The Hidden Hierarchy of Innovation: How Physicians Ranked Pharma, Diagnostics, and Devices in Driving Patient Outcomes (1990-2015)

1. The Core Axis: Why 56% Is More Than a Number—It’s a Strategic Signal

In January 2019, the Journal of Managed Care & Specialty Pharmacy published a physician survey that quantified a hierarchy long suspected but rarely measured. The study, conducted by Wamble, Ciarametaro, and Dubois and funded by the National Pharmaceutical Council, asked physicians with a mean of 21.4 years of clinical practice to attribute post-diagnosis patient outcome gains across eight major conditions to four categories of medical innovation: pharmaceuticals and biopharmaceuticals, diagnostics, medical devices and procedures, and a residual category termed “cannot allocate” (Source 1: [Primary Data - Wamble et al., 2019]).

The result was unambiguous. Physicians attributed 56% of post-diagnosis outcome improvements to pharmaceuticals and biopharmaceuticals. Diagnostics, the second-ranked category, received only 20%. Medical devices and procedures trailed further, with the remainder assigned to public health interventions, lifestyle changes, and system-level factors that physicians could not confidently allocate to any single technology class.

This 56% figure is not merely a descriptive statistic; it is a strategic signal reflecting two converging forces. First, the period 1990-2015 corresponds to the blockbuster era of pharmaceutical innovation—statins for ischemic heart disease, antiretroviral therapy for HIV, targeted biologics for breast cancer and lung cancer, and insulin analogs for diabetes. Second, the attribution pattern reveals a cognitive bias inherent in physician training: physicians are trained to prescribe. The act of diagnosis is a prerequisite, but the act of prescribing is the therapeutic intervention. Diagnostics, by contrast, produce information, not treatment. Their value is distal—enabling earlier intervention—and thus harder for physicians to mentally assign causal weight in outcome improvement.

The eight conditions selected were not arbitrary. The study anchored to the Centers for Disease Control and Prevention’s identification of the conditions with the largest effect on US morbidity and mortality from 1990 to 2014: breast cancer, ischemic heart disease, HIV infection, diabetes, unipolar depression, chronic obstructive pulmonary disease, cerebrovascular disease, and lung cancer (Source 2: [CDC Morbidity and Mortality Data, 1990-2014]). These conditions represent the disease burden that drove healthcare resource allocation, R&D investment, and health policy for a generation.

Image suggestion: Bar chart comparing attribution percentages (pharma 56%, diagnostics 20%, devices/procedures, other, cannot allocate) with small icons for each category.

2. Dual-Track Selection: This Demands a Slow-Analysis Deep Audit

Readers must recognize that this study was published in 2019. Its timeliness as a news event has expired. Its value now lies in what it reveals about the structural assumptions embedded in healthcare value assessment frameworks. This demands a slow-analysis deep audit—not a recapitulation of the survey results, but an interrogation of their implications for current resource allocation.

The data challenges the parity assumptions used by organizations such as the Institute for Clinical and Economic Review (ICER) in the United States and the National Institute for Health and Care Excellence (NICE) in the United Kingdom. These value assessment frameworks typically evaluate drugs, diagnostics, and devices through cost-effectiveness analysis using quality-adjusted life years (QALYs). In theory, these frameworks are technology-agnostic. In practice, the pharmaceutical industry generates the vast majority of clinical trial data suitable for cost-effectiveness submissions, while diagnostic companies face structural barriers in generating comparable evidence of downstream outcome improvement.

The physician perception data reveals a hidden economic logic: if physicians themselves attribute 56% of outcome gains to pharmaceuticals, then investors and policymakers who allocate R&D resources proportionally to perceived impact may be rational. However, this logic contains a circularity. Physician perceptions are shaped by the data they see—and the data they see is disproportionately generated by pharmaceutical companies with the largest clinical trial budgets.

The survey’s own authors acknowledged this tension. The study included the statement: “Developments in diagnostics, medical devices, procedures, and prescription drugs have increased life expectancy and quality of life after diagnosis for many diseases.” (Source 1: [Primary Data]). Yet physicians still gave pharmaceuticals the majority share, suggesting that the perception gap between technological impact and attribution is not random—it is structural.

