Characteristics and outcomes of critically-ill medical patients admitted to a tertiary medical center with restricted ICU bed capacity.

BACKGROUND: In the emergency department (ED) critically-ill medical patients are treated in the resuscitation room (RR). No studies described the outcomes of critically-ill RR patients admitted to a hospital with low capacity of intensive care unit (ICU) beds. METHODS: We included all medical patients above 18 who were admitted to a RR of a tertiary hospital during 2011-2012. We conducted multivariate logistic and Cox regressions and propensity score (PS) matched analysis to analyze parameters associated with the study outcomes. RESULTS: In-hospital mortality rate was 32.4% in ICU admitted patients compared to 52.0% of the non-ICU critically-ill patients (p<0.001). Age above 80, female and recent ED encounters were associated with non-ICU admissions (p<0.05 for all). ICU admission had a statistically significant effect on in-hospital mortality in PS matched analysis (OR 0.36, 95% CI 0.21-0.61). A marginal effect was evident in one-year survival in PS matched landmark analysis (HR 0.50 95% CI 0.23-1.06). CONCLUSION: ED critically-ill medical patients who were treated in the RR had high mortality rates in an institute with restricted ICU beds availability. However, those who were admitted to an ICU showed prolonged short and perhaps long term survival compared to those who were not.

Balancing volume and duration of information consumption by physicians: The case of health information exchange in critical care.

BACKGROUND: The realization of the potential benefits of health information exchange systems (HIEs) for emergency departments (EDs) depends on the way these systems are actually used. The attributes of volume of information and duration of information processing are important for the study of HIE use patterns in the ED, as cognitive load and time constraints may result in a trade-off between these attributes. Experts and non-experts often use different problem-solving strategies, which may be consequential for their system use patterns. Little previous research focuses on the trade-off between volume and duration of system use or on the factors that affect it, including user expertise. OBJECTIVES: This study aims at exploring the trade-off of volume and duration of use, examining whether this relationship differs between experts and non-experts, and identifying factors that are associated with use patterns characterized by volume and duration. METHODS: The research objectives are pursued in the context of critically-ill patients, treated at a busy ED in the period 2010-2012. The primary source of internal and external data is an HIE linked to 14 hospitals, over 1300 clinics, and other clinical facilities. We define four use profiles based on the attributes of duration and volume: quick and basic, quick and deep, slow and basic, and slow and deep. The volume and duration of use are computed using HIE log files as the number of screens and the time per screen, respectively. Each session is then classified into a specific profile based on distances from predefined profile centroids. Experts are physicians that are board-certified in emergency medicine. We test the distribution of use profiles and their associations with multiple variables that describe the patient, physician, situation, information available in the HIE system, and use dynamics within the encounter. RESULTS: The quick and basic profile is the most prevalent. While available admission summaries are associated with quick and basic use, lab and imaging results are associated with slower or deeper use. Physicians who are the first to use the system or are sole users during an encounter are less inclined to quick and deep use. These effects are intensified for experts. DISCUSSION: A trade-off between volume and duration is identified. While system use is overall similar for experts and non-experts, the circumstances in which a certain profile is more likely to be observed vary across these two groups. Information availability and multiple-physician dynamics within the encounter emerge as important for the prediction of use profiles. The findings of this study provide implications for the design, implementation, and research of HIE use.

Use patterns of health information exchange systems and admission decisions: Reductionistic and configurational approaches.

BACKGROUND: Research that endeavors to identify the value of electronic health information exchange (HIE) systems to the healthcare industry and, specifically, to clinical decision making is often inconclusive or theory-based. Studies seeking to identify how clinical decisions relate to patterns of actual HIE use, often by analyzing system log files, generally rely on dichotomous distinctions between system use and no-use, disregard the availability of information in the system, and control for few user characteristics. OBJECTIVE: We aim at empirically exploring the associations between use patterns of HIE systems and subsequent clinical decisions on the basis of broad definitions of use patterns, available information, and control variables. METHODS: We examine the decision to admit critically-ill patients either to the intensive care unit (ICU) or to another ward at a busy emergency department in the period 2010-2012. Using HIE log files, use patterns are characterized by the variables of number of users, volume, diversity, granularity, duration, and content. We test the association between HIE use patterns and the admission decision, after controlling for multiple demographic, clinical, physician, and situational variables and for available HIE information. This association is examined by taking a reductionistic approach that focuses on independent use variables and a configurational approach that focuses on use profiles. RESULTS: Five use profiles were identified, the largest of which (46.95% of encounters) described basic HIE access. ICU admission is more probable when the HIE system is perused by multiple users (odds increase by 31%) and when use profiles include prolonged screen viewing (odds increase by 159%) or access to diverse and multiple types of information, specifically on test results, procedures, and previous encounters. DISCUSSION: Reductionistic and configurational approaches yield complementary insights, which advance the understanding of how actual HIE use is associated with clinical decision making. The study shows that congruent profiles of HIE use enhance the predictability of the admission decision beyond what can be explained by independent variables of HIE use.

Use patterns of health information exchange through a multidimensional lens: conceptual framework and empirical validation.

Insights about patterns of system use are often gained through the analysis of system log files, which record the actual behavior of users. In a clinical context, however, few attempts have been made to typify system use through log file analysis. The present study offers a framework for identifying, describing, and discerning among patterns of use of a clinical information retrieval system. We use the session attributes of volume, diversity, granularity, duration, and content to define a multidimensional space in which each specific session can be positioned. We also describe an analytical method for identifying the common archetypes of system use in this multidimensional space. We demonstrate the value of the proposed framework with a log file of the use of a health information exchange (HIE) system by physicians in an emergency department (ED) of a large Israeli hospital. The analysis reveals five distinct patterns of system use, which have yet to be described in the relevant literature. The results of this study have the potential to inform the design of HIE systems for efficient and effective use, thus increasing their contribution to the clinical decision-making process.

A design process for using normative models in shared decision making: a case study in the context of prenatal testing.

BACKGROUND: Shared decision making (SDM) encourages the patient to play a more active role in the process of medical consultation and its primary objective is to find the best treatment for a specific patient. Recent findings, however, show that patient preferences cannot be easily or accurately judged on the basis of communicative exchange during routine office visits, even for patients who seek to expand their role in medical decision making (MDM). OBJECTIVE: The objective of this study is to improve the quality of patient-physician communication by developing a novel design process for SDM and then demonstrating, through a case study, the applicability of this process in enabling the use of a normative model for a specific medical situation. DESIGN: Our design process goes through the following stages: definition of medical situation and decision problem, development/identification of normative model, adaptation of normative model, empirical analysis and development of decision support systems (DSS) tools that facilitate the SDM process in the specific medical situation. CASE STUDY: This study demonstrates the applicability of the process through the implementation of the general normative theory of MDM under uncertainty for the medical-financial dilemma of choosing a physician to perform amniocentesis. DISCUSSION: The use of normative models in SDM raises several issues, such as the goal of the normative model, the relation between the goals of prediction and recommendation, and the general question of whether it is valid to use a normative model for people who do not behave according to the model's assumptions.