AI in Healthcare: Building Trust

The Future Health Index is Ten Years Old

In the past decade, the Future Health Index has examined the role of technology in addressing key trends in healthcare systems. Initially a benchmark for connected care implementation, the index has evolved to explore how technology can shape the future of healthcare, based on the perspectives of healthcare leaders, professionals, and patients across diverse demographics and healthcare systems.

Key milestones:

• 2016: Integration of services, data, and systems for a healthier population.
• 2017: Delivering more effective 'connected care'.
• 2018: Leveraging data and telehealth for better outcomes.
• 2019: Utilizing diverse perspectives on digital health technologies.
• 2020: Empowering the next generation of healthcare professionals to transform care; the impact of COVID-19 on younger healthcare professionals.
• 2021: Building resilient and sustainable futures.
• 2022: Utilizing advanced analytics to improve healthcare delivery.
• 2023: Innovating through new care delivery models.
• 2024: Addressing workforce shortages with AI-based innovation and a new care delivery model.

Foreword

Healthcare systems worldwide are under pressure due to increasing workforce shortages, rising costs, and a growing demand for care driven by an aging population and chronic conditions, including in the Netherlands. Simultaneously, patients expect more personalized care, shorter waiting times, and a better experience. Healthcare professionals feel this pressure daily, and the urgency to find solutions together is greater than ever.

At Philips, we believe technology can play a key role in making healthcare future-proof, not as an end in itself, but as a means to better support healthcare professionals and help patients more effectively. This requires collaboration among healthcare institutions, governments, health insurers, banks, technology companies, and patients. Only by working together can we implement solutions for the challenges in healthcare.

In this tenth edition of the Future Health Index – the largest global study of its kind – we explore how countries worldwide are addressing healthcare challenges and the role artificial intelligence (AI) can play in this transformation. In the Netherlands, healthcare professionals are positive about working in healthcare and find satisfaction in caring for patients. However, they are also under significant pressure.

AI can be a solution to provide better care to more people. Dutch healthcare professionals are very optimistic about the potential of AI to improve patient outcomes compared to their international colleagues. However, they also express concerns about data bias and legal liability associated with AI use. Patients also see the potential benefits of AI in healthcare, provided there is human oversight.

The insights from this report indicate that we are at a tipping point. AI technology offers unprecedented possibilities, from automating administrative tasks to supporting clinical decision-making and enhancing the patient experience. However, technology alone is not enough. It requires vision, leadership, and cross-sector collaboration to translate this innovation into real impact.

At Philips, we are committed to providing better care for more people every day. With our technology, knowledge, and partnerships, we contribute to solutions that make a difference – for patients, healthcare professionals, and society as a whole. Dutch healthcare professionals also acknowledge this, stating they are well-involved in technology development and implementation. Yet, a significant portion feels that technology is not designed to meet their needs. This presents a challenge for us and the sector to improve together.

We invite you to read this anniversary edition of the Future Health Index and join us in thinking about the future of healthcare.

Léon Kempeneers
Managing Director
Philips Benelux

"Technology like AI offers unprecedented possibilities. It requires vision, leadership, and cross-sector collaboration to translate this innovation into real impact in practice."

Research Approach

This is the largest global study of its kind, analyzing the priorities and perspectives of healthcare professionals and patients. In this 10th edition, the Future Health Index 2025 examines how artificial intelligence (AI) can enable healthcare professionals to deliver better care to more people. The report highlights the key challenges facing healthcare professionals today. It also discusses their views on the rise of AI and identifies important gaps that need to be addressed to increase trust in the integration of AI in patient care.

We also examine the patient perspective, assessing how comfortable they are with AI in healthcare and identifying opportunities to strengthen their trust in technological advancements.

For this year's Future Health Index, we conducted our own quantitative research, involving approximately 1,900 healthcare professionals and over 16,000 patients across 16 countries.

Countries surveyed: Australia, Brazil, Canada, China, France, Germany, India, Indonesia, Japan, Netherlands, Saudi Arabia, Spain, South Africa, South Korea, United Kingdom, and United States.

Survey participants: Approximately 1,900 healthcare professionals and over 16,000 patients.

Chapter 1: The Possibilities of AI for Healthcare

The Dutch healthcare system continues to face significant pressure, impacting both healthcare professionals and patients. However, healthcare professionals are optimistic about AI's potential to streamline administrative tasks, improve patient care, and help prevent hospital admissions.

