Evaluating Disparities in Oncology Clinical Trials Using Haddon’s Matrix

Evaluating Disparities in Oncology Clinical Trials Using Haddon’s Matrix

Medical Oncology & Disparities in Clinical Trials

Considering Disparities Using Haddon’s Matrix

October 15, 2021

Wes Chapman

Introduction – Disparities, Implications for Clinical Trials

This is the second in a series of blogs looking at value-based medical oncology care, specifically addressing the key clinical/process elements in a value-based program that create value for the patient and the rest of participants in the care ecosystem. In this blog I examine the structural and racial/ethnic based disparities that limit access to clinical trials for minority populations.

Clinical trials in oncology are a critical element of value creation in the ecosystem for oncology care. This is particularly important for patients, for whom clinical trials offer access to potentially life extending therapies. Additionally clinical trials have been demonstrated to both lower total cancer care costs and improve outcomes (overall survival) for participants (here) and (here). Despite these benefits, 80% of trials are delayed due to recruitment problems, the average dropout rate is over 30% and 50% of sites enroll either one or zero patients (here).  Reported oncology clinical trial participation rates by oncology patients in the US range from less than 3% (here)to 8% (here).  Minorities are underrepresented in clinical trials by 50% relative to their population (here), and less based on incidence of specific cancer types (here). Finally, as of 2018, the participation in US based genome-wide association studies was 80% European, 10% Asian and only 2% African. This systematic bias can lead to drug approvals which do not reflect the underlying population, with the potential for systematic population generalization based errors as in the recent Plavix case in Hawaii (here).

Companies have begun to adopt aggressive postures towards ensuring representative sample proportions in their clinical trials. Most recently Moderna had to delay the trial of its COVID-19 vaccine (here) due to inadequate minority representation in the trial. Delays such as this impede the time to market of any new therapeutic – costing money. Addressing disparities is now both a moral and economic necessity for the pharmaceutical industry.

The potential economic impact of oncology clinical trials on the entire care ecosystem is the potential to reduce drug costs by the amount of “free” drug contributed to the system by industry trial sponsors. We estimate that increasing trial participation by 5-10% of the adult population would produce a net savings in purchased drug costs of $2.75-5.5 billion (here). This level of participation is reasonable considering that children participation in clinical trials has averaged over 50% since the 1970’s.

Barriers to Trial Participation

In their paper, Role of Clinical Trial Participation in Cancer Research: Barriers, Evidence, and Strategies, Unger, Cook, Tai and Bleyer, (here) developed the following flow chart of barriers to entry to clinical trials by minorities and other trial participants.

Figure 1

This construct has the advantage of being a reasonable simple estimation of the flow process map of clinical trial enrollment. It deals with Structural (access) and Clinical (eligibility criteria) as barriers to enrollment in trials, which they certainly are. The Physician and Patient barriers are dealt with as attitudinal and informational in nature. The problem with a barrier-based model, is that it: 1) does not recognize either incentives or dis-incentives (economic or otherwise) for participation at any of the steps, 2) does not address any of the historical or current issues of structural or individual racism or bias that might influence behaviors by actors within the ecosystem, 3) does not consider any issues beyond enrollment – (what happens to continued trial participation?), 4) ignores the important role that regulation and sponsor activities play in the process, 5) ignores the role of the host clinic and /or hospital plays in the clinical trial process, 6) does not consider the impact of just-in-time (JIT) trials (also known as decentralized trials) or RWE evidence trials and how they might impact the structural and clinical barriers.

Unger et al expanded on this work with a 2019 paper, Systematic Review and Meta-Analysis of the Magnitude of Structural, Clinical, and Physician and Patient Barriers to Cancer Clinical Trial Participation, (here), which did meta level analysis to examine the relative impact of each of these barriers and quantifying their potential impact on trial enrollment.

Figure 2

The analysis presented above (Figure 2) is based on meta-level research looking at the barrier model, and the relative impact of each of the barriers sequentially and when grouped based on broadly structural and decision-based criteria. This analysis clearly indicates that the structural barriers are critical gating factors in narrowing the base of potential trial participants down to those ultimately enrolled.

Haddon’s Matrix

I have considered the problem of minority enrollment as an issue of public health and epidemiology using the tools and methods developed by Dr. William Haddon.  William Haddon Jr, a public health physician, was instrumental in applying scientific methods to the study of injuries, particularly motor vehicle injuries. The Haddon matrix is a conceptual model for the systematic exploration of the etiology of accidents, countermeasures, and provides an integrated approach to injury control. Dr. Haddon brought the tools of epidemiology and public health to the study of accidents – particularly on highways. His career in public health began in the National Highway Safety Administration and later transitioned into the insurance industry. His methods have proven relentlessly successful since 1968 in transportation and public health applications in the US (see Figure 3 below).

