Counterpoint

On November 13, 2024, we published “7 Strategic Healthcare Implications of the 2024 Presidential Election” to “identify key strategic issues for health economy stakeholders based on hindsight analysis of the health policy initiatives in the previous Trump Administration, the Biden Administration and the 118th Congress in the context of certain key trends identified in our 2024 Trends Shaping the Health Economy Report.”  

To mark the 100th day of the Trump Administration, we asked ChatGPT to grade my predictions with the benefit of hindsight analysis.  

Read the full conversation →

Here are the results.

With the caveat that ChatGPT is seemingly confused about what the first Trump Administration tried to do with the Inpatient Only list in 2020, ChatGPT provided the following analysis of my predictions. 

1. Affordability Will Dominate the Agenda

• Prediction: Healthcare affordability will become a central focus, driven by economic concerns. 
• Grade: A 
• Assessment: This prediction has materialized. Both federal and state policymakers are emphasizing cost containment in healthcare, with initiatives aimed at reducing patient expenses and overall system costs. 

2. Commercial Price Controls Are on the Table 

• Prediction: A Trump administration might consider implementing price controls in the commercial insurance market, influenced by studies from the CBO and RAND. 
• Grade: B+ 
• Assessment: While direct federal price controls have not been enacted, discussions around regulating commercial healthcare prices have gained momentum, especially at the state level. Republican-led states like Indiana have explored measures to limit healthcare costs, reflecting the article’s foresight. 

3. Site-Neutral Payments May Reemerge

• Prediction: Policies promoting site-neutral payments could be revived, aiming to equalize reimbursement rates across different care settings. 
• Grade: A- 
• Assessment: This prediction aligns with current policy trends. Bipartisan support for site-neutral payment reforms has grown, with legislative proposals introduced to address payment disparities between hospital outpatient departments and other care settings.

4. The Inpatient Only (IPO) List Will Stay Gone

• Prediction: The Trump administration would maintain the elimination of the IPO list, while the Biden administration might consider reinstating it. 
• Grade: B 
• Assessment: As of now, the IPO list remains eliminated, and there has been no significant movement to reinstate it. The article accurately anticipated the Trump administration’s stance, though the prediction about potential reversal under a Biden administration is not applicable in the current context. 

5. Price Transparency Efforts Will Continue

• Prediction: Both parties will support ongoing efforts to enhance price transparency in healthcare. 
• Grade: A 
• Assessment: This prediction has proven accurate. Initiatives to improve price transparency have continued, with policies requiring hospitals and insurers to disclose pricing information, aiding consumers in making informed healthcare decisions. 

6. Prescription Drug Policy Will Stay Political

• Prediction: Drug pricing will remain a contentious and politically charged issue. 
• Grade: A 
• Assessment: The prediction holds true. Debates over prescription drug prices persist, with legislative and regulatory actions aimed at controlling costs, reflecting the ongoing political focus on this issue. 

7. The FDA Will Stay in the Spotlight

• Prediction: The FDA will continue to face scrutiny, particularly regarding accelerated approvals and the integration of AI/ML in drug development. 
• Grade: B+ 
• Assessment: The FDA has remained under attention, especially concerning its approval processes and the oversight of emerging technologies in drug development. While not the central focus of healthcare policy debates, the agency’s actions continue to be closely monitored. 

Next, we asked ChatGPT what my list of predictions omitted based upon the first 100 days of the Trump Administration. 

1. Medicare Advantage Scrutiny 

• Why it matters: MA is consuming a larger share of Medicare spending and has drawn bipartisan scrutiny over coding intensity, overpayments, and quality bonuses. 
• What could have been predicted: Greater regulatory oversight and audits targeting MA plans, especially as CMS and OIG ramp up reviews.
• Missed Opportunity: The article discussed affordability but didn’t distinguish between commercial and government plan reforms—Medicare Advantage is a major frontier in cost containment. 

2. State-Led Healthcare Policy Innovation 

• Why it matters: In a divided federal government or deregulatory Trump scenario, states (particularly red states) are increasingly driving healthcare policy—on transparency, scope of practice, facility fees, and hospital regulation. 
• What could have been predicted: A surge in state legislation that fills policy gaps left by federal inaction. 
• Missed Opportunity: The article framed healthcare as a federal policy issue but didn’t explore the rising influence of state-level health policy. 

3. Workforce Shortages and Scope of Practice Expansion

• Why it matters: The clinical labor crisis is one of the most acute operational constraints in healthcare, and states are expanding scope of practice laws (e.g., for NPs and PAs) to ease pressure. 
• What could have been predicted: Deregulatory pushes—especially under a GOP-led HHS—supporting expanded scope and reduced licensing barriers. 
• Missed Opportunity: Workforce policy is a major strategic concern for hospitals, but it wasn't discussed.

4. Behavioral Health as a Bipartisan Priority

• Why it matters: Mental and behavioral health have strong bipartisan momentum, particularly in rural and pediatric populations. Federal investments continue regardless of administration. 
• What could have been predicted: Sustained or increased funding for mental health infrastructure, school-based programs, and telebehavioral health. 
• Missed Opportunity: The article’s scope was mostly economic (pricing, reimbursement), missing the social determinants and public health issues gaining traction. 