The Long Waiting Time for Patients

The aging population, the increase in chronic diseases, and a growing gap between patient demand and the availability of healthcare professionals are putting healthcare systems worldwide under increasing pressure. Last year's Future Health Index revealed that more than three-quarters of healthcare leaders globally reported care delays due to workforce shortages. This year's findings show that delays in patient care remain a serious problem worldwide. In the Netherlands, the average longest waiting time for a specialist appointment is 63 days. While this is below the global average of 70 days, it is significantly higher than in many Asian countries, such as China, Indonesia, and Japan, where the longest waiting times are 22 days or less.

65% of patients have waited to see a specialist.

Average longest waiting times for a specialist appointment (days):

• Australia: 81
• Brazil: 18
• Canada: 131
• China: 98
• France: 99
• Germany: 93
• India: 40
• Indonesia: 19
• Japan: 22
• Netherlands: 63
• Saudi Arabia: 51
• Spain: 40
• South Africa: 33
• South Korea: 128
• UK: 109
• USA: 70 (global average)

Despite the Pressure, Healthcare Professionals Find Meaning in Their Work

Most healthcare workers in the Netherlands have a positive outlook on their role: more than two-thirds feel valued (67%), motivated (70%), and confident (79%) in their work – all significantly higher than the global average. When asked about their work, 90% used exclusively positive words (compared to 65% globally).

Given the pressure healthcare professionals face, these high figures are encouraging. The Dutch healthcare sector is projected to face a shortage of 266,000 healthcare workers by 2035. Healthcare professionals are experiencing the consequences of these shortages. Four out of ten healthcare professionals now spend less time with patients and more time on administrative tasks compared to five years ago, while only a quarter spend more time with patients.

Globally, healthcare professionals who spend less time with patients than five years ago feel significantly more stressed compared to their colleagues. It distracts them from what they initially studied medicine for: helping patients recover and providing comfort and support (mentioned by 38% and 44% of Dutch healthcare professionals, respectively, as a source of joy and meaning).

What gives healthcare professionals joy and meaning in their work?

• 44%: Providing comfort and support to patients.
• 38%: Helping patients recover and improve their health.

Healthcare professionals are losing time for patients due to administration:

• 40% spend less time with patients and more time on administration.
• 34% spend equal time with patients and administration.
• 26% spend more time with patients and less time on administration.

Healthcare Professionals Develop New Technology, but Needs Remain Unmet

Healthcare professionals are open to new digital health technologies but are critical of their usefulness in daily medical practice. While 93% of healthcare professionals are involved in the development of new digital health technologies, significantly above the global average, only 43% state that the new technology is designed to meet their needs.

This indicates that, in many cases, there is still a significant gap in translating clinical needs into practical, usable solutions that seamlessly support the daily workflow of healthcare professionals. A deeper understanding is needed to bridge this gap and solve operational and logistical problems.

More than 9 out of 10 healthcare professionals are actively involved in developing new technology within their organizations.

More than 4 out of 10 healthcare professionals believe that new technologies are designed to meet their needs.

AI Can Help Address Challenges in Healthcare

Healthcare professionals in the Netherlands are convinced that AI, if implemented correctly, can solve many of the challenges they face. One of the benefits is that it can alleviate repetitive tasks and shorten procedure times, allowing departments to increase their capacity to treat more patients faster. They are concerned that slower AI adoption could lead to greater backlogs for patients (41%) and more burnout among clinical staff due to non-clinical tasks (44%).

Trust in AI's potential is much higher than the global average: In the Netherlands, 9 out of 10 healthcare professionals (91%) are optimistic about AI's ability to improve patient outcomes, compared to 79% globally. Furthermore, 94% of Dutch healthcare professionals indicate they understand where and how AI is being used in their department. This may contribute to their positive attitude towards the technology. This optimism applies to all workflows: 80% or more of Dutch healthcare professionals believe AI can effectively support virtually all facets of care, from logistics to diagnostic decision support, treatment planning, and monitoring.

Where AI Can Have a Positive Impact at the Departmental Level:

• 97%: Improving staff skills and knowledge.
• 96%: Improving patient access to care through increased capacity and throughput.
• 96%: Shorter procedure times and waiting times, plus more face-to-face time.
• 89%: Reducing administrative burdens and overtime.

