BACKGROUND:Clinical care in modern intensive care units (ICUs) combines multidisciplinary expertise and a complex array of technologies. These technologies have clearly advanced the ability of clinicians to do more for patients, yet so much equipment also presents the possibility for cognitive overload.PURPOSE:The aim of this study was to investigate clinicians' experiences with and perceptions of technology in ICUs.METHODOLOGY/APPROACH:We analyzed qualitative data from 30 interviews with ICU clinicians and frontline managers within four ICUs.RESULTS:Our interviews identified three main challenges associated with technology in the ICU: (a) too many technologies and too much data; (b) inconsistent and inaccurate technologies; and (c) not enough integration among technologies, alignment with clinical workflows, and support for clinician identities. To address these challenges, interviewees highlighted mitigation strategies to address both social and technical systems and to achieve joint optimization.CONCLUSION:When new technologies are added to the ICU, they have potential both to improve and to disrupt patient care. To successfully implement technologies in the ICU, clinicians' perspectives are crucial. Understanding clinicians' perspectives can help limit the disruptive effects of new technologies, so clinicians can focus their time and attention on providing care to patients.PRACTICE IMPLICATIONS:As technology and data continue to play an increasingly important role in ICU care, everyone involved in the design, development, approval, implementation, and use of technology should work together to apply a sociotechnical systems approach to reduce possible negative effects on clinical care for critically ill patients.

Background and Aims:The progress of artificial intelligence (AI) in endoscopy is at a crossroads. The positive results of randomized controlled trials of computer-aided detection (CADe) have not been replicated in multiple pragmatic CADe trials, including ours. This gap between efficacy and effectiveness remains to be understood. We surveyed and interviewed our trial's colonoscopists to gain insight into human-AI interactions. Methods:We used a sequential, mixed-methodology design. After the trial, we administered Survey 1, focusing on attitudes and beliefs before and after trying CADe. The trial's null results were disclosed, and we then administered Survey 2 and conducted open-ended interviews, focusing on reactions to the null results. Responses were analyzed overall and by baseline adenoma detection rate (ADR) tertile. We identified key themes using thematic analysis and qualitative software. Results:Nearly all colonoscopists responded (22 and 21 of 24 [92% and 88%] for Surveys 1 and 2, respectively). Most (96%) regarded endoscopic ability as critical to their professional identity. Large majorities conveyed trust in and enthusiasm for AI before and after trying CADe (82%-87%) and desired to have CADe available (72%). Nearly two-thirds (62%) were surprised by the null results. There were few differences by ADR. No unifying explanation for the null results emerged from surveys or individual interviews. Colonoscopists expressed a range of expectations for AI in endoscopy. Conclusions:Lack of enthusiasm or mistrust of AI/CADe do not explain our pragmatic CADe trial's null results. AI may need to target dimensions beyond optical recognition to realize its promise in endoscopy.

Research on data annotation for artificial intelligence (AI) has demonstrated that biases, power, and culture impact the ways that annotators apply labels to data and subsequently affect downstream AI systems. However, annotators can only apply labels that are available to them in the annotation classification scheme. Drawing on a 3-year ethnographic study of an R&D collaboration between medical and AI researchers, we argue that the construction of the classification schema itself -- decisions about what kinds of data can and cannot be collected, what activities can and cannot be detected in the data, what the possible annotation classes ought to be, and the rules by which an item ought to be classified into each class -- dramatically shape the annotation process, and through it, the AI. We draw on Bowker and Star's [9] classification theory to detail how the creation of a training data codebook for a computer vision algorithm in hospital intensive care units (ICUs) evolved from its original, clinically-driven goal of classifying complex clinical activities into a narrower goal of identifying physical objects and simpler activities in the ICU. This work reinforces how trade-offs and decisions made long before annotators begin labeling data are highly consequential to the resulting AI system.

Stanford University, USA; University of California San Diego, USA.