5. AI/Automation in Provider Operations

• Why it matters: As healthcare organizations look to reduce cost structures, AI is being deployed for documentation, prior auth, and even diagnostics—raising policy questions about regulation and liability. 
• What could have been predicted: Regulatory activity from HHS, ONC, or FDA regarding AI/ML applications in healthcare operations—not just drug development. 
• Missed Opportunity: The FDA was discussed in the context of clinical trials, but not broader regulatory or workforce-displacement impacts of automation. 

6. Employer Market Pressures

• Why it matters: Employers are increasingly vocal about rising healthcare costs and lack of value, and they’re lobbying for changes—such as reference pricing or direct contracting. 
• What could have been predicted: Policy activity (or at least political rhetoric) around empowering employers or reining in large hospital systems on pricing. 
• Missed Opportunity: While commercial price controls were mentioned, the employer voice in shaping the future of coverage models was absent. 

How do I grade ChatGPT’s analysis of the relevance of the themes that I (supposedly) did not address? 

First, the purpose of my article was to focus on the strategic implications of the outcome of 2024 Presidential election. In hindsight, which is 20/15, I failed to distinguish between policy initiatives that would continue in the new Administration – like scrutiny of Medicare Advantage – and those that would change. ChatGPT overlooked what I think is the most notable Medicare Advantage policy development of the first 100 days of the Trump Administration, which is CMS’s announcement of a 5.06% increase in Medicare Advantage payments for 2026. I would never have predicted such a stunning increase in an administration that is otherwise obsessed with cost reductions. 

Second, I agree with ChatGPT that state-led healthcare policy initiatives in the past 100 days have potentially massive strategic implications, which raises the question of how ChatGPT gave me no credit for including this state-led policy initiative in my article: 

Last week President-elect Trump won 58.6% of the vote in Indiana, while Senator Mike Braun (R) was elected Governor. Indiana is a reliably “red” state, as evidenced by the fact that Republicans have had a “trifecta” for the past 16 years, holding the Governor’s office and control of both chambers in the General Assembly. The fact that the Indiana General Assembly has enacted several laws that attempt to limit healthcare costs in the last two years suggests that lowering healthcare costs is a priority at both the Federal and state levels. Notably, the well-known RAND Corporation study that revealed that the average commercial reimbursement rate for hospital services is 255% of Medicare was a touchstone for the Indiana General Assembly. Providers whose commercial reimbursement rates exceed 250% of Medicare could increasingly face scrutiny from elected officials, employers and consumers.  

While my article cited the Congressional Budget Office’s 2022 “Policy Approaches to Reduce What Commercial Insurers Pay for Hospitals’ and Physicians’ Services” as part of my reasoning, I failed to predict that the Indiana General Assembly’s response to the CBO would be the redneck’s credo: “Hey y’all, watch this.” Since then, I have tried to atone for my mistake with a series of articles about the potentially catastrophic impact of state-led price controls: 

• The Category 5 Hurricane Forming in the Indiana State Legislature 

• Price Controls for Hospitals, the Only Health Economy Stakeholder With Unlimited Downside Risk 

• The Doom Loop of Price Caps: Past, Present and Future 

I could not agree more with ChatGPT that self-insured employers will be an increasingly important stakeholder in the future of U.S. healthcare, which I have written about extensively over the past two years based on our insights from the health plan price transparency files published under CMS’s Transparency in Coverage initiative. Employers have long been subject to fiduciary duties under ERISA, but price transparency is the catalyst for determining whether employers have met ERISA’s fiduciary standards. 

Workforce issues are paramount, but the challenge results from the decades-long medieval guild-like mindset of professional medical societies. There is no data to suggest that the crisis of physician and nurse shortages can be solved. As a result, the strategic issue is when and how providers will decide which service lines to exit. 

Finally, as to the importance of AI in provider operations, ChatGPT is undoubtedly correct in theory but is incapable of understanding the barriers to widespread adoption of AI in the health economy. Perhaps more importantly, many health economy stakeholders mistakenly think that ChatGPT and AI are synonymous.  

ChatGPT is built on a machine learning model called a transformer neural network, specifically designed for natural language processing tasks. It learns to generate human-like text by being trained on vast amounts of text data from the internet. During training, it predicts the next word in a sentence based on the context of the previous words, allowing it to generate coherent and contextually relevant responses.¹ 

If you understand what that paragraph means, then you will understand that ChatGPT’s impact on the health economy might be rather limited. On the other hand, because ChatGPT “understands” that “AI” is an umbrella term that includes natural language processing, machine learning, neural networks, image recognition, etc., ChatGPT recognizes the potential of AI to automate workflows. 

However, ChatGPT cannot “understand” the rate-limiting factor of AI adoption to automate the numerous repetitive, time-consuming and wasteful administrative tasks in healthcare. While some AI, like ambient AI, can be deployed unilaterally to automate clinical documentation, using AI to automate many other healthcare administrative tasks requires multiple parties to share a precise understanding of the rules for those tasks. There are few healthcare use cases more susceptible to automation through AI than prior authorization, and the current disputes about deploying AI for that use case reveal that humans have not yet agreed on the applicable rules. 

One thing is certain in the uncertainty created by the unpredictable and frequently chaotic first 100 days of the second Trump Administration: change is coming. Every health economy stakeholder should remember these words from General Eric Shinseki: “If you don’t like change, you’re going to like irrelevance even less.”