The Potential of AI Beyond Hospital Walls: Reducing Hospital Admissions

Against the backdrop of rising healthcare costs, which are expected to double in the Netherlands between 2022 and 2050, the greatest potential for AI may lie in preventing people from being admitted to the hospital in the first place. More than eight out of ten healthcare professionals in the Netherlands believe that digital health technologies – including AI – will reduce hospital admissions in the future. Moreover, they are concerned that slower AI adoption could lead to missed opportunities for early intervention.

A particularly promising area for AI is remote patient monitoring. In the 2024 edition of the Future Health Index, we saw healthcare leaders investing in a wide range of remote patient monitoring solutions, driven by the growing demand for personalized, preventive home care. This year's findings show that these investments are embraced by healthcare workers, who believe that remote patient monitoring – combined with AI and predictive analytics – can reduce hospital admissions and save lives.

Healthcare professionals also recognize the benefits of wearable technology in supporting medication adherence (90%), saving time in monitoring patient health and enabling earlier intervention (89%), and helping patients take their diagnosis seriously (88%). However, assessing and interpreting all this data takes time and leads to an overload of information. This underscores the need for AI to generate meaningful insights for clinical care.

86% of healthcare professionals believe that digital health technologies will reduce the number of acute or emergency medical procedures and interventions.

84% of healthcare professionals believe that AI and predictive analytics can save lives by enabling earlier interventions.

Chapter 2: Optimally Utilizing AI for Healthcare

While AI is promising, many obstacles still stand in the way of successful implementation. These include improving data functionality, building trust through partnerships, and establishing clear guidelines. Patients emphasize the importance of maintaining human contact and prefer technologies that improve care rather than replace face-to-face interaction.

Addressing Data Problems

Reliable, representative, and easily accessible data forms the basis of AI systems. Three-quarters of Dutch healthcare professionals, however, are concerned about data bias in AI applications leading to greater disparities in health outcomes. This is significantly higher than in Germany (43%), Japan (49%), and South Korea (54%). To increase trust in AI, about one-third of Dutch healthcare professionals want certainty about the security of their data.

Data inefficiency is one of the biggest frustrations for both healthcare professionals and patients in the Netherlands. Approximately six out of ten (57%) indicate that it takes too many clicks to access the data they need to provide care to their patients, and 56% have lost clinical time due to problems with incomplete or inaccessible patient data. This highlights the need for better interoperability between systems. These issues underscore the urgent need to simplify data management and thus recover lost time.

56% of healthcare professionals have lost clinical time due to incomplete or inaccessible data.

20+ minutes of clinical time per day are lost by one-third (32%) of them.

61% of patients experience negative consequences from healthcare workers' lack of access to real-time data.

Building Trust Through Partnerships

As Dutch healthcare systems scale up their use of AI, partnerships are essential to build trust within the healthcare ecosystem, including organizations, healthcare professionals, and patients. As many as 92% of healthcare professionals state that they would have more confidence in using AI tools developed in collaboration with a trusted company.

Meanwhile, patients seek human oversight of AI to feel comfortable with its use in their treatment. Our findings indicate that patients prefer to receive information and reassurance about AI from their doctors and nurses, rather than from news channels or social media. This underscores the crucial role of healthcare professionals in building patient trust in AI.

Patients trust information about AI the most from healthcare professionals:

• Doctors: 74%
• Nurses: 70%
• News media: 49%
• Social media: 29%

Healthcare Professionals Want Clarity on Liability and Evidence

Another persistent concern among healthcare professionals is: who is responsible if an AI system makes a mistake in diagnosis or treatment? Issues such as hallucinations in generative AI systems that compromise accuracy and reliability are a major concern according to our research findings, due to the uncertainty surrounding legal liability.

Our research shows that Dutch healthcare professionals, in addition to clarity on legal liability, need clear guidelines for AI use and its limitations, scientific evidence for its effectiveness, and monitoring and evaluation.

Although nearly a quarter of healthcare professionals also want job security, their overall opinion is clear: With the right measures, AI-driven healthcare offers the possibility of improved care, efficiency, and better patient outcomes.

78% of healthcare professionals are concerned or uncertain about legal liability and AI.

