On this page

To measure a direct-to-patient program, follow each patient from the ad to a paid fill and refills using step events that the telehealth and pharmacy partners report, because cash-pay fills often never reach insurance claims. The method has six steps: define the funnel, agree the events and fields with each partner, report the match rate first, compute cost per patient start and cost per paid fill by campaign, compare partners on paid-fill rate and speed, and track month-3 and month-6 refills in aggregate. Curve Compliance supports this method: it keeps the ad click attached across the handoff to a partner's intake step, takes in the step events partners report, and puts Meta and Google Ads spend next to them, so the drugmaker receives aggregate results by channel, campaign and partner.

Book a call. Planning how a direct-to-patient program will be measured? Curve's team does the setup for the click matching, the partner events and the campaign reporting, typically live in about a week, case by case, under the BAA Curve signs on every plan. Book a call with Curve.

Why can't claims data measure a direct-to-patient program?

Because many cash-pay fills never become claims. HealthVerity, a real-world data company, put it plainly in January 2026: “Patients who continue therapy through direct-to-consumer programs can disappear from closed payer claims data even though treatment has not stopped.” The same post warns that “Coverage loss can appear as treatment discontinuation” (HealthVerity).

IQVIA's June 2026 insight brief on obesity evidence lists “poor insight into patient access routes” among the limits of commonly used real-world data, naming out-of-pocket purchasing and online pharmacies (IQVIA, June 2026). For a program that runs through a telehealth visit and a pharmacy partner, the facts the brand needs (who started, who filled, who refilled) sit with the partners. Jennifer Goldsack, CEO of the Digital Medicine Society (DiMe), wrote in December 2025 that this pathway “is expanding far faster than the oversight, standards, and evidence needed to keep pace” (pharmaphorum). DiMe's initiative with leading pharmaceutical companies lists, among its goals, to “define how to measure patient access and affordability” (DiMe). Measurement has to be designed in, at the partner, from the start.

Step 1: what funnel should a direct-to-patient program define?

Write the funnel down before launch, with one definition per step that every partner uses. A typical cash-pay path has six steps:

StepWho records itWhat the event should carry
Ad click and site visitThe brand siteChannel, campaign and the ad click
Visit startedTelehealth partnerThe campaign that brought the patient, the date
Consult completedTelehealth partnerCampaign, partner, date
Prescription writtenTelehealth partnerCampaign, partner, date
Paid fillPharmacy partnerCampaign, partner, fill date
Refill at month 3 and month 6Pharmacy partnerCampaign, partner, refill number

Agree what each step means, too. Is a paid fill the day payment clears or the day the pharmacy ships? Is a start the completed consult or the first paid fill? Pick one definition per step and hold every partner to it, or the same patient will count differently at different partners. Leave out anything the math doesn't need: dose, diagnosis and free text add risk without adding a metric.

Step 2: what should each partner agree to report?

Put a short data agreement in place with each partner. It covers four things: the events (the steps above, with the same names everywhere), the fields on each event, how they arrive (an incoming webhook as each step happens, or a scheduled file upload), and how often. A daily feed is a sensible default; the faster the feed, the sooner a broken handoff shows up.

Keep the fields minimal on purpose. The step, the date and the campaign that brought the patient are what the analysis needs, with no name or contact details. Before you sign, ask how small groups will be handled in reporting. A minimum group size below which numbers are withheld, and coarse time windows such as weeks or months rather than exact dates, both lower the chance that a reported number points to one person; small-group rules explains how those rules work. Data-sharing terms between a drugmaker and its partners raise legal questions of their own, so talk to your counsel before partner events start flowing.

Step 3: why should you report the match rate first?

Because every later number rests on it. The match rate is the share of partner events that tie back to an ad click or a tracked site visit. If one partner's events match well and another's barely match, the second partner will look worse on every cost metric for a reason that has nothing to do with patients.

Report the match rate by partner and by week, before any cost metric. The usual causes of a low rate are a click lost when the visitor moves from the brand site to the partner's intake page, a partner event that lacks the campaign field, and two sides defining a step differently. Fix the plumbing, then read the costs. The attribution window matters as well: a refill at month 6 sits far outside a 7-day click window, so set the window to fit the journey (attribution windows for healthcare).

Step 4: how do you compute cost per patient start and cost per paid fill?

Divide ad spend by the outcomes it produced, campaign by campaign. Patient starts are the industry's own yardstick: in DiMe's April 2026 release, PHIL's Josh Zeidman said manufacturers that run these programs well “will see the impact where it matters most: patient starts, adherence, and long-term therapy outcomes” (DiMe release). The core metrics, each read by channel, campaign and partner:

MetricHow to compute itWhat it tells you
Match ratePartner events tied to an ad click, divided by all partner eventsWhether the cost numbers can be trusted
Cost per patient startAd spend divided by patients who started, by your program's definition of a startWhat a new patient on therapy costs, by campaign
Cost per paid fillAd spend divided by all paid fills credited to the campaign in the periodSpend efficiency through the pharmacy step
Paid-fill ratePaid fills divided by prescriptions written, by partnerWhere patients stall between prescription and pharmacy
Time to first fillDays from completed consult to first paid fill, by partnerFriction at the pharmacy handoff
Month-3 and month-6 refill ratePatients with a paid refill at month 3 or month 6, divided by patients who started in that cohortWhether a campaign brings patients who stay

Spend comes from the ad platforms and outcomes come from the partners, so keep both in one place where they line up by campaign and by date. Count first fills and refills separately so the two cost metrics stay distinct. The clinic version of the same arithmetic is in how to calculate cost per started patient.

