On this page
Ask a direct-to-patient measurement vendor four things first: where each step event comes from, which fields it carries, what your team receives, and what stops anyone from tying a report back to a patient. Those answers decide whether a drugmaker's program can see paid fills and refills by campaign and partner without handling patient identities. Below are 12 questions for any vendor, what a strong answer includes, and how Curve Compliance answers the ones its method covers.
Book a call. Comparing measurement options for a direct-to-patient program? Curve's team can walk your marketing, analytics and privacy leads through the method, from partner step events to aggregate results by channel, campaign and partner. Book a call with Curve.
Why does a direct-to-patient program need its own measurement?
Because the usual data stops at the handoff. When a patient pays cash through a manufacturer's direct-to-patient platform, the prescription may never become an insurance claim. HealthVerity 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” (HealthVerity). IQVIA's June 2026 obesity evidence brief lists “Poor insight into patient access routes, including out-of-pocket purchasing, online pharmacies” among the places traditional data falls short (IQVIA).
Industry groups see the same gap. In April 2026 the Digital Medicine Society launched an initiative with leading pharma organizations that will “define how to measure patient access and affordability” (DiMe release). Its CEO wrote in December 2025 that the direct-to-patient pathway “is expanding far faster than the oversight, standards, and evidence needed to keep pace” (pharmaphorum). Until shared standards exist, the questions your team asks a measurement vendor become the standard for your own program.
Which questions should you ask a direct-to-patient measurement vendor?
Use these 12 in a first call or a request for proposal, and ask for each answer in writing, with a sample report or field list where one exists.
| # | Question to ask | What a strong answer includes |
|---|---|---|
| 1 | Where do the step events come from? | The partner's own systems send them, through a webhook or a file, not a browser pixel on the partner's site. |
| 2 | Which fields does each event carry? | A written field list per event: the step, the time and the campaign that brought the patient, with no name or contact details. |
| 3 | How does the ad click survive the handoff to a partner? | A clear way to keep the click attached from the brand site to a partner's intake, with later events still matching. |
| 4 | What does our team receive? | Aggregate results by channel, campaign and partner, with no patient-level records needed to answer budget questions. |
| 5 | What is the smallest group a report will show? | A minimum cell size, written down, with counts below it suppressed or combined. |
| 6 | How precise are the dates in our reports? | Coarse reporting windows, such as weeks or months, rather than exact dates next to small counts. |
| 7 | What do the contracts say about re-identification? | Terms that bar every recipient from re-identifying patients or linking reports to other data it holds. |
| 8 | What reaches each ad platform, and in which fields? | A fixed field list per platform, neutral event names, contact identifiers off unless you choose otherwise, and health details stripped first. |
| 9 | Can we see what was sent and what each platform accepted? | A log of every event sent to each platform and the platform's response. |
| 10 | How is cost per paid fill calculated? | Ad spend next to paid fills per campaign, the list of fills behind any number, and unmatched fills on their own line. |
| 11 | Can retention be reported by partner? | Refills at month 3 and month 6 by partner, in aggregate, with cohorts compared like for like. |
| 12 | Who signs the BAA, what security report backs it, and who does the setup? | A signed BAA, a current SOC 2 Type II report, and a named team that connects each partner. |
For a wider view of attribution tools built for healthcare advertisers, see HIPAA-compliant attribution software for healthcare and how Curve compares.
What should you ask about the data that comes in from partners?
Questions 1 to 3 decide whether the numbers can be trusted at all. A program with several telehealth and pharmacy partners needs every partner to send the same steps with the same definitions: a visit, a consult, a prescription, a paid fill and, later, each refill. If one partner counts a consult when it is booked and another when it is completed, the partner columns stop being comparable. Ask to see the field list for each event before any partner connects.
The handoff is the weak point. A patient clicks an ad, lands on the brand site and moves to a partner's intake on another domain. If the click does not travel with them, every later step arrives without a campaign, and cost per paid fill can't be computed for that patient. Ask how the click is carried across and whether a paid fill recorded days later still matches the original ad click.
What should the drugmaker team receive?
The safest design is one where your team never needs patient-level records to answer its questions. Results in aggregate by channel, campaign and partner answer the budget questions: which campaigns produce paid fills, what each paid fill costs, and how paid-fill rates and refills differ by partner.
Aggregates still carry risk when groups are small or dates are exact. HHS's de-identification guidance advises experts “to consider how data sources that are available to a recipient of health information [...] could be utilized for identification of an individual” (HHS, content last reviewed February 3, 2025). The same logic applies to any report: a group of three patients with an exact start date is far easier to recognize than a few hundred over a quarter. For background on the terms, see de-identified, anonymized and aggregated data.
So ask for three practices in writing, from the vendor and from each partner: a minimum cell size below which counts are suppressed or combined, reporting windows of weeks or months rather than exact dates, and contract terms that bar re-identification by anyone who receives a report. Where those terms belong in your partner and vendor contracts is a decision for your privacy and legal teams, so talk to your counsel before the first partner connects. This page is general information, not legal advice.
What should a vendor send to Meta, Google and other ad platforms?
