Visibility-based layout of a hospital unit - An optimization approach.

A patient fall is one of the adverse events in an inpatient unit of a hospital that can lead to disability and/or mortality. The medical literature suggests that increased visibility of patients by unit nurses is essential to improve patient monitoring and, in turn, reduce falls. However, such research has been descriptive in nature and does not provide an understanding of the characteristics of an optimal inpatient unit layout from a visibility-standpoint. To fill this gap, we adopt an interdisciplinary approach that combines the human field of view with facility layout design approaches. Specifically, we propose a bi-objective optimization model that jointly determines the optimal (i) location of a nurse in a nursing station and (ii) orientation of a patient's bed in a room for a given layout. The two objectives are maximizing the total visibility of all patients across patient rooms and minimizing inequity in visibility among those patients. We consider three different layout types, L-shaped, I-shaped, and Radial; these shapes exhibit the section of an inpatient unit that a nurse oversees. To estimate visibility, we employ the ray casting algorithm to quantify the visible target in a room when viewed by the nurse from the nursing station. The algorithm considers nurses' horizontal visual field and their depth of vision. Owing to the difficulty in solving the bi-objective model, we also propose a Multi-Objective Particle Swarm Optimization (MOPSO) heuristic to find (near) optimal solutions. Our findings suggest that the Radial layout appears to outperform the other two layouts in terms of the visibility-based objectives. We found that with a Radial layout, there can be an improvement of up to 50% in equity measure compared to an I-shaped layout. Similar improvements were observed when compared to the L-shaped layout as well. Further, the position of the patient's bed plays a role in maximizing the visibility of the patient's room. Insights from our work will enable understanding and quantifying the relationship between a physical layout and the corresponding provider-to-patient visibility to reduce adverse events.

Interruptions in a dental setting and evaluating the efficacy of an intervention: A pilot study.

INTRODUCTION: Interruptions during dental treatment are frequent, and often impact provider satisfaction and processing times We investigate the source and duration of such interruptions at a German dental clinic. METHODS: A pre-post approach was adopted at this dental clinic. This included direct observations of 3 dentists and 3 dental hygienists, and a survey of providers. Following that, an intervention (switchable 'Do Not Enter' sign) was chosen, and a pilot study was conducted to evaluate if the chosen intervention can reduce processing time and improve provider satisfaction. Additional observations and surveys were performed afterwards. RESULTS: Pre-intervention data indicated that interruptions have the highest negative impact on provider satisfaction at this clinic as well as on processing time during longer and more complex treatments, where a minor error due to an interruption could lead to rework of 30 minutes and more. The total number of interruptions dropped by 72.5% after the intervention, short interruptions (< 1min) by 86%. Provider survey indicated improvement due to the intervention in perceived workload, provider work satisfaction, patient safety and stress. CONCLUSIONS: This study demonstrates that a switchable sign can substantially reduce the number of interruptions in this dental clinic. It also shows the potential of improving the work environment by reducing interruptions to the dental providers.

Locating trauma centers considering patient safety.

Trauma continues to be the leading cause of death and disability in the U.S. for those under the age of 44, making it a prominent public health problem. Recent literature suggests that geographical maldistribution of Trauma Centers (TCs), and the resultant increase of the access time to the nearest TC, could impact patient safety and increase disability or mortality. To address this issue, we introduce the Trauma Center Location Problem (TCLP) that determines the optimal number and location of TCs in order to improve patient safety. We model patient safety through a surrogate measure of mistriages, which refers to a mismatch in the injury severity of a trauma patient and the destination hospital. Our proposed bi-objective optimization model directly accounts for the two types of mistriages, system-related under-triage (srUT) and over-triage (srOT), both of which are estimated using a notional tasking algorithm. We propose a heuristic based on the Particle Swarm Optimization framework to efficiently derive a near-optimal solution to the TCLP for realistic problem sizes. Based on 2012 data from the state of Ohio, we observe that the solutions are sensitive to the choice of weights for srUT and srOT, volume requirements at a TC, and the two thresholds used to mimic EMS decisions. Using our approach to optimize that network resulted in over 31.5% reduction in the objective with only 1 additional TC; redistribution of the existing 21 TCs led to 30.4% reduction.

Nurses' work with interruptions: an objective model for testing interventions.

