What actually drives specialty pharmacy access delays

When a health system decides time-to-therapy is too long, the first instinct is almost always to blame prior authorization. It is a satisfying explanation. It is also, most of the time, the smaller half of the problem.

Payer review has a clock on it. Depending on the plan and the request type, that clock is measured in days, and it is one of the few parts of this process with a defined limit. The parts without a clock are where the delay actually accumulates.

Where the days go

Before the referral ever leaves the clinic

A referral that arrives incomplete does not fail. It waits. Missing diagnosis specificity, missing prior therapy documentation, missing weight or lab values, missing the attestation a payer policy requires: each one generates a callback, and each callback costs a day or more because both ends of the call are busy.

This is the single highest-yield place to intervene, and it is usually the least instrumented. Most organizations cannot tell you what percentage of their specialty referrals arrive complete, which means they cannot tell whether it is improving.

Benefits investigation

Before anyone can submit anything, someone has to establish whether the therapy runs through the medical benefit or the pharmacy benefit, whether the patient's plan carves out specialty pharmacy to a specific network, what the patient's out-of-pocket exposure looks like, and whether the site of care is covered.

When this work is done well it takes hours. When it is done as an afterthought, it gets redone partway through the process, and the second pass costs more than the first would have.

The handoff between clinic, pharmacy, and hub

Every organizational boundary the case crosses is a place it can sit. Clinic to financial clearance. Financial clearance to pharmacy. Pharmacy to manufacturer hub. Hub back to clinic for a missing signature. None of these transitions is long on its own. Collectively they are often the largest single block of elapsed time, and they are invisible in most reporting because each function measures only its own turnaround.

A useful diagnostic: ask each team what their average turnaround is, add them up, and compare the sum to the actual referral-to-first-dose interval. The gap between those two numbers is the handoff problem, and it is usually substantial.

Financial clearance happening after scheduling instead of before

When a patient is scheduled before financial clearance is complete, one of two things happens. Either the appointment is kept and the organization takes the financial risk, or the appointment is canceled and rescheduled, which resets the patient's clock and burns a chair slot. Sequencing clearance ahead of scheduling is unglamorous and it removes more delay than most technology projects.

Assistance enrollment treated as a separate process

Patient assistance, copay support, and foundation enrollment are frequently run as a parallel track that starts only after coverage is established. For patients who need them, that sequencing adds a second full cycle. Running assistance screening concurrently with benefits investigation, rather than after it, collapses that.

Denials handled as correspondence

A denial that goes into a work queue behaves very differently from a denial that triggers a same-day peer-to-peer. Many denials are documentation failures rather than coverage decisions, and they are recoverable quickly if someone owns them immediately. When denials are batched, the recoverable ones age alongside the genuine ones.

What to measure

Four measures are usually enough to locate the problem:

1. Referral-to-first-dose, in days, by therapeutic class. The headline number. Segment it, because one class will be dragging the average.
2. Referral completeness rate at intake. The upstream lever.
3. Denial rate and denial reason. Reason matters more than rate. Documentation denials and coverage denials call for different fixes.
4. Abandonment, with the stage at which it happened. A patient who abandons after approval is a different failure than one who abandons during benefits investigation.

These are not hard to build. Most of the data already exists across the electronic health record, the pharmacy system, and the hub portal. The work is in agreeing on definitions and picking one owner for the report.

What usually fixes it

In rough order of effort against effect:

• Define referral completeness and enforce it at intake, with a standard template per class.
• Move financial clearance ahead of scheduling.
• Run assistance screening concurrently with benefits investigation.
• Assign single-case ownership across the whole pathway so the case has a name attached to it at every handoff.
• Create a same-day denial triage path with a clinician available for peer-to-peer.
• Build the four-measure report and review it monthly with clinic, pharmacy, and revenue cycle in the same room.

None of these require new technology, and none of them require a vendor. Most access programs that are underperforming are not missing a tool. They are missing a defined intake standard and a single owner across the handoffs.

Robert Walters, PharmD, MBA is the founder of Walters Rx Strategy and is Apexus Advanced 340B Certified. He spent fifteen years leading health-system pharmacy operations across acute care, a cancer center, infusion, and emergency department pharmacy. Book a consult or read more insights.

This article is general and informational. It is not legal, regulatory, clinical, or medical advice, and it does not recommend or promote any specific product or therapy.