From drug discovery to patient care – Global X ETF

We believe the health system is at a tipping point, leveraging digital health technologies to make long-standing inefficiencies in health care a thing of the past. Artificial intelligence is at the center of this transformation, with the potential to improve healthcare outcomes by 30-40% while saving the United States as much as 10% in annual healthcare costs.^{1, 2}

Important points

• Drug discovery is a long and expensive process with many inefficiencies and potential pitfalls. AI has the potential to help at virtually every step of the process, from early-stage computer simulation to late-stage trial design, patient recruitment, and data analysis.
• The healthcare industry is large and complex, with many insurance companies and providers. This creates friction, inefficiency, and paperwork, which AI can reduce and potentially save significant time and money.
• Given the size and diversity of the healthcare sector, it can be difficult to pick individual winners from AI’s disruption of the industry. Therefore, ETFs that offer broader exposure to this megatrend may be an attractive option for investors.

AI in drug development

Drug discovery is an extremely difficult process. Despite advances in technology, developing a new drug still takes him 10-15 years and costs him, on average, $1.3 billion.^{3, 4} Additionally, 90% of investigational drugs fail in human trials because they are ineffective or have too many side effects.^Five Complicating matters, clinical trials are notoriously difficult to design and run, often hindering the approval of treatments.

AI in drug discovery: Decades of powering computer-aided drug discovery

Since the 1990s, computer simulation software has been used in the development of investigational drugs, successfully reducing costs and increasing drug discovery success rates.⁶ However, new AI-enabled models suggest even more favorable unit economics. By running millions of scenarios, AI software can reduce preclinical drug development costs by 20-40% and accelerate drug candidate design and validation by as much as 15 times.^{7, 8}

AI provides drug development with:

• Build better molecules: AI software predicts the 3D structure of target proteins, drug-protein interactions, and activity in new therapeutics. Improving molecular structure can ensure high levels of efficacy for investigational drugs in preclinical studies. But most importantly, AI will help ensure maximum correlation between preclinical studies in mice and late-stage human clinical trials.
• maximum research effort: AI software can help design multiple target drug molecules and predict drug reuse to maximize therapeutic efficacy. Emerging therapeutic categories such as GLP-1 (glucagon-like peptide 1) are successfully treating conditions such as type 2 diabetes, obesity, and cardiovascular disease with a single molecule. AI can help predict whether a treatment for one disease will also help other diseases, potentially making it easier to identify blockbuster treatments early in the drug discovery phase.
• Enhanced patient identification: AI software can help predict the toxicity and efficacy of treatments through genomic profiling and identify characteristics of patients likely to benefit. Establishing these guidelines early in the drug development process helps provide a clear overview of clinical trial participants and increases the likelihood that an investigational drug will be successful. If approved, it would also open a clear path to marketing the treatment.

The impact of broader AI drug discovery is significant and has the potential to change the drug discovery process as we know it. By 2025, an estimated 30% of new drugs will be discovered using AI technology, up from zero today.⁹ Increased use of AI drug discovery is expected to lead to the development of 50 additional new treatments over the next 10 years, which alone could lead to more than $50 billion in revenue.^Ten

AI in Clinical Trial Operations: Modernizing Human Clinical Trials

The design and conduct of clinical trials is known for its inherent complexity and challenges. Artificial intelligence reduces the cost of running clinical trials and provides scalable solutions to some of the biggest obstacles in the drug development process.

• trial design: Determining the cause of negative trial results can be difficult, as unknown flaws in trial design can mask a drug's true effectiveness. AI can identify patterns in test results that humans cannot, helping to determine, for example, whether a drug is only suitable for certain patient populations.
• inefficient recruitment: One of the reasons why an estimated 86% of clinical trials fail to meet their registration deadlines is because 85% of patients are unaware of the existence of clinical trials in which they can participate.¹¹ Nearly one-third of Phase III trials fail due to enrollment difficulties, including problems identifying suitable patients and retaining participants.¹² AI can help reduce the time and cost of recruiting participants by analyzing large datasets of patient records to identify eligible candidates for clinical trials. Tempus, an AI precision medicine company, is accelerating clinical trial enrollment by achieving trial activation within 10 business days, a significant improvement from the industry average of 8 months.¹³ AI is projected to save $13 billion in costs in identifying clinical trial participants.¹⁴
• insufficient data: Each investigational treatment uses a rubric to evaluate its effectiveness. For some diseases, we rely on surrogate endpoints to measure the effectiveness of treatment that may correlate with clinical improvement, but a relationship is not necessarily guaranteed. An estimated 80% of cancer treatment trials rely on surrogate endpoints.¹⁵ AI-powered wearables can help measure drug effectiveness and monitor patients during clinical trials, potentially eliminating the need for frequent trips to testing centers. Greater integration with electronic patient-recorded data and telemedicine is expected to make AI models even more powerful.