What healthcare professionals need to build trust in AI:

• 39%: Continuous monitoring and evaluation of AI systems.
• 36%: Clear guidelines for AI use and limitations.
• 34%: Clarity on legal liability when using AI.
• 34%: Reassurance for staff regarding job security.
• 33%: Reassurance regarding data security.
• 31%: Scientific evidence to support effectiveness.
• 31%: Reliable IT/support helpdesk.
• 30%: Transparency about the basis of AI recommendations.
• 26%: Addressing data bias and data quality issues.

Maintaining Human Contact

Our research shows that nearly eight out of ten patients are happy with the use of more technology, such as AI, in healthcare if they see the benefits. This is not surprising, as the Netherlands has one of the highest levels of digital literacy in Europe.

However, it is also important to maintain the human touch in healthcare. Only 22% of patients would visit the doctor more often if they could have an audio or video appointment instead of an in-person visit. Nearly half (49%) are concerned that an audio or video appointment will not be as good as a face-to-face appointment.

Despite their enthusiasm for technological advancements, personal interactions remain crucial for patients. Interestingly, two-thirds (67%) of patients would like more choice between virtual and face-to-face care appointments, highlighting the need for flexible care models that seamlessly integrate AI and digital technologies.

49% of patients say that increased trust in technology means they spend less time with their doctor.

88% of patients are concerned about appointments that are not in person.

Trust in Data is Essential for the Success of Wearables

Our data also shows that wearable consumer products such as smartwatches and fitness trackers are used to a limited extent for medical treatments and prevention. Only 28% of Dutch patients in our survey use wearables, and just over four in ten have shared or would be willing to share their wearable data. These figures are lower than the global average. Four in ten patients use wearable devices, and more than six in ten have shared or would be willing to share their data. These findings highlight the need to address patients' concerns about wearable technology in healthcare to fully leverage its potential for preventive care.

Recommendations

How Can We Build Trust in AI for Healthcare Among Patients and Professionals?

1. Prioritize People in AI Design

   AI should be designed based on the needs of both patients and healthcare professionals. To create trust and acceptance, it is essential to involve the right stakeholders from the beginning and throughout the process. Solutions should seamlessly support patients' health routines and integrate into the workflows and IT infrastructures of healthcare. This creates a smooth experience for healthcare professionals and improves patient outcomes.

2. Enhance Human-AI Collaboration

   The true potential of AI lies in enhancing the skills of healthcare professionals and enabling patients and caregivers to manage their health and well-being. While AI agents can perform certain tasks autonomously, human oversight remains essential when it comes to health. Healthcare professionals play a crucial role in building patient trust by communicating transparently about AI's role, supported by comprehensive training from the start of their education.

3. Demonstrate Effectiveness and Fairness

   Both healthcare professionals and patients want assurance that AI works as intended, while regulators demand evidence that the technology meets safety and performance requirements. Consistent performance across all relevant patient groups and clinical contexts is essential, as are safeguards against bias to support non-discriminatory outcomes. The use of representative, high-quality datasets during development and validation can help reduce bias and ensure fair outcomes for every patient.

4. Enable Innovation with Clear Frameworks

   To accelerate the delivery of potentially life-saving AI to patients, regulations need to be adapted to strike a balance between the pace of innovation and safeguards that protect patients and build trust. Global harmonization of regulatory frameworks can reduce complexity and enable faster access to innovation without compromising patient safety. Approaches such as regulatory sandboxes can enable the responsible development and monitoring of AI while maintaining the consistent application of medical device regulations.

5. Build Strong Cross-Sector Partnerships

   In healthcare, no one can succeed alone. Close collaboration among all players in the ecosystem – including healthcare institutions and professionals, patient groups, payers, policymakers, regulators, researchers, and the health technology sector – is crucial to stimulate innovation and create solutions that meet the needs of stakeholders and build trust. Aligned goals and incentives, including payment models, are essential to focus on what matters most: improving the health and well-being of patients and healthcare professionals.

Appendices

Research Methodology

Accenture Song, a global technology-driven creative group, conducted two quantitative studies* using online surveys (CAWI).

The studies were conducted from December 2024 to March 2025 in 16 countries (Australia, Brazil, Canada, China**, France, Germany, India, Indonesia, Japan, Netherlands, Saudi Arabia, Spain, South Africa, South Korea, the United Kingdom, and the United States).

Survey 1:

1,926 healthcare professionals participated in a 15-minute online survey.

