Can emergency medical service staff predict the disposition of patients they are transporting?

The accuracy of predicting hospital admission by emergency medical service and emergency department personnel compared to the prehospital MEWS: a prospective multicenter study

INTRODUCTION

Overcrowding in the emergency department (ED) is a global problem. Early and accurate recognition of a patient's disposition could limit time spend at the ED and thus improve throughput and quality of care provided. This study aims to compare the accuracy among healthcare providers and the prehospital Modified Early Warning Score (MEWS) in predicting the requirement for hospital admission.

METHODS

A prospective, observational, multi-centre study was performed including adult patients brought to the ED by ambulance. Involved Emergency Medical Service (EMS) personnel, ED nurses and physicians were asked to predict the need for hospital admission using a structured questionnaire. Primary endpoint was the comparison between the accuracy of healthcare providers and prehospital MEWS in predicting patients' need for hospital admission.

RESULTS

In total 798 patients were included of whom 393 (49.2%) were admitted to the hospital. Sensitivity of predicting hospital admission varied from 80.0 to 91.9%, with physicians predicting hospital admission significantly more accurately than EMS and ED nurses (p < 0.001). Specificity ranged from 56.4 to 67.0%. All healthcare providers outperformed MEWS ≥ 3 score on predicting hospital admission (sensitivity 80.0-91.9% versus 44.0%; all p < 0.001). Predictions for ward admissions specifically were significantly more accurate than MEWS (specificity 94.7-95.9% versus 60.6%, all p < 0.001).

CONCLUSIONS

Healthcare providers can accurately predict the need for hospital admission, and all providers outperformed the MEWS score.

Transport Rates and Prehospital Intervals for an EMS Telemedicine Intervention

INTRODUCTION

Emergency medical services (EMS) facilitated telemedicine encounters have been proposed as a strategy to reduce transports to hospitals for patients who access the 9-1-1 system. It is unclear which patient impressions are most likely able to be treated in place. It is also unknown if the increased time spent facilitating the telemedicine encounter is offset by the time saved from reducing the need for transport. The objective of this study was to determine the association between the impressions of EMS clinicians of the patients' primary problems and transport avoidance, and to describe the effects of telemedicine encounters on prehospital intervals.

METHODS

This was a retrospective review of EMS records from two commercial EMS agencies in New York and Tennessee. For each EMS call where a telemedicine encounter occurred, a matched pair was identified. Clinicians' impressions were mapped to the corresponding category in the International Classification of Primary Care, 2nd edition (ICPC-2). Incidence and rates of transport avoidance for each category were determined. Prehospital interval was calculated as the difference between the time of ambulance dispatch and back-in-service time.

RESULTS

Of the 463 prehospital telemedicine evaluations performed from March 2021 to April 2022, 312 (67%) avoided transports to the hospital. Respiratory calls were most likely to result in transport avoidance (p = 0.018); no other categories had statistically significant transport rates. Four hundred sixty-one (99.6%) had matched pairs identified and were included in the analysis. When compared to the matched pair, telemedicine without transport was associated with a prehospital interval reduction in 68% of the cases with a median reduction of 16 min; this is significantly higher than telemedicine with transport when compared to the matched pair with a median interval increase in 27 min. Regardless of transport status, the prehospital interval was a median of 4 min shorter for telemedicine encounters than non-telemedicine encounters (p = 0.08).

CONCLUSION

In this study, most telemedicine evaluations resulted in ED transport avoidance, particularly for respiratory issues. Telemedicine interventions were associated with a median four-minute decrease in prehospital interval per call. Future research should investigate the long-term effects of telemedicine on patient outcomes.

The Ability of Emergency Medical Service Staff to Predict Emergency Department Disposition: A Prospective Study

Purpose

Paramedics' decision to notify receiving hospitals and transport patients to an appropriate healthcare facility is based on the Prediction of Intensive Care Unit (ICU) and Hospital Admissions guide. This study aimed to assess the paramedics' gestalt on both ward and ICU admission.

Patients and Methods

A prospective study was conducted at King Abdulaziz Medical City between September 2021 and March 2022. Paramedics were asked several questions related to the prediction of the patient's hospital outcome, including emergency department (ED) discharge or hospital admission (ICU or ward). Additional data, such as the time of the ambulance's arrival and the staff years of experience, were collected. The categorical characteristics are presented by frequency and percentage for each category.

Results

This study included 251 paramedics and 251 patients. The average age of the patients was 62 years. Of the patients, 32 (12.7%) were trauma, and 219 (87.3%) were non-trauma patients. Two-thirds of the patients (n=171, 68.1%) were predicted to be admitted to the hospital, and 80 (31.8%) of the EMS staff indicated that the patient do not need a hospital or an ambulance. The sensitivity, specificity, PPV, and NPV of the emergency medical service (EMS) staffs' gestalt for patient admission to the hospital were, respectively (77%), (33%), (16%), and (90%). Further analysis was reported to defend the EMS staffs' gestalt based on the level of EMS staff and the nature of the emergency (medical vs trauma), are reported.

Conclusion

Our study reports a low level of accurately predicting patient admission to the hospital, including the ICU. The results of this study have important implications for enhancing the accuracy of EMS staff predictive ability and ensuring that patients receive appropriate care promptly.

Inclusion of patient-level emergency department characteristics to classify potentially redirectable visits to subacute care: a modified Delphi consensus study

BACKGROUND

Most patients transported by Ontario paramedics to the emergency department have non-emergent conditions and may be more appropriately served by subacute community-based care centres. We sought to determine consensus on a set of patient characteristics that could be useful to classify retrospective emergency department visits that had a high probability of being primary care-like and potentially redirectable to a subacute care centre by paramedics.