Figure 3

In my model, I consider each failure to enroll a minority participant in a clinical trial as an “accident” or “failure” and look at barriers to prevent those failures. The matrix looks at the types of factors used by Haddon but focused on the issues relevant to clinical trials and the analysis of factors leading to failure with minorities and disadvantaged communities.

A Conceptual Matrix for Minority Clinical Trial Participation

  Influencing Factors
Phase Host Economic Regulatory/Clinical Social
Pre-Trial Enrollment Patient Limitations by insurance, transportation, etc. Enrollment Criteria, understanding consent, language barriers. Treatment facility has no trials. Trials not offered. Dreadful history, different language, limited understanding of process or benefits. Staff/physician prejudices and bias
Sponsor Speed to maximum enrollment Limited diversity requirements Limited interaction/engagement, limited population diversity requirements
Treating Physician & Staff Wants to keep patient, no other incentive Trials not offered. Time and cost to enroll patient, staff training, physical plant compliance No trials experience, not updated on trials benefits. Prejudices and bias
Hospital or Clinic Payment too small to justify participation, additional system cost, staff training CoC accreditation and tumor boards do not require novel compound trials RVU compensation system does not reward trials.
Continued Participation During Trial Patient Additional unreimbursed costs. Compensation reflecting historical inequities Confusing and difficult compliance Issues of perceived benefit
Sponsor Limited flexibility for compensation No evaluation based on diversity success Limited interaction and engagement
Treating Physician Original physician lost patient. Disincentive to support trial New site physician may see needs of trial first New site team has limited patient history and engagement.
Hospital or Clinic Trial may interfere with other more lucrative treatment Additional cost and complication for compliance New patients need special services
Post-Trial Follow up Patient Limited financial capabilities/benefit Complicated follow up requirements Need to travel, complex communications
Sponsor Limited financial impact Limited ongoing oversight Easier to not engage
Treating Physician Limited financial benefit Limited ongoing oversight Easier to not engage
Hospital or Clinic Trial may interfere with other more lucrative treatment Limited ongoing oversight Trial may interfere with ongoing relationship


I regard that there are three distinct phases of clinical trial engagement for minority participants: 1) Pre-trial enrollment, 2) Continued participation during the trial, and 3) Continued engagement in post-trial follow up. Each of these phases has unique challenges and value to the minority participant, sponsor, and clinical team.

I have considered three influencing factors: 1) Economic, 2) Regulatory/Clinical, and 3) Social.  The influences that we considered potentially most relevant fell into one of these categories, and the categories are distinct enough to have little obvious correlation.

Considerations, Examples & Recommendations

The host for each of the patient, trial sponsor, treating physician and staff and finally the facility – a distinction most important for large clinics and hospitals. Each of these participants has a unique view and set of incentives and related influencing factors. As examples:

  • Sponsor Incentives. Every effort should be put in place to work with sponsors through incentives to encourage rapid trial enrollment which includes enough representation of minorities to ensure that the science behind the evaluation of tests therapeutics is valid and done in a fair and equitable manner. Increasingly we see that oncology drugs are approved in early phase trials (phase 1, 1b and 2) and using some form of enhanced approval to shorten the time to approval. The FDA is increasingly requiring participation in trials sufficient to ensure at least proportional representation for minorities to ensure scientific validity. This can be encouraged through expanded access to JIT trials in geographies targeted for minority recruitment.
    1. Recommendation 1. That the FDA require adequate minority representation in clinical trials to access accelerated approvals and that the agency consider encouraging the design and utilization of JIT trials to help accomplish this objective.
    2. Recommendation 2. That trial protocols include a complete genetic profile for all oncology trial participants and that these profiles be included in all submissions to the agency. Limited genetic panels can no longer be considered adequate for the understanding of efficacy, safety, or statistical adequacy, particularly as applied to minority participants.
  • Insurance Compatibility. Insurance must be compatible with clinical trial participation for insured patients to participate in trials. Beginning in 2007, Medicare agreed to pay all normal and customary medical costs for trial participants – leaving only extraordinary costs and experimental drugs to be paid for by the trial sponsor (here). The extension for normal commercial insurance was included in ACA in 2014 (here). A facility may have a trial and the patient may meet the inclusion criteria, but if the patient is covered by Medicaid, they probably will not have compatible insurance coverage for covering reasonable and necessary medical expenses associated with trial participation until 2022 (here). Access is a nuanced issue, not adequately addressed by simple trial presence in a facility. The Medicaid issue was not resolved until December of last year and has a significantly disproportionate impact on minority trial participants based on the demographics of the Medicaid population – minority participants have twice the Medicaid participation as their proportion of the total population. Despite the recent expansion of insurance coverage for clinical trials out of network requirements for care are not covered for clinical trials.
    1. Recommendation 3. That all payers pay for reasonable and customary expenses at their normal “in-network” costs for clinical trial patients, and those providers and facilities accept those rates for those patients. Sponsors should normally include the patient’s portion of any copays and deductibles associated with clinical trial participants. This will equalize the opportunity for patients with high deductible plans, and disproportionately benefit poor populations.
    2. Recommendation 4. That in cases where drug is provided free through a clinical trial to a trial participant covered by commercial insurance, and prior authorization has been used to establish the base treatment plan to be paid for by the commercial insurer, that an equitable apportionment of monetary savings realized by the insurer for the free drug used in the allocated to reasonable and necessary expenses of the trial.
  • Compensation for Minority Trial Participants. The compensation for minority participants in clinical trials is an issue has not been adequately addressed and remains a barrier to participation for poor and underserved minorities. This is an issue which can be used to compensate for expenses and hardship but cannot be used to unduly influence. There is limited guidance as to how this might be used to reflect the additional hesitancy based on racially based exploitive history or the actual additional costs based on current situations for different populations (here).
    1. Recommendation 5. That payment for clinical trial participation be adjusted for trial participants to reflect actual incremental costs and the aggregate impact of cancer care on minority participants financial situation and capabilities.
  • Sponsor and Physician Responsibilities. Patient navigation was originally designed and implemented to help bridge the social and cultural gaps that separated African American patients from adequate cancer care (here). The program has been successful and has been expanded to be designated best care practice. The issues of cost incurred by minority populations participating in trials is further complicated by educational and communication differences that make understanding what is really going on with clinical trial participation difficult for disadvantaged minority groups. This same problem was endemic in the delivery of clinical care and led to the development of patient navigation for minority populations in a program started by Dr. Harold P. Freeman in the 1980’s (here). In 1989 the American Cancer Society published a report looking at care for poor and minorities which concluded five points which are relevant today for minority participants in clinical trials:
    1. Poor people face substantial barriers to obtaining cancer care and often do not seek care if they cannot pay for it.
    2. Poor people and their families often make extreme personal sacrifices to obtain and pay for care.
    3. Fatalism about cancer is prevalent among the poor and may prevent them from seeking care.
    4. Cancer education programs are often culturally insensitive and irrelevant to many poor people.
    5. Poor people endure greater pain and suffering from cancer than other Americans.
      1. Recommendation 6. That navigation for minority participants in clinical trials be provided through the clinic of the patient’s physician, with the financial support and engagement of the sponsor.

Summary and Conclusions

Haddon’s matrix is a proven tool to identify the etiology of accidents and failures in complex public health applications. We have adapted the work of Dr. Haddon to reflect our views of the challenges and opportunities for increased clinical trial participation, commensurate with good science, equity, and the evolving needs of these minority and disadvantaged communities. Using Dr. Haddon’s methodology, I have identified 6 low-cost, high-impact recommendations to address historical and current racial and social inequities in the clinical trial process.


Reference Authors, date URL
A multi-year look at the cost burden of cancer care


Dieguez, Ferro & Pyenson, 2017 https://www.milliman.com/en/insight/2017/a-multi-year-look-at-the-cost-burden-of-cancer-care
AACR CANCER DISPARITIES PROGRESS REPORT 2020 2020 https://cancerprogressreport.aacr.org/wp-content/uploads/sites/2/2020/09/AACR_CDPR_2020.pdf
Addressing Systemic Racism Through Clinical Preventive Service Recommendations From the US Preventive Services Task Force Doubeni, Simon et al https://jamanetwork.com/
Annual Report to the Nation on the Status of Cancer, Part II: Progress Toward Healthy People 2020 Objectives for 4 Common Cancers Henley et al, 2020 https://acsjournals.onlinelibrary.wiley.com/doi/full/10.1002/cncr.32801
Bristol-Myers, Sanofi ordered to pay Hawaii $834 million over Plavix warning label


Reuters, 2/15/2021 https://www.reuters.com/article/us-bristol-myers-sanofi-plavix/bristol-myers-sanofi-ordered-to-pay-hawaii-834-million-over-plavix-warning-label-idUSKBN2AF1YI
Cancer Care Costs in the United States Are Projected to Exceed $245 Billion by 2030 AACR, 2020 https://ascopost.com/news/june-2020/national-cost-of-cancer-care-in-the-united-states-expected-to-rise-to-over-246-billion-by-2030/
Cancer patients receiving chemotherapy: Opportunities for better management