Interruptions experienced by nurses may lead to errors as their focus and attention to multiple patient needs are disrupted. As quantitative models to understand the dynamics of interruptions are lacking, the objective of this study is a model of a nurse's work with interruptions, generating insights into the onset of interruptions and evaluating suggested interventions. We observed nurses in a US Level I trauma center for 47.3 h, including 259 interruptions (9.1% of total time) across 580 nursing activities. A stochastic, non-stationary, model of a nurse's work was developed considering source and activity-dependent interruptions, with parameters clustered across similar periods of day. Two interventions emulating 'do not disturb' strategies were evaluated, along with a more focused intervention from suggestions that nurses' phone calls be 'triaged'. Modeled outcomes included the increase in interruptions in other activities due to deferment and changes to the beneficial/detrimental interruption (B/D) ratio. Across-the-board sequestering of nurses by deferring interruptions during medication increased the B/D ratio 17% (1.35 vs. 1.58), but resulted in an unforeseen 73% (1.04/h vs. 1.80/h) increase in interruptions during direct care. In contrast, the focused intervention (deferring only those interruptions arriving via cell phone during medication and direct care), netted a 31% improvement in the B/D ratio (1.29 vs. 1.69) and with moderated (< 0.13/h) impact on interruptions during other activities. Modeling the dynamics of the onset of interruptions reveals the potentially negative impact of across-the-board interventions, and the advantage of focused interventions anticipating unmet needs before they present as interruptions.

An early warning tool for predicting at admission the discharge disposition of a hospitalized patient.

OBJECTIVES: To develop an early warning discharge disposition prediction tool based on clinical and health services factors for hospitalized patients. Recent study results suggest that early prediction of discharge disposition (ie, whether patients can return home or require placement in a facility) can improve care coordination, expedite care planning, and reduce length of stay. STUDY DESIGN: Retrospective analysis of inpatient data; development of multiple logistic regression model and an easy-to-use score. METHODS: We used retrospective data from all patients who were admitted in 2013 to the general medical service at the Veterans Affairs Boston Healthcare System and discharged alive. A derivation-validation approach was used to build a multiple logistic regression model, which was transformed into a score for potential implementation. RESULTS: Of the 4760 patients discharged in 2013, 485 (10.2%) were discharged to a facility other than home. Correlates of discharge to a facility included a primary admission diagnosis of neoplasm (odds ratio [OR], 2.71; 95% CI, 1.73-4.25), diseases of the nervous system (OR, 2.53; 95% CI, 1.26-5.08), and musculoskeletal diseases (OR, 2.55; 95% CI, 1.52-4.27), as well as discharge to a facility during previous hospitalization. Patients with a prior primary diagnosis of circulatory disorder and those with comorbidity of hypertension, either complicated or uncomplicated, were less likely to be discharged to a facility. A value of 5 or greater on the 20-point scale indicated discharge to a facility with 83% sensitivity and 48% specificity. CONCLUSIONS: A validated, easy-to-use score can assist providers in identifying upon admission those patients who may not be able to go directly home after hospitalization, thus facilitating early discharge planning and coordination, potentially reducing length of hospital stay and improving patient experience.

Evaluating e-consultation implementations based on use and time-line across various specialties.

BACKGROUND: Electronic Consultation (e-consults) can provide improved access, enhance patient and provider satisfaction, and reduce beneficiary travel expenses. We explored how e-consults were implemented across three specialty areas, diabetes (Diab), gastroenterology (GI), and neurosurgery (Neuro), at two Veterans Affairs hospitals in terms of strategies for use and time-lines. METHODS: We conducted observations and electronically shadowed patient e-consultations submitted to a specialty care service by primary care provider(s) at the two sites during a thirteen-month period. We divided the e-consult process in each specialty into three broad milestones; Request (from primary to specialty), Response (from specialty back to primary), and Follow up (from primary to patient), and recorded the flow and time in each category. An overall hierarchy of e-consults was developed to illustrate the many ways an e-consult was used. The Kolmogorov-Smirnov test was used to compare the distribution of time across specialties. RESULTS: A total of 394 consults submitted between April 14, 2012 and May 2, 2013 were reviewed (Diab=152, GI=169, Neuro=73). Of the 152 diabetes specialty clinic e-consults, 35% required some sort of direct contact with the patient by the specialty clinic before a recommendation was provided. Overall, 58% of the e-consults were completed within 20days, while 68% were completed within 30days. The Response times between Diab and GI were significantly different (median=0 vs. 3days; p<0.0001) and so were Follow up times (median=0 vs. 4days; p<0.0001). All three stages were statistically different between Diab and Neuro; however, there was not enough evidence to suggest any differences between GI and Neuro. CONCLUSIONS: The use of an e-consult is likely to vary based on the specialty, but the often significant variations in time may continue to hinder prompt access to care. E-consult design, implementation, documentation, training, self-learning, and monitoring should be tailored to get the most benefit out of this system.