A direct-to-patient program runs two separate data flows. Step events from partners feed your reporting. Conversion events sent to ad platforms from your brand sites are a different decision, with their own field list. Meta checks what it receives, and Stape notes that “These checks apply to both Meta Pixel and Conversions API, and Conversions API does not bypass the data sharing rules” (Stape, updated March 3, 2026). Ask which events reach each platform, under which names and with which fields, and how health details in URLs, event names and custom fields are caught before anything leaves. For why moving tags to a server is only part of the answer, see server-side tracking alone is not HIPAA compliance.
Then ask for the evidence: a log of each event sent and whether the platform accepted it, so your privacy team can check what left.
How should cost per paid fill and retention be reported?
Cost per paid fill is ad spend divided by the paid fills credited to a campaign. It needs both halves in one place: spend from the ad platforms and fills from the partners, joined at the ad click. Ask which platforms' spend the vendor pulls in, and which attribution models it offers. Fills that can't be tied to a click should appear as their own line, not spread across campaigns.
Retention is the second number your team will want. Refills at month 3 and month 6, by partner and in aggregate, show what happens after the first fill. Compare cohorts that started in the same window, on the same product and form.
How Curve Compliance helps
Here is how Curve's method answers the questions it covers. In this method, each telehealth or pharmacy partner sends a stripped-down event at each step (visit, consult, prescription, paid fill) carrying the campaign that brought the patient, with no name or contact details. Partners send those events through an incoming webhook or a file upload, and Curve credits each to the original ad click. Curve keeps the ad click attached when a visitor leaves your site for an intake step on another system, and outcomes recorded days later still match; common booking tools are built in, and other systems are connected through setup by Curve's team.
Your team receives aggregate results by channel, campaign and partner, including cost per paid fill, and refills at month 3 and month 6 can be reported by partner, in aggregate. Campaign Reporting puts Meta and Google Ads spend next to Curve-tracked conversions per platform and campaign, with five attribution models. Partners connect once and see their own results, and the partner tools are open to every partner in the program.
On your brand sites, Curve replaces browser ad pixels with one script and sends conversions server-side to Meta Conversions API, Google Ads, TikTok, Microsoft Advertising and LinkedIn. Each platform receives only a fixed list of fields, contact identifiers are off by default and SHA-256 hashed when enabled, and events can use neutral names. Curve strips PHI before data reaches an ad platform and flags PHI-like patterns so they can be stopped at the source, and Event Logs show what Curve sent to each platform and what the platform accepted.
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. Questions 5 to 7 are practices to agree in writing with every vendor and partner, so bring your privacy team's thresholds to the conversation. Book a call to walk through the 12 questions with Curve's team.
See how Curve Compliance answers these in direct-to-patient pharma attribution.
Why minimum cell sizes and re-identification bans matter: HIPAA expert determination for marketing data.
For the questions applied to a health-data measurement provider, see Veeva Crossix and direct-to-patient cash-pay fills.
Sources, checked November 20, 2026: HealthVerity, January 14, 2026; IQVIA insight brief, June 2026; DiMe release, April 16, 2026; pharmaphorum, December 2, 2025; HHS de-identification guidance, last reviewed February 3, 2025; Stape, updated March 3, 2026.
Frequently Asked Questions
What should you ask a direct-to-patient measurement vendor?
Ask where each partner step event comes from, which fields it carries, what your team receives, and which safeguards stop re-identification. Then ask how cost per paid fill and refills by partner are calculated, what reaches each ad platform, and who signs the BAA.
What is a direct-to-patient measurement vendor?
A company that connects a drugmaker's advertising to the steps patients take with its telehealth and pharmacy partners, such as consult, prescription and paid fill, and reports the results.
Why can't claims data measure a direct-to-patient program?
Cash purchases often never become insurance claims. HealthVerity notes that patients in these programs “can disappear from closed payer claims data even though treatment has not stopped” (HealthVerity), so the measurement has to come from the partners.
Should a drugmaker receive patient-level data from its partners?
The budget questions (which campaigns produce paid fills, and how partners differ on paid-fill rate and refills) can be answered with aggregate results by channel, campaign and partner. Ask vendors to design for aggregates from the start.
What is a minimum cell size?
The smallest group a report will show; counts below it are suppressed or combined so a number can't point to one person. The right threshold is one to agree with your privacy team.
Can partners see their own results?
In Curve's method, yes. Partners connect once and see their own results, and the partner tools are open to every partner in the program.
Does Curve Compliance sign a BAA?
Yes. Curve signs a BAA on every plan and holds SOC 2 Type II.
What are the best marketing attribution platforms for a pharma direct-to-patient program?
Look for one that takes step events from your partners, joins them to Meta and Google Ads spend at the ad click, and reports in aggregate by channel, campaign and partner. Put the 12 questions above to each one, Curve Compliance included.
Talk to Curve Compliance
Book a call. Evaluating vendors for a direct-to-patient program? Bring these 12 questions to a call, and Curve's team will walk through how partner step events, aggregate results by channel, campaign and partner, and Meta and Google Ads spend come together to show cost per paid fill by campaign. Book a call with Curve.