Fitch, Pyenson, 2010 https://us.milliman.com/en/insight/research/health/cancer-patients-receiving-chemotherapy-opportunities-for-better-management/
Cancer Statistics, 2018 Siegel et al, 2018 https://pubmed.ncbi.nlm.nih.gov/29313949/
Cancer Statistics, 2020 Siegel, Miller & Jamal, 2020 https://acsjournals.onlinelibrary.wiley.com/journal/15424863
Cancer Won’t Wait BUILDING RESILIENCE IN CANCER SCREENING AND DIAGNOSTICS IN EUROPE BASED ON LESSONS FROM THE PANDEMIC IQVIA Institute, 2021 https://www.iqvia.com/insights/the-iqvia-institute/reports/cancer-wont-wait
Cost Drivers of Cancer Care: A Retrospective Analysis of Medicare and Commercially Insured Population Claim Data 2004-2014 Fitch, Pelizzari & Pyenson, 2016 https://www.milliman.com/en/insight/cost-drivers-of-cancer-care-a-retrospective-analysis-of-medicare-and-commercially-insured
Delivering innovation: 2020 oncology market outlook McKinsey & Company, 2020 https://www.mckinsey.com/industries/pharmaceuticals-and-medical-products/our-insights/delivering-innovation-2020-oncology-market-outlook
Direct Contracting Is the Key to Bypassing the Payer Rinde, 2020 NA
Employer Health Benefits, 20119 Summary of Findings Kaiser Family Foundation, 2019 https://www.kff.org/report-section/ehbs-2019-summary-of-findings/#:~:text=The%20average%20annual%20premiums%20for,%25%20and%20inflation%20increased%202%25.
Employer Health Benefits, 2019 Annual Survey Kaiser Family Foundation, 2019 https://www.kff.org/health-costs/report/2019-employer-health-benefits-survey/
Equity-Driven Approaches to Optimizing Cancer Care Coordination and Reducing Care Delivery Disparities in Underserved Patient Populations in the United States Oyer, Smeltzer et al, 2020 https://ascopubs.org/doi/full/10.1200/OP.20.00895
Facts & Figures 2020 Reports Largest One-year Drop in Cancer Mortality Simon, 2020 https://www.cancer.org/latest-news/facts-and-figures-2020.html#:~:text=The
Global Oncology Trends 2017 QuintilesIMS, 2017 https://communityoncology.org/wp-content/uploads/2017/06/QIIHI_Oncology_Trend_Report_2017_Advances_Complexity_Cost.pdf
Global Oncology Trends 2019 IQVIA, 2019 https://www.iqvia.com/insights/the-iqvia-institute/reports/global-oncology-trends-2019
Global Trends in R&D IQVIA, 2020 https://www.iqvia.com/insights/the-iqvia-institute/reports
Population Health Informatics Can Advance Interoperability: National Program of Cancer Registries Electronic Pathology Reporting Project Pollack et al, 2020 https://pubmed.ncbi.nlm.nih.gov/33125274/
Precision medicine in practice: Strategies for rare cancers McKinsey & Co., 2020 https://www.mckinsey.com/industries/pharmaceuticals-and-medical-products/our-insights/precision-medicine-in-practice-strategies-for-rare-cancers
Should next-generation sequencing tests be performed on all cancer patients? McKenzie et al, 2019 https://www.tandfonline.com/loi/iero20
The 2019 Genentech Oncology Trend Report 10th Edition Genentech, 2019 https://www.genentech-forum.com/trend-reports.html
The Costs of Cancer, 2020 Edition American Cancer Society, 2020 https://www.fightcancer.org/sites/default/files/National%20Documents/Costs-of-Cancer-2020-10222020.pdf
The Drugs at the Heart of Our Pricing Crisis Bach, Trusheim, 2021 https://www.nytimes.com/2021/03/15/opinion/how-to-control-drug-prices.html
The Future of Oncology 2020 https://pharmaphorum.com/deep_dive/
The oncology practice of the future: Six ways to prepare for success in value-based care CardinalHealth, 2020 https://www.cardinalhealth.com/en/services/specialty-physician-practice/resources/whitepapers-and-downloads/oncology-practice-of-the-future.html


Lisa Woods, Jonathan R. Slotkin, and M. Ruth Coleman, 2019 https://hbr.org/2019/03/how-employers-are-fixing-health-care
Walmart has no shelf space for hospitals’ shoddy healthcare Molly Gamble. 2019 https://www.beckershospitalreview.com/strategy/walmart-has-no-shelf-space-for-hospitals-shoddy-healthcare.html
Wes Chapman
Written by Wes Chapman

No comments yet.

No one have left a comment for this post yet!