1. Consumers have doubts about using GenAI for health information

Consumer adoption of GenAI for health reasons declined from 40 per cent in 2023 to 37 per cent in 2024, according to a survey of 2,000 adults in the US conducted by the Deloitte Center for Health Solutions. Deloitte noted that “one of the most prominent and growing reasons for the stagnant adoption is distrust in the information that the tool produces… When asked why they’re not using gen AI for health and wellness purposes, more consumers chose ‘I don’t trust the information’ in this year’s survey (30%) than they did in 2023 (23%).”

2. Public health R&D produces ‘blockbuster returns’, analysis finds

US government funding into global health research over the past 16 years has led to “dozens of breakthrough innovations” and billions of dollars in economic benefits while also helping to spur billions more in industry investments, an analysis by the Global Health Technologies Coalition revealed. “Our analysis shows that the $46 billion invested by the US government between 2007 and 2022 has not only delivered a steady stream of new drugs, vaccines and prevention tools, but has also supported basic research that is expected to attract an additional $102 billion investment from industry,” says Nick Chapman, CEO of Policy Cures Research, one of the coalition’s global network partners.

3. LLMs could reduce doctor workloads, but need editing for safety

Large language models could reduce doctors’ workloads and provide informative messages for doctors responding to patient inquiries, a study by Mass General Brigham found. The study used OpenAI’s GPT-4 to model possible responses to patients with cancer, and those responses tended to be longer, more educational but less directive than physician-written replies. However, the researchers also found that, “[i]f left unedited, 7.1 percent of LLM-generated responses could pose a risk to the patient and 0.6 percent of responses could pose a risk of death, most often because GPT-4’s response failed to urgently instruct the patient to seek immediate medical care.”

4. Healthcare should focus on helpful, rather than cutting-edge, tech: Survey

Electronic health records (EHRs) and digital communication tools for healthcare professionals are currently the most effective technologies for helping to save staff time, according to a survey of clinical staff across the UK conducted by The Health Foundation. The top time-saving tool identified by respondents was videoconferencing for discussions with colleagues (51 per cent), followed by EHRs (50 per cent) and digital messaging tools (29 per cent). “This contrasts with the typical health policy emphasis on cutting-edge clinical and patient-facing technologies,” The Health Foundation noted. “Maximising opportunities to free up time in the NHS will require focusing on technologies that can help with administrative and operational tasks and inter-professional communication as well as clinical tasks.”

5. AI analysis of social media posts accurately assesses community rates of depression, anxiety

Compared to the common method of population surveys, using AI in combination with analysis of social media posts improved assessments of depression and anxiety across communities by 10 per cent, a study led by researchers at Stony Brook University found. The study analysed nearly 1 billion Tweets from more than 2 million users across every state in the US.

6. US doctors ready to use GenAI, but want information transparency

Although 40 per cent of US doctors surveyed by Wolters Kluwer say they are ready to use GenAI to interact with patients, many physicians remain wary. The survey found that, before they would use GenAI outputs for clinical decisions, 91 per cent of respondents want assurances that the information sources were created by medical experts. And 89 per cent say GenAI tool vendors should be transparent about information sources.

7. Automation, virtual care needed in face of staff shortages: Survey

Technology is needed to address healthcare staff shortages that are hurting employee morale and contributing to delays in care, according to a Royal Philips survey of nearly 3,000 healthcare leaders across 14 countries. Respondents say automating repetitive tasks (88 per cent) is critical and also support remote patient monitoring for chronic disease management (45 per cent) and the use of virtual care (40 per cent). And almost all (96 per cent) believe that data-driven insights could help to reduce disparities in health outcomes.

8. Nursing professionals question AI’s potential impact on perceptions of empathy

A survey of more than 1,100 nursing professionals and students found that more than half have reservations about the use of AI in healthcare and 38 per cent question how the technology might benefit their field. Among their concerns—identified in a survey by Cross Country Healthcare and Florida Atlantic University’s Christine E. Lynn College of Nursing—are job displacement risks, data security issues and how AI might lead to a perceived lack of empathy.

9. Asia-Pacific survey shows wide support for sustainable healthcare practices

Of more than 3,000 people surveyed by Royal Philips and Kantar Profiles Network across the Asia-Pacific, 85 per cent say it’s important to adopt sustainable healthcare practices and 87 per cent acknowledge how climate impacts and health are connected. And almost all (93 per cent) believe that technology will be critical for achieving sustainable healthcare practices. Among the technologies they identified were efficient usage and logistics (27 per cent), more efficient use of energy (25 per cent) and reduced carbon dioxide emissions and other strategies such as virtual care (20 per cent).

10. Many consumers ready to share personal data for accurate predictions of health risks

More than half of healthcare consumers in the US say they are open to sharing personal health data to get more accurate predictions about their risk for heart and cardiovascular conditions (61 per cent), cancer (57 per cent) or diabetes (52 per cent), a survey by Huron Consulting Group found. The survey also identified nearly 60 per cent of respondents as being “moderately or hyper-digital based on their usage and preferences for digital tools and services.”