Step 5: how should you compare partners on paid-fill rate and speed?

On rates, not counts. A partner that receives more traffic will have more fills; the useful question is what share of its prescriptions become paid fills, and how fast. In a March 2026 Pharmaceutical Commerce interview, AssistRx's Jordan Armstrong named speed first, “getting that patient onto therapy as quickly as possible,” and the article points to “conversion rates, refill adherence, and integration with other access and affordability programs” as the measures that show a program's value (Pharmaceutical Commerce).

Compare like with like: the same start month, the same product form and the same payment route. Use partner metrics to fix the program, such as a slow pharmacy handoff or an intake page that fails on mobile, not to move patients toward one provider. Patients keep choosing where they are seen, and prescribing stays with the clinician.

Step 6: how do you track month-3 and month-6 retention without patient records?

Count refills by cohort, in aggregate. Group patients by start month, campaign and partner, then report the share of each group with a paid refill at month 3 and at month 6. No one outside the partner needs a name to do this: the partner reports the step and the campaign; report rates only for groups large enough to share.

Expect cohorts to differ for reasons that have nothing to do with marketing. IQVIA's February 2026 post on direct-to-patient programs cites a 180-day adherence analysis in which commercial patients averaged 155 days on Zepbound® (Eli Lilly) pens, the direct-to-patient vial channel was “comparably persistent” at 141 days, and the cash card cohorts averaged 128 days (Zepbound pen) and 121 days (Wegovy® (Novo Nordisk)) (IQVIA, February 2026). The averages differ by cohort, so compare partners and campaigns within the same cohort, and read month-6 results only once the cohort is old enough.

How Curve Compliance helps

Curve Compliance supports this method. Curve keeps the ad click attached when a visitor leaves the brand site for a booking or intake step on another system. That is built in for common booking tools, other systems are connected through setup by Curve's team, and outcomes recorded days later still match. Telehealth and pharmacy partners can report step events such as consult, prescription and paid fill through an incoming webhook or a file upload, and Curve credits each one to the original ad click. In this method, each event carries the campaign that brought the patient and no name or contact details.

Campaign Reporting puts Meta and Google Ads spend next to Curve-tracked outcomes per platform and campaign, with five attribution models; that is where cost per paid fill by campaign comes from. The drugmaker receives aggregate results by channel, campaign and partner, including cost per paid fill, and month-3 and month-6 refills can be reported by partner in aggregate. Partners connect once and see their own results, and the partner tools are open to every partner in the program.

Curve strips PHI before data reaches an ad platform and flags PHI-like patterns so they can be stopped at the source. Curve signs a BAA on every plan, holds SOC 2 Type II, and its team does the setup, typically live in about a week, case by case. See Curve conversion tracking and Curve for pharma and medical device teams, or book a call to plan the measurement for your program.

To run this across your telehealth and pharmacy partners, see direct-to-patient pharma attribution.

For the reason the cash channel has to be measured at the partner, see why cash-pay patients disappear from claims data.

Step five in more depth: telehealth partner metrics: paid fills and retention.

Before any step event reaches an ad platform, check what a direct-to-patient program can send to ad platforms.

For the partners behind each step, see pharma direct-to-patient programs.

Sources, checked November 20, 2026: HealthVerity, January 14, 2026; IQVIA insight brief, June 2026; pharmaphorum, December 2, 2025; DiMe project page; DiMe release, April 16, 2026; Pharmaceutical Commerce, March 26, 2026; IQVIA, February 10, 2026.

Frequently Asked Questions

How do you measure a direct-to-patient program?

Follow each patient from the ad to a paid fill and refills using step events that the telehealth and pharmacy partners report, matched to the campaign that brought them. Then read the match rate, cost per patient start, cost per paid fill, paid-fill rate and month-3 and month-6 refills by channel, campaign and partner, in aggregate.

Why does claims data miss direct-to-patient patients?

Cash-pay prescriptions often fall outside pharmacy benefit workflows, so no claim is created. HealthVerity notes that patients in these programs “can disappear from closed payer claims data even though treatment has not stopped” (HealthVerity).

What does the match rate tell you?

How much of each partner's activity can be tied to a campaign at all. Read it by partner and by week before any cost metric; a low or falling rate usually means a click lost at the handoff or a missing campaign field.

What is the difference between cost per patient start and cost per paid fill?

Cost per patient start divides ad spend by new patients, using the step your program defines as a start. Cost per paid fill divides the same spend by every paid fill credited to the campaign, first fills and refills, so it improves as patients stay on therapy.

What data should a telehealth or pharmacy partner share for measurement?

A stripped-down event per step: the step, the date and the campaign that brought the patient, with no name or contact details. Ask for a minimum group size and coarse time windows in reporting, so results stay in aggregate.

How long should you track retention?

At least to month 6. IQVIA's February 2026 post compares cohorts over 180 days, and refill rates at month 3 and month 6 show whether a campaign brings patients who stay (IQVIA, February 2026).

Can Curve Compliance support this method?

Yes. Curve keeps the ad click attached across the handoff, takes in partner step events through incoming webhooks or file uploads, and puts Meta and Google Ads spend next to them in Campaign Reporting.

Talk to Curve Compliance

Book a call. Want cost per paid fill by channel, campaign and partner, read in aggregate? Curve's team does the setup for the click matching, the partner events and Campaign Reporting, typically live in about a week, case by case, under the BAA Curve signs on every plan. Book a call with Curve.