic sildenafil identified explicitly as the fallback if the daily cost proves prohibitive once his pharmacy benefit runs the actual claim.
Not fully settled: whether starting with the more familiar, lower-total-exposure on-demand agent would have been the more conventional first move for a treatment-naive patient. The pharmacologist would have defaulted to it absent Ray’s specific schedule constraint; the group agreed the constraint was concrete enough to skip that default step rather than treat it as the safer starting point.