Clinical trials can also benefit from digital twins. These are AI-generated virtual models of real patients that simulate different dosing regimens and patient progression. The impact of digital twins is so high that they can reduce the size of control groups in clinical trials by 30%. This means more clinical trial participants can receive the active ingredient rather than a placebo.¹⁶

AI in healthcare operations: the missing link to transformative digitization

Although the healthcare industry has made great strides in digitizing its operations, an estimated 80% of healthcare documents in the United States are still sent by mail or fax.¹⁷ Although industry documentation is becoming increasingly digital with the introduction of electronic medical records (EMR), the process remains inefficient and labor-intensive. Not surprisingly, 78% of physicians report health IT-related burnout and fatigue.¹⁸

AI offers a wide range of benefits by reducing the daily tasks of healthcare providers and allowing them to spend more time treating patients. Among them, we highlight a few excerpts that have had a notable impact.

• Extract relevant information from medical records to make medical decisions
• Assist in creating notes such as progress notes and discharge summaries for EMR integration
• Provide care instructions to patients, including dietary restrictions prior to surgery
• Reduce the burden of pre-approval and insurance claims
• Summarize a medical journal article
• Patient referrals, transfers, prescription writing, medical meal planning, and visit scheduling.

AI brings significant cost savings across the healthcare ecosystem in all everyday applications.

• Private insurers could save 7 to 9 percent of their total costs, potentially generating savings of $80 to $110 billion annually over the next five years.¹⁹
• Physician groups could save 3% to 8% of their costs, equating to savings of $20 billion to $60 billion.²⁰
• Hospitals could potentially save 4 to 11 percent of their costs, or $60 billion to $120 billion annually.^{twenty one}

Automating medical documentation: Freeing up doctors' time

Physicians spend an estimated 39% of their time documenting patient information in electronic medical records.^{twenty two} Digital health companies have prioritized automating patient records so doctors can spend more time on patient care. Teladoc specifically partnered with Microsoft's Nuance, an AI solution that automatically transcribes patient visits, saving an average of seven minutes per appointment. Nuance found that participating physicians' feelings of burnout were reduced by his 70%.^{twenty three}

Traditional transcription services can take 72 hours or more, which is why we've seen other tech giants step into this space.^{twenty four} Amazon Web Services recently launched HealthScribe, a new service that uses voice recognition, machine learning, and AI to summarize doctor visits.^{twenty five} OpenAI, which developed ChatGPT, is also entering this field. OpenAI is working with Hint Health to develop new features that will allow doctors to record appointments and automatically transcribe consultation notes.²⁶

AI-powered chatbots: Advancing patient care

We have also seen conversational AI tools customized for the healthcare industry come to market in hopes of simplifying the complex nature of healthcare and primarily enabling insurance coverage. Pre-approval is a very difficult process through which insurance companies review and approve certain medical services and products before they provide them. Although this process can be time-consuming, the new AI service reduces the cost per transaction by 70%, from $10 to $3.²⁷

Conversational AI tools allow doctors to enter information in their own words, and the service takes care of the rest. Digital healthcare company Doximity recently announced DocsGPT, an AI-powered chatbot tool that facilitates prior authorizations, claims, and patient communication. Once the doctor approves her AI-generated message, the platform will automatically send it to the corresponding party as well. Recent research shows that when doctors utilize her AI-based chatbots, responses are typically longer, higher quality, and more empathetic, resulting in improved overall bedside manner. It has been.²⁸

conclusion

Although it is still early days for investors to quantify the impact of artificial intelligence, the potential use cases for AI healthcare promise to revolutionize entire industries. But picking individual winners can be difficult in a field as vast as healthcare, which includes everything from drug discovery to insurance to treatment and more. ETFs can provide broad exposure to this megatrend, whether investors want to focus on digital health, genomics, or purer AI.

Related ETFs

GNOM – Global X Genomics & Biotechnology ETF

EDOC – Global X Telemedicine & Digital Health ETF

AIQ – Global X Artificial Intelligence & Technology ETF

BOTZ – Global X Robotics & Artificial Intelligence ETF

Click on the fund name above to display its current holdings. Stocks held may change. Current and future holdings involve risks.