• The healthcare professionals comprised a mix of doctors (including surgeons), nurses, and medical assistants.
• Respondents worked across a broad range of specialties within the private and public healthcare system.

Survey 2:

16,144 patients aged 18 and older participated in a 10-minute online survey.

• Respondents were largely representative of their specific countries in terms of age and gender.
• 99% of respondents had visited a doctor in the past two years.

If relevant, surveys were translated into the local language. In some cases, specific questions were slightly adapted for relevance within specific countries. Care was taken to ensure the meaning of the question remained as close as possible to the original English version.

In both cases – healthcare professionals and patients – the sample sizes were weighted to ensure representative results at a global level.

* Two separate studies were conducted, but for simplicity, the report refers to data from a single 'survey'.

** The survey data is representative of mainland China only and does not include Taiwan or Hong Kong.

Weighting

Weighting is a statistical technique used to adjust sample data so that it accurately reflects the larger population. This process is crucial when certain groups are over- or under-represented in the sample compared to their actual share in the population.

• Improves accuracy: Weighting corrects for potential biases that may arise from unequal sample sizes across different markets.
• Ensures representative representation: This ensures that the insights obtained better align with the demographics and characteristics of the overall population.
• Enables comparability: By weighting the data, we can make fair comparisons between different markets and demographics, leading to more reliable conclusions.

The tables below show both the unweighted and weighted sample sizes, as well as the estimated margin of error*** at a 95% confidence level.

Please note that this report uses weighted data from both healthcare professional and patient surveys to provide insights that are representative of the various markets analyzed.

Healthcare Professionals Survey:

Patient Survey:

*** The estimated margin of error is the margin of error that would apply to a sample of this size for the respondent population in each country.

Glossary

Artificial Intelligence (AI)

An AI system is a machine-based system that, for explicit or implicit purposes, derives from the input it receives how output should be generated, such as predictions, content, recommendations, or decisions that can affect physical or virtual environments. Different AI systems vary in their degree of autonomy and adaptability after implementation.

AI Algorithms

AI algorithms instruct a computer on how to make decisions, perform a function, or carry out another task autonomously.

AI Hallucinations

Answers produced by AI systems that are misleading, incorrect, or nonsensical, but are presented as facts.

Automation

The use of technology and software solutions to perform tasks and processes with limited human involvement. This can include the deployment of digital tools, machines, and computer systems to streamline and optimize various aspects of healthcare delivery, administration, and management.

Data

Used here to refer to a variety of clinical and/or operational information collected from various sources, including, but not limited to, electronic medical records (EMRs), medical devices, and workflow management tools.

Data Bias

An error that occurs when certain elements in a dataset are missing, underrepresented, or overrepresented.

Digital Health Technology

A technology that transmits, shares, and/or analyzes health data. The technology can take various forms, including (but not limited to) home health monitors, digital health records, hospital/healthcare facility equipment, and health or fitness trackers.

Generative AI

AI systems that can create original content in response to a user's request or prompt.

Healthcare Leader

An executive or senior manager who works in a hospital, medical practice, imaging center/office laboratory, or emergency department and is the ultimate decision-maker or influences decision-making.

Healthcare Organization

The hospital or healthcare facility where the healthcare professional works or is employed.

Healthcare Professional

Individuals directly involved in providing healthcare to patients (including doctors, nurses, surgeons, specialists, technologists, technicians, etc.).

Out-of-Hospital Care

Services provided outside the traditional hospital setting, such as at home, in clinics, outpatient clinics, or other community locations, either in person or virtually.

Patient Flow

The efficiency with which a patient moves through a healthcare facility, from arrival to discharge.

Predictive Analytics

A branch of advanced analytics that makes predictions about future events, behaviors, and outcomes.

Remote Patient Monitoring

Technology that allows patients' health to be monitored and diagnosed remotely.

Specialist

A doctor or other healthcare professional who is trained and qualified in a specific field. Examples of specialists include oncologists (cancer specialists) and cardiologists (heart specialists).

Staff

Includes all employees within a healthcare organization, such as healthcare professionals, IT, finance, administrative support, facilities, etc.

Workflows

A process involving a series of tasks performed by different people within and across work environments to deliver care. To complete each task, actions may be required from one person, between people, or between organizations. This can occur sequentially or concurrently.