Nursing Interruptions in a Trauma Intensive Care Unit: A Prospective Observational Study.

OBJECTIVE: The aims of this study were to identify and analyze elements that affect duration of an interruption and likelihood of activity switch as experienced by nurses in an ICU. BACKGROUND: Although interruptions in the ICU impact patient safety, little is known regarding the complex situations that drive them. METHODS: RNs were observed in a 23-bed surgical ICU. We observed 206 interruptions, and analyzed for duration and activity switch. RESULTS: RNs were interrupted on the average every 21.8 minutes. Attending physicians/residents caused fewer, but longer, interruptions to the RN. Longer interruptions were more likely to result in an activity switch. During complex situations such as when an RN is documenting, interruptions by a physician led to longer durations. Interruptions by a device led to higher switches. CONCLUSIONS: A deeper understanding of individual factors and their complex interactions related to interruptions experienced by ICU RNs are vital to understanding the clinical significance of these interruptions and intervention design.

The n-by-T Target Discharge Strategy for Inpatient Units.

BACKGROUND: Ineffective inpatient discharge planning often causes discharge delays and upstream boarding. While an optimal discharge strategy that works across all units at a hospital is likely difficult to identify and implement, a strategy that provides a reasonable target to the discharge team appears feasible. METHODS: We used observational and retrospective data from an inpatient trauma unit at a Level 2 trauma center in the Midwest US. Our proposed novel n-by-T strategy-discharge n patients by the Tth hour-was evaluated using a validated simulation model. Outcome measures included 2 measures: time-based (mean discharge completion and upstream boarding times) and capacity-based (increase in annual inpatient and upstream bed hours). Data from the pilot implementation of a 2-by-12 strategy at the unit was obtained and analyzed. RESULTS: The model suggested that the 1-by-T and 2-by-T strategies could advance the mean completion times by over 1.38 and 2.72 h, respectively (for 10 AM ≤ T ≤ noon, occupancy rate = 85%); the corresponding mean boarding time reductions were nearly 11% and 15%. These strategies could increase the availability of annual inpatient and upstream bed hours by at least 2,469 and 500, respectively. At 100% occupancy rate, the hospital-favored 2-by-12 strategy reduced the mean boarding time by 26.1%. A pilot implementation of the 2-by-12 strategy at the unit corroborated with the model findings: a 1.98-h advancement in completion times (P<0.0001) and a 14.5% reduction in boarding times (P = 0.027). CONCLUSION: Target discharge strategies, such as the n-by-T, can help substantially reduce discharge lateness and upstream boarding, especially during high unit occupancy. To sustain implementation, necessary commitment from the unit staff and physicians is vital, and may require some training.

Differentiating between detrimental and beneficial interruptions: a mixed-methods study.

INTRODUCTION: Efforts to understand interruptions now span much of the last decade and a half. Often thought to negatively impact patient safety, some now acknowledge that interruptions may be beneficial and actually necessary for safety and high quality care. This study seeks a framework for differentiating between interruptions that are detrimental and those that are beneficial. METHODS: A mixed-methods approach at a US Level 1 trauma centre included direct observation of 13 registered nurses (RNs), survey of 47 RNs, retrospective observation of hands-free communication devices, and modelling of observed interruptions to key performance measures. RESULTS: On average, RNs were interrupted every 11 min, with 20.3% of their workload triggered by interruptions. While 85% of RNs agreed that interruptions place their patients at risk, only 21% agreed that all should be eliminated. During one 90-min period, 18 original events spawned 68 interruptions, 50 of these repeat messages. A statistical model, with patient measures of time and comfort, revealed that alarms and call lights returning RN's attention to the patient outside the patient room are beneficial, while interruptions in the patient room are generally detrimental. Triangulating the results, we present an emerging framework for differentiating between beneficial and detrimental interruptions based on the impact of interruptions on the RN's steady treatment and attention to the patient. CONCLUSIONS: A mixed-methods approach can help distinguish between detrimental and beneficial interruptions. While interruptions breaking the delivery of steady treatment and attention to the patient are detrimental, those returning the RN's focus to the patient, as well as those supporting patient-clinician and clinician-clinician communications are beneficial. This insight may be helpful to healthcare delivery teams tasked with improving interruption-laden processes.

Impact of Diabetes E-Consults on Outpatient Clinic Workflow.