This structural gap has direct consequences for R&D investment strategy. If a venture capital firm or corporate R&D board allocates capital based solely on perceived physician impact, it will overweight drug development relative to diagnostic innovation. This may be economically rational in the short term, given higher reimbursement margins for drugs versus diagnostics. But it creates a strategic blind spot: the diagnostic innovations that enable earlier detection and more precise treatment selection may be systematically undervalued, leading to underinvestment in the very technologies that amplify pharmaceutical efficacy.

Image suggestion: Diagram showing a linear timeline 1990-2015 with key drug approvals (antiretrovirals, statins, biologics) and diagnostic milestones (PCR, imaging improvements) for the eight conditions.

3. The Diagnostic Blind Spot: Why 20% Is a Market Signal for Undervalued Innovation

The 20% attribution to diagnostics requires careful unpacking. The “cannot allocate” category—which captured public health interventions, lifestyle modifications, and system-level improvements—accounted for the remainder of post-diagnosis outcome gains. This residual category is, in economic terms, a confession: physicians recognize that factors beyond their direct control contribute significantly to patient outcomes, but they cannot attribute these gains to specific technology classes.

This admission of attribution limits has important economic implications. Consider HIV infection as a case study. The introduction of antiretroviral therapy (ART) in the mid-1990s transformed HIV from a fatal disease to a chronic condition. But ART’s efficacy depended critically on viral load monitoring—a diagnostic innovation that enabled physicians to adjust drug dosing in real time. Without viral load diagnostics, ART would have been dosed blindly, reducing its effectiveness and accelerating resistance development. Yet in the physician perception survey, the drug received attribution for the outcome gain, while the diagnostic that enabled optimal dosing remained invisible.

The same logic applies to breast cancer. Mammography screening and subsequent tumor biomarker testing (e.g., HER2, estrogen receptor status) enable targeted therapy selection. The diagnostic determines which drug to use and for which patient. But the physician attributes the outcome to the drug, not to the diagnostic that made the drug selection possible.

This diagnostic blind spot creates a market signal for mispriced innovation. Diagnostics, particularly in vitro diagnostics, typically command lower reimbursement prices than pharmaceuticals and face shorter product life cycles. A novel diagnostic may sell for $100-$500 per test, while a targeted therapy for the same condition may cost $50,000-$200,000 per patient annually. The physician perception survey suggests that this pricing disparity is reinforced by cognitive attribution patterns: if physicians do not mentally assign outcome improvement to the diagnostic, they will not advocate for higher diagnostic reimbursement, creating a self-reinforcing cycle of undervaluation.

The strategic recommendation for R&D portfolio managers and policymakers is clear: diagnostic innovation should not be evaluated solely on its direct revenue potential, but on its multiplier effect on pharmaceutical efficacy. A diagnostic that enables 20% more patients to receive the correct first-line therapy may generate more outcome improvement than a new drug that offers a 10% relative risk reduction. Yet current value assessment frameworks, and the physician perceptions that influence them, systematically underweight this multiplier effect.

Image suggestion: Venn diagram showing overlapping circles for diagnostics, pharmaceuticals, and patient outcomes, with diagnostics positioned as the enabling condition for pharmaceutical efficacy.

4. Funding Source Context and Survey Design Limitations: The Analytical Caveat

Any analysis of this study must address the funding source. The National Pharmaceutical Council (NPC) provided financial support for the research. The NPC is a trade association representing pharmaceutical and biopharmaceutical companies. This does not invalidate the study’s findings, but it requires the reader to consider potential design biases.

The survey methodology asked physicians to allocate percentage contributions across four categories. The researchers did not ask physicians to rank the cost-effectiveness of each category, only the attribution of outcome improvement. This distinction is critical. Physicians may attribute 56% of outcome gains to pharmaceuticals while simultaneously believing that diagnostics offer superior value per dollar spent. The study did not capture value-per-dollar perceptions, only absolute attribution.