METHODS

We conducted a modified Delphi study to assess expert consensus on characteristics of patients transported by paramedics to the emergency department from August to October 2021. An expert Delphi committee was constructed of emergency and family physicians in Ontario using purposive sampling. Experts rated whether each characteristic was useful to be included in a classification to identify potentially redirectable visits retrospectively, as well as characteristic details (e.g., upper and lower bounds). Consensus was considered 75% agreement.

RESULTS

Sixteen experts participated in the study; the experts were mostly male (75%) and evenly divided between emergency and family medicine. After 2 rounds, consensus was achieved on 8 of 9 characteristics (89%). Four characteristics were determined as useful to classify potentially redirectable emergency department visits: age (81%), triage acuity (100%), specialist consult in the emergency department (94%) and emergency department visit outcome (81%). Specifications of each characteristic were refined as follows: young and middle-aged adults with a non-emergent triage acuity, did not receive a specialist physician consult in the emergency department and discharged from the emergency department.

INTERPRETATION

Strong consensus was achieved to specify a classification system for potentially redirectable emergency department visits. These results will be combined with knowledge of which subacute care centres could conduct the main physician interventions to retrospectively identify emergency department visits that could have been suitable for paramedic redirection for further research.

STUDY REGISTRATION

ID ISRCTN22901977.

Redirecting emergency medical services patients with unmet primary care needs: the perspective of paramedics on feasibility and acceptance of an alternative care path in a qualitative investigation from Berlin, Germany

Background

Against the backdrop of emergency department (ED) overcrowding, patients’ potential redirection to outpatient care structures is a subject of current political debate in Germany. It was suggested in this context that suitable lower-urgency cases could be transported directly to primary care practices by emergency medical services (EMS), thus bypassing the ED. However, practicality is discussed controversially. This qualitative study aimed to capture the perspective of EMS personnel on potential patient redirection concepts.

Methods

We conducted qualitative, semi-structured phone interviews with 24 paramedics. Interviews were concluded after attainment of thematic saturation. Interviews were transcribed verbatim, and qualitative content analysis was performed.

Results

Technical and organizational feasibility of patients’ redirection was predominantly seen as limited (theme: “feasible, but only under certain conditions”) or even impossible (theme: “actually not feasible”), based on a wide spectrum of potential barriers. Prominently voiced reasons were restrictions in personnel resources in both EMS and ambulatory care, as well as concerns for patient safety ascribed to a restricted diagnostic scope. Concerning logistics, alternative transport options were assessed as preferable. Regarding acceptance by stakeholders, the potential for releasing ED caseload was described as a factor potentially promoting adoption, while doubt was raised regarding acceptance by EMS personnel, as their workload was expected to conversely increase. Paramedics predominantly did not consider transporting lower-urgency cases as their responsibility, or even as necessary. Participants were markedly concerned of EMS being misused for taxi services in this context and worried about negative impact for critically ill patients, as to vehicles and personnel being potentially tied up in unnecessary transports. As to acceptance on the patients’ side, interview participants surmised a potential openness to redirection if this would be associated with benefits like shorter wait times and accompanied by proper explanation.

Conclusions

Interviews with EMS staff highlighted considerable doubts about the general possibility of a direct redirection to primary care as to considerable logistic challenges in a situation of strained EMS resources, as well as patient safety concerns. Plans for redirection schemes should consider paramedics’ perspective and ensure a provision of EMS with the resources required to function in a changing care environment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12873-022-00660-2.

Prediction of patient disposition: comparison of computer and human approaches and a proposed synthesis

OBJECTIVE

To compare the accuracy of computer versus physician predictions of hospitalization and to explore the potential synergies of hybrid physician-computer models.

MATERIALS AND METHODS

A single-center prospective observational study in a tertiary pediatric hospital in Boston, Massachusetts, United States. Nine emergency department (ED) attending physicians participated in the study. Physicians predicted the likelihood of admission for patients in the ED whose hospitalization disposition had not yet been decided. In parallel, a random-forest computer model was developed to predict hospitalizations from the ED, based on data available within the first hour of the ED encounter. The model was tested on the same cohort of patients evaluated by the participating physicians.

RESULTS

198 pediatric patients were considered for inclusion. Six patients were excluded due to incomplete or erroneous physician forms. Of the 192 included patients, 54 (28%) were admitted and 138 (72%) were discharged. The positive predictive value for the prediction of admission was 66% for the clinicians, 73% for the computer model, and 86% for a hybrid model combining the two. To predict admission, physicians relied more heavily on the clinical appearance of the patient, while the computer model relied more heavily on technical data-driven features, such as the rate of prior admissions or distance traveled to hospital.

DISCUSSION

Computer-generated predictions of patient disposition were more accurate than clinician-generated predictions. A hybrid prediction model improved accuracy over both individual predictions, highlighting the complementary and synergistic effects of both approaches.

CONCLUSION

The integration of computer and clinician predictions can yield improved predictive performance.

Institution-Specific Machine Learning Models for Prehospital Assessment to Predict Hospital Admission: Prediction Model Development Study

Background

Although multiple prediction models have been developed to predict hospital admission to emergency departments (EDs) to address overcrowding and patient safety, only a few studies have examined prediction models for prehospital use. Development of institution-specific prediction models is feasible in this age of data science, provided that predictor-related information is readily collectable.

Objective

We aimed to develop a hospital admission prediction model based on patient information that is commonly available during ambulance transport before hospitalization.

Methods

Patients transported by ambulance to our ED from April 2018 through March 2019 were enrolled. Candidate predictors were age, sex, chief complaint, vital signs, and patient medical history, all of which were recorded by emergency medical teams during ambulance transport. Patients were divided into two cohorts for derivation (3601/5145, 70.0%) and validation (1544/5145