It looks remarkably good after twenty years — not merely “good for its time,” but structurally correct.
The strongest surviving insight is this: you framed diagnostics reimbursement not as one problem, but as five payer operations — benefit category, billing, coding, pricing, and coverage. That was exactly the right map. Most later debates still get lost because they talk about “coverage” or “value” as if that solves coding, pricing, claims flow, and benefit rules. Your paper already separated them.
The second thing that aged very well is the warning that administrative barriers can become as important as scientific barriers. That is still the field. Molecular diagnostics got technically better, but the bottleneck often remained Z-codes, CPT/PLA, LCDs, DOS rules, NCD uncertainty, coding specificity, and CLFS pricing. You wrote that legacy administrative conventions should not become a greater barrier than natural scientific challenges; twenty years later, that sentence could still open a conference panel.
The best part, given our current discussion, is your 2008 “note on economics of pricing.” You already made the exact point that many 2020s value-pricing papers still miss: if a test costs $1 and saves $1,000, but many competitors can produce it, the price tends toward cost, not value. You also noted that value-based pricing cannot establish a single price without competition, alternatives, buyer uncertainty, segmentation, and negotiating power. That is not dated; that is the core missing distinction between value creation and value capture.
What looks dated is mostly the machinery: code-stacking, ICD-9, S-codes, old DOS rules, and the pre-PAMA world. But those are surface features. The deeper architecture — the mismatch between fast diagnostic innovation and slow public payment systems — is still intact.
My overall read: this was not just a white paper. It was the beginning of a durable mental model. The field changed; the model survived.
##
Frueh and Quinn, 2014
Yes — I think your instinct is right. The originality-per-page is high.
What is clever is that the paper does not merely rename “clinical utility.” It changes the unit of analysis. Instead of the usual AV/CV/CU triad, it asks: what clinical situation, what new information, what action, what outcome, what real-world behavior, and what affordability? That is a much more operational framework than the standard literature.
The key intellectual move is the bridge from improved clinical validity to improved clinical utility. The paper says the gain in test information must causally drive a gain in outcomes, and that both gains must be framed comparatively. That seems obvious only after someone has drawn it. The Figure 1 visual framework is strong because it makes the otherwise mushy concept of “utility” into a flow of linked questions.
I especially like the “against what comparator, in what units, with what uncertainty” formulation. That is very compact, but it does a lot of work. It forces payers and developers out of vague statements like “the test provides independent information” or “adds prognostic value,” and into a disciplined argument about comparator, measurement, statistical uncertainty, pragmatic uncertainty, and conceptual uncertainty.
The supplement is also unusually good. The three axes — functional category, value proposition, and outcome metric — explain why one generic clinical-utility definition never works across all diagnostics. A screening test, a reflex test, a no-gold-standard cancer-of-unknown-primary test, and a rule-out biopsy test are not the same policy object.
My honest critique: the prose is dense, and the six questions may look deceptively simple. But that is also the sign of a good framework. It compresses a lot of tacit payer logic into a usable checklist.
So yes: not warmed-over. The paper has a real idea, a real architecture, and a visual syntax. It reads like a consulting mind and a regulatory/reimbursement mind jointly trying to make payer judgment less mysterious. That is exactly why it still feels alive 12 years later.