Additionally, the physician sample—with a mean 21.4 years in practice—represents clinicians trained and practicing during the 1990-2015 period. This cohort witnessed the introduction of statins, SSRIs, antiretrovirals, and biologic therapies. They did not witness the full impact of genomic diagnostics, liquid biopsies, or AI-assisted imaging, which emerged primarily after 2015. A survey conducted today, covering 2015-2025, might yield different attribution ratios as diagnostic and digital innovations mature.

The authors themselves cautioned: “Physician perceptions indicated that attention should be paid to value assessments of innovative diagnostics, devices, and surgical procedures, as well as to pharmaceuticals and biopharmaceuticals.” (Source 1: [Primary Data]). This statement, embedded in the study’s conclusions, acknowledges that the survey results should not be interpreted as a recommendation to deprioritize non-pharmaceutical innovation.

5. Economic Implications for Value Assessment and R&D Allocation

The 56-20 attribution split has direct implications for how healthcare systems should weight innovation categories in value assessment frameworks.

Current frameworks such as ICER’s value assessment and NICE’s technology appraisal evaluate interventions on a case-by-case basis using incremental cost-effectiveness ratios (ICERs). This approach assumes that each intervention can be evaluated independently. The physician perception data suggests that this independence assumption is flawed. Diagnostic innovations enable drug efficacy; drug innovations create demand for diagnostic precision. The two categories are economically complementary, not substitutable.

A portfolio-level value assessment would recognize that investments in diagnostics yield returns that are partially captured by pharmaceuticals. Conversely, pharmaceutical R&D without corresponding diagnostic innovation may yield diminishing returns as treatment precision becomes the binding constraint on efficacy improvement.

For R&D allocation, the data supports a diversification strategy weighted toward pharmaceuticals but with a diagnostic floor. A portfolio that allocates 56% of resources to drug development, 20% to diagnostics, and the remainder to devices and digital health may reflect both physician perception and actual outcome contribution. But this allocation should be dynamic, not static. As diagnostic technologies mature—particularly liquid biopsies for early cancer detection, continuous glucose monitors for diabetes management, and genomic profiling for psychiatric medication selection—the diagnostic attribution share may increase.

6. Market Predictions and Strategic Positioning

Three predictions emerge from this analysis for the period 2025-2035.

First, diagnostic companies will invest in outcomes evidence generation. The physician perception data reveals that diagnostics suffer from an evidence deficit, not a value deficit. Diagnostic manufacturers will increasingly fund randomized controlled trials that measure downstream patient outcomes, not just analytical sensitivity and specificity. This will raise the evidentiary bar for diagnostic reimbursement and market access.

Second, pharmaceutical companies will acquire or partner with diagnostic firms at an accelerated rate. The complementarity revealed by the physician survey—drugs get the attribution, but diagnostics enable the efficacy—creates a strategic rationale for vertical integration. Large pharmaceutical firms will seek to internalize diagnostic capabilities to control the enabling technology that amplifies their drug revenue.

Third, value assessment frameworks will shift from intervention-level to pathway-level evaluation. ICER, NICE, and similar bodies will begin evaluating diagnostic-therapeutic combinations as bundled interventions, recognizing that the attribution of outcome improvement cannot be cleanly separated between the two categories. This will create new reimbursement models, such as bundled payments for diagnostic-guided therapy, that align financial incentives with outcome attribution.

The 56% figure from the Wamble et al. study will age not as a fixed truth, but as a baseline measurement from which future perceptions deviate. Whether that deviation moves toward greater diagnostic attribution or reinforces pharmaceutical dominance will depend on the evidence generated, the regulatory frameworks established, and the investment strategies pursued over the next decade.

Sources cited:

Source 1: Primary Data - Wamble DE, Ciarametaro M, Dubois R. "Physician Perceptions of the Relative Contributions of Medical Technology Innovations to Patient Outcome Improvements." Journal of Managed Care & Specialty Pharmacy, January 2019.
Source 2: CDC. "National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey, 1990-2014."