Hospital providers often use workarounds to circumvent processes so that patients can receive care. Workarounds in response to operational failures enable care to continue and therefore may be indicative of workers' commitment. On the other hand, workarounds in the absence of operational failures may signal an ineffective approach associated with lower quality of care and worse patient outcomes. Working closely with healthcare providers, we developed a survey to measure workaround behaviors and operational failures on medical/surgical units. The lead author surveyed over 4,000 nurses from 63 hospitals throughout the United States. We matched this data with audit data on the incidence of pressure injuries among over 21,000 patients on 262 nursing units in 56 survey hospitals. Hospital‐acquired pressure injuries are a significant risk to patient health and hospital costs. We do not find support for our hypothesis that workarounds are associated with a higher rate of hospital‐acquired pressure injuries. However, when we take into account the moderating role of operational failures on the relationship between workarounds and pressure injuries, we find significant results. When nursing units have lower levels of operational failures, workarounds are associated with higher rates of hospital‐acquired pressure injuries. Our results provide evidence that workarounds may be associated with negative patient outcomes, if they stem from a process‐avoiding approach. The best results can be achieved by reducing both operational failures and workarounds via instilling a process‐focused approach.

How much media information, of what kinds and delivered on what devices, do Americans consume? We measure each consumer information stream using three different measures of what is consumed: hours, words and bytes, and sum across each recipient. We estimate that in 2008 Americans consumed about 1.3 trillion hours of information outside of work, an average of almost 12 hours per person per day. Media consumption totaled 3.6 zettabytes and 1,080 trillion words, corresponding to 100,500 words and 34 gigabytes for the average person on an average day. We measure information flows, not stocks, and find that information consumption measured in bytes grew at an annual rate of 5.4% from 1980 to 2008, only a few percentage points greater than GDP growth over this period. We report our findings for different media types, including television, the Internet and computer games, and discuss the utility of analyzing contrasting measures of information consumption in totaling how much media information Americans consume.

How much media information, of what kinds, and delivered on what devices, do Americans consume? We measure each consumer information stream using 3 different measures of what is consumed-hours, words, and bytes-and sum across each recipient. We estimate that, in 2008, Americans consumed about 1.3 trillion hours of information outside of work, an average of almost 12 hours per person per day. Media consumption totaled 3.6 zettabytes and 1,080 trillion words, corresponding to 100,500 words and 34 gigabytes for the average person on an average day. We measure information flows, not stocks, and find that information consumption measured in bytes grew at an annual rate of 5.4% from 1980 to 2008, only a few percentage points greater than GDP growth over this period. We report our findings for different media types, including television, the Internet and computer games, and discuss the utility of analyzing contrasting measures of information consumption in totaling how much media information Americans consume.

. Severe haemophilia results in increased mortality and poorer quality of life. Factor prophylaxis leads to a more normal life, but is very costly; most of the cost is due to the high cost of replacement factor. Despite its high cost, factor prophylaxis has been adopted throughout the developed world even in different health care systems. We argue that there are at least five possible reasons why societies may value factor prophylaxis despite its cost: (i) it is directed towards an inherited disease, (ii) the treatment is largely directed towards children, (iii) the disease is rare and the overall cost to society is small, (iv) the treatment is preventative, and v) the high cost is largely the result of providing safe products. In an era of rising health care costs, there is a strong research agenda to establish the factors that determine the value of expensive therapies for rare diseases like haemophilia.

Facebook and other social networking sites (SNS) do not object to all forms of nation-state or international legislation. Facebook must manage the perception that these fragments of a user's digital self presented on Facebook can be picked up and reconstituted by others with ill effects for the user. In reality, Facebook is private. Evidence of the importance companies like Facebook place upon maintaining this architecture of disclosure is their strong lobbying efforts against privacy legislation in the United States, as the "do not track" legislation example can attest. Addressing challenges with how Facebook uses information requires a global response. Government regulation is a minor challenge when compared with larger societal trends that may work against Facebook's architecture of disclosure. Facebook as a political tool provides a platform for the airing of marginalized voices. An enhancement of a democratic …

In the first 5 years of virtual reality application research at the California Institute for Telecommunications and Information Technology (Calit2), we created numerous software applications for virtual environments. Calit2 has one of the most advanced virtual reality laboratories with the five-walled StarCAVE and the world's first passive stereo, LCD panel-based immersive virtual reality system, the NexCAVE. The combination of cutting edge hardware, direct access to world class researchers on the campus of UCSD, and Calit2's mission to bring the first two together to make new advances at the intersection of these disciplines enabled us to research the future of scientific virtual reality applications. This chapter reports on some of the most notable applications we developed.