BACKGROUND: An e-consult is an electronic communication system between clinicians, usually a primary care physician (PCP) and a medical or surgical specialist, regarding general or patient-specific, low complexity questions that would not need an in-person consultation. The objectives of this study were to understand and quantify the impact of the e-consult initiative on outpatient clinic workflow and outcomes. METHODS: We collected data from 5 different Veterans Affairs (VA) outpatient clinics and interviewed several physicians and staff members. We then developed a simulation model for a primary care team at an outpatient clinic. A detailed experimental study was conducted to determine the effects of factors, such as e-consult demand, view-alert notification arrivals, walk-in patient arrivals, and PCP unavailability, on e-consult cycle time. RESULTS: Statistical tests indicated that 4 factors related to outpatient clinic workflow were significant, and levels within each of the 4 significant factors resulted in statistically different e-consult cycle times. The arrival rate of electronic notifications, along with patient walk-ins, had a considerable effect on cycle time. Splitting the workload of an unavailable PCP among the other PCPs, instead of the current practice of allocating it to a single PCP, increases the system's ability to handle a much larger e-consult demand. CONCLUSIONS: The full potential of e-consults can only be realized if the workflow at the outpatient clinics is designed or modified to support this initiative. This study furthers our understanding of how e-consult systems can be analyzed and alternative workflows tested using statistical and simulation modeling to improve care delivery and outcomes.

Scheduling of advanced practice providers at Level 1 trauma centers.

BACKGROUND: Advanced practice providers (APPs) are essential to the provision of trauma care services, particularly in the wake of residency hour restrictions. Demand for these APPs fluctuates with cyclic patient arrivals; however, most trauma teams continue to staff APPs in a linear fashion. Failure to plan for variable arrivals may contribute to excessive patient wait times and emergency department overcrowding. This study used both qualitative and quantitative approaches to evaluate the impact of APP scheduling on patient wait time and to find schedules minimizing delays in reaching the needed care at the right time. METHODS: A retrospective observation of the availability of APPs and the flow of 2,249 trauma patients at a Level 1 trauma center, using both visual overlays and computer modeling, allowed us to evaluate the baseline condition, two what-if schedules, and two model-generated schedules minimizing patient time without any additional APP hours. RESULTS: A visual overlay of APP staffing on 2010 patient arrivals indicated substantial times of mismatch. Trauma managers considered adding an APP during weekday evenings that would have resulted in a 14.8% increase in APP hours and yielded a 27% reduction in patient wait times according to our model. An alternate schedule was developed and implemented in 2012 with a 10.5% increase in APP hours and yielding a 73% reduction in wait times. We also delineated two schedule options with 57% and 78% reductions in wait time and no increase in APP work hours. CONCLUSION: Evaluating alternate shift times and assignments using visual overlays and computer modeling can provide APP staffing solutions with up to 78% reduction in trauma patient wait time without additional APP labor. Knowing that care at the right time is crucial to arriving patients, making sure APP staffing is synchronized with arriving patients is something trauma center managers cannot ignore. LEVEL OF EVIDENCE: Care management study, level IV.

Predictive factors and models for trauma patient disposition.

BACKGROUND: Hospital length of stay for trauma patients can be unnecessarily prolonged due to delays in disposition planning. Demographic characteristics, comorbidities, and other patient variables may help in planning early during hospitalization. MATERIALS AND METHODS: The data of 2836 trauma patients were retrospectively analyzed. Analysis of variance and the chi-square test were used to determine univariate predictors of discharge location (i.e., home, nonhome, and rehabilitation), and multivariable logistic regression was used to determine independent predictors. Clinical decision rules for discharge location were developed for two models: (1) a regular discharge (RD) model to predict discharge location based on demographic and clinical characteristics at the completion of hospital stay and (2) an admission planning discharge (APD) model based on data available shortly after admission. RESULTS: The discharge locations differed on age, sex, certain comorbidities, and various hospital and clinical variables. Increased age, female sex, longer intensive care unit and hospital stays, and the comorbidities of neurologic deficiencies, coagulopathy, and diabetes were independent predictors of nonhome discharge in the RD model. For the APD model, increased age, female sex, the comorbidities of neurologic deficiencies, diabetes, coagulopathy, and obesity were independent predictors of nonhome discharge. The RD and APD models correctly predicted the discharge location 87.2% and 82.9% of the time, respectively. CONCLUSIONS: Demographic and clinical information for trauma patients predicts disposition early in the hospital stay. If the clinical decision rules are validated, discharge steps can be taken earlier in the hospital course, resulting in increased patient satisfaction, timely rehabilitation, and cost savings.

