he prospect of improved clinical outcomes and more efficient health systems has fueled a rapid rise in the development and evaluation of AI systems over the last decade. Because most AI systems within healthcare are complex interventions designed as clinical decision support systems, rather than autonomous agents, the interactions among the AI systems, their users and the implementation environments are defining components of the AI interventions' overall potential effectiveness. Therefore, bringing AI systems from mathematical performance to clinical utility needs an adapted, stepwise implementation and evaluation pathway, addressing the complexity of this collaboration between two independent forms of intelligence, beyond measures of effectiveness alone 1 . Despite indications that some AI-based algorithms now match the accuracy of human experts within preclinical in silico studies 2 , there

Technology is changing at a rapid rate, opening up new possibilities within the health care domain. Advances such as open source hardware, personal medical devices, and mobile phone apps are creating opportunities for custom-made medical devices and personalized care. However, they also introduce new challenges in balancing the need for regulation (ensuring safety and performance) with the need to innovate flexibly and efficiently. Compared with the emergence of new technologies, health technology design standards and regulations evolve slowly, and therefore, it can be difficult to apply these standards to the latest developments. For example, current regulations may not be suitable for approaches involving open source hardware, an increasingly popular way to create medical devices in the maker community. Medical device standards may not be flexible enough when evaluating the usability of mobile medical devices that can be used in a multitude of different ways, outside of clinical settings. Similarly, while regulatory guidance has been updated to address the proliferation of health-related mobile phone apps, it can be hard to know if and when these regulations apply. In this viewpoint, we present three examples of novel medical technologies to illustrate the types of regulatory issues that arise in the current environment. We also suggest opportunities for support, such as advances in the way we review and monitor medical technologies.

This paper empirically investigates the relationship between institutional holdings and capital structure. Institutions may affect capital structure through their monitoring and information-gathering roles. At the same time, institutions may gravitate toward firms with specific capital structures, forming leverage-based investment clienteles. Using implied trades generated from mutual fund outflows as an instrument for institutional holdings, and a semi-natural experiment in which addition to the S&P 500 Index provides an exogenous shock to institutional holdings, we find that institutional holdings are a significant determinant of firms' capital structures: A change in institutional holdings causes an opposite change in leverage. Moreover, using dynamic panel estimation, we find that while institutions affect capital structure decisions, changes in leverage do not affect institutional holdings, and there is no evidence of a clientele effect. Finally, we find that firms lower their leverage in response to increased institutional holdings by becoming more likely to issue equity, and less likely to increase debt. While our findings are consistent with models in which institutions substitute for debt by monitoring and reducing information asymmetry problems, further evidence suggests that the effect on asymmetric information dominates.

In 2004, the National Patient Safety Agency (NSPA) released a safety alert relating to the management and use of infusion devices in England and Wales. The alert called for the standardisation of infusion devices and a consideration of using centralised equipment systems to manage device storage. There has also been growing interest in smart pump technology, such as dose error reduction software (DERS) as a way to reduce IV medication errors. However, questions remain about the progress that has been made towards infusion device standardisation and the adoption of DERS.We report on the results of a survey investigating the extent to which the standardisation of infusion devices has occurred in the last 10 years; how far centralised equipment libraries are being used in practice; and about the prevalence of DERS use within the UK.The findings indicate that while reported standardisation levels are high, the use of centralised equipment libraries remains low, as does DERS usage. 2 Key phrases1. Infusion device standardisation, the use of centralised equipment libraries and DERS have all been suggested as ways to improve patient safety but there has been little research on establishing the prevalence of all three on a national level.2. Progress has been made towards infusion device standardisation, however "standardisation" does not always mean that only one type of device is being used, and there is still some variability in the devices used across whole organisations.3. Due to specific clinical areas requiring different devices or alternative configurations of the same device centralised equipment libraries are not the most common method of device storage management across entire hospitals.4. Due to the significant practical and organisational challenges that face institutions wishing to implement DERS, only a small number of hospitals are using this technology, especially across entire trusts and health boards.5. Obstacles to the implementation of DERS include existing device contracts, the significant time and resources required, not being convinced of the technology, and complications related to a lack of standardisation.3