In most technologies and most industries, experiments play a central role in organizational learning as a source of knowledge and as a check before changes are implemented. There are four primary types of experiments: controlled, natural, ad-hoc, and evolutionary operation. This paper discusses factors that affect learning by experimentation and how they influence learning rates. In some cases, new ways of experimenting can create an order of magnitude improvement in the rate of learning. On the other hand, some situations are inherently hard to run experiments on, and therefore learning remains slow until basic obstacles are solved. Examples of experimentation are discussed in four domains: product development, manufacturing, consumer marketing, and medical trials.

Most flying activities today are based on extensive knowledge, embodied in smart devices and algorithms to supplement and sometimes supplant pilots. Control developed in five principal stages. Initially flying was a pure craft, with high variability and low safety. In the 1930s, rules were developed, and instruments replaced human senses. Rule-based control proved inadequate to handle the complexity of WW II aircraft, and the result was the development of standard procedures. These three stages all used the human pilot to do the actual control. Two further stages use automated control. But higher stages of flying control revert to lower stages in some situations.

Summary. Background: The need for clearly reported studies evaluating the cost of prophylaxis and its overall outcomes has been recommended from previous literature. Objectives: To establish minimal ‘‘core standards’’ that can be followed when conducting and reporting economic evaluations of hemophilia prophylaxis. Methods: Ten members of the IPSG Economic Analysis Working Group participated in a consensus process using the Nominal Groups Technique (NGT). The following topics relating to the economic analysis of prophylaxis studies were addressed; Whose perspective should be taken? Which is the best methodological approach? Is micro‐ or macro‐costing the best costing strategy? What information must be presented about costs and outcomes in order to facilitate local and international interpretation? Results: The group suggests studies on the economic impact of prophylaxis should be viewed from a societal perspective and be reported using a Cost Utility Analysis (CUA) (with consideration of also reporting Cost Benefit Analysis [CBA]). All costs that exceed $500 should be used to measure the costs of prophylaxis (macro strategy) including items such as clotting factor costs, hospitalizations, surgical procedures, productivity loss and number of days lost from school or work. Generic and disease specific quality of lífe and utility measures should be used to report the outcomes of the study. Conclusions: The IPSG has suggested minimal core standards to be applied to the reporting of economic evaluations of hemophilia prophylaxis. Standardized reporting will facilitate the comparison of studies and will allow for more rational policy decisions and treatment choices.

Making goods evolved over several centuries from craft production to complex and highly automated manufacturing processes.A companion paper by R. Jaikumar documents the transformation of firearms manufacture through six distinct epochs, each accompanied by radical changes in the nature of work.These shifts were enabled by corresponding changes in technological knowledge.This paper models knowledge about manufacturing methods as a directed graph of cause-effect relationships.Increasing knowledge corresponds to more numerous variables (nodes) and relationships (arcs).The more dense the graph, the more variables can be monitored and controlled, with greater precision.This enables higher production speeds, tighter tolerances, and higher quality.Changes in knowledge from epoch to epoch tend to follow consistent patterns.More is learned about key classes of phenomena, including measurement methods, feedback control methods, and disturbances.As knowledge increases, control becomes more formal, and operator discretion is reduced or shifted to other types of activity.Increasing knowledge and control are two dimensions of a shift from art towards science.Evolution from art to science is not monotonic.The knowledge graphs of new processes are riddled with holes; dozens of new variables must be identified, understood, and controlled.Frederick Taylor pioneered three key methods of developing causal knowledge in such situations: reductionism, using systems of quantitative equations to express knowledge, and learning by systematic experimentation.Using causal networks to formally model knowledge appears to also fit other kinds of technology.But even as vital aspects of manufacturing verge on "full science," other technological activities will remain nearer to art, as for them complete knowledge is unapproachable.3 Know-how and know-why are often referred to as procedural knowledge and causal knowledge.See the discussion of [23] later.An additional category is declarative knowledge. 4Many other ways of classifying knowledge are used.For example, the distinction between collective and individual knowledge is important for designing knowledge management systems.

Rapid change in the geographical location of production raises important questions regarding the welfare, development potential, and competitive position of different countries and regions. This paper explores in detail the geography of economic activity in a specific industry, the hard disk drive (HDD) component of the computer industry. Firms in the HDD industry are breaking the production system into ever smaller distinct steps, and spreading the physical location of these steps around the world. Firms from the United States dominate the industry. Our findings suggest that globalization has enabled US firms to sustain their dominant position in the industry, preserve employment in the United States (and possibly expand it), and increase employment worldwide, most notably in Southeast Asia.