Analyzing discharge strategies during acute care: a discrete-event simulation study.

BACKGROUND: We developed a discrete-event simulation model of patient pathway through an acute care hospital that comprises an ED and several inpatient units. The effects of discharge timing on ED waiting and boarding times, ambulance diversions, leave without treatment, and readmissions were explicitly modeled. We then analyzed the impact of 1 static and 2 proactive discharge strategies on these system outcomes. RESULTS: Our analysis indicated that although the 2 proactive discharge strategies significantly reduced ED waiting and boarding times, and several other measures, compared with the static strategy (P < 0.01), the number of readmissions increased substantially. Further analysis indicated that these findings are sensitive to changes in patient arrival rate and conditions for ambulance diversion. CONCLUSIONS: Determining the appropriate time to discharge patients not only can affect individual patients' health outcomes, but also can affect various aspects of the hospital. The study improves our understanding of how individual inpatient discharge decisions can be objectively viewed in terms of their impact on other operations, such as ED crowding and readmission, in an acute care hospital.

Application of behavior-based ergonomics therapies to improve quality of life and reduce medication usage for Alzheimer's/dementia residents.

Behavior-based ergonomics therapy (BBET) has been proposed in the past as a viable individualized non-pharmacological intervention to manage challenging behaviors and promote engagement among long-term care residents diagnosed with Alzheimer's/dementia. We evaluate the effect of BBET on quality of life and behavioral medication usage in an 18-bed dementia care unit at a not-for-profit continuing care retirement community in West Central Ohio. Comparing a target cohort during the 6-month pre-implementation period with the 6-month post-implementation period, our study indicates that BBET appears to have a positive impact on the resident's quality of life and also appears to correlate with behavioral medical reduction. For instance, the number of days with behavioral episodes decreased by 53%, the total Minimum Data Set (MDS) mood counts decreased by 70%, and the total MDS behavior counts decreased by 65%. From a medication usage standpoint, the number of pro re nata (PRN) Ativan doses decreased by 57%.

Individualized behavior management program for Alzheimer's/dementia residents using behavior-based ergonomic therapies.

Person-centered, nonpharmacological interventions for managing Alzheimer's/dementia-related behavioral disturbances have received significant attention. However, such interventions are quite often of a single type limiting their benefits. We develop a comprehensive nonpharmacological intervention, the Behavior-Based Ergonomic Therapy (BBET), which consists of multiple therapies. This low-cost, 24/7 program uses learning, personality, and behavioral profiles and cognitive function of each resident to develop a set of individualized therapies. These therapies are made available through an accessible resource library of music and video items, games and puzzles, and memory props to provide comfort or stimulation depending on an individual resident's assessment. The quantitative and qualitative benefits of the BBET were evaluated at the dementia care unit in a not-for-profit continuing care retirement community in west central Ohio. The 6-month pilot study reduced falls by 32.5% and markedly reduced agitation through increased resident engagement.

Acute kidney injury (AKI) and risk of readmissions in patients with heart failure.

Heart failure (HF) contributes to a high rate of hospitalizations. Acute kidney injury (AKI), with or without chronic kidney disease (CKD), is a common complication in patients with HF. The link between AKI and the risk for readmission for repeat episodes of HF is not well studied. In this study, 6,535 patients discharged with primary diagnoses of HF derived from a statewide inpatient database were examined. The association between AKI, with and without CKD, and risk for 30-day readmission with HF was assessed. Logistic regression was used to test the relations between predictor variables and outcomes. The mean age was 73.8 ± 14.6 years, and 51% of patients (n = 3,351) were women. AKI occurred in 6.5% of patients during the index hospitalization, whereas 16% had CKD. Nine hundred seventy-seven patients (15%) required readmission within 30 days for HF. Index hospital mortality was 1.7% in those without AKI or CKD compared to 11% and 13% in those with AKI without and with CKD, respectively (p <0.0001). Patients with AKI had a 30-day readmission rate of 21%, compared to 14% in those without AKI (p <0.0001). On multivariate analysis, AKI without CKD was associated with the highest risk for readmission (odds ratio 1.81, 95% confidence interval 1.35 to 2.39) compared to those with neither of the 2 diagnoses. In conclusion, patients with HF who have AKI experience a high rate of 30-day readmission for repeat episodes of HF. Reducing the risk for AKI, and follow-up monitoring after AKI, may improve care and reduce health care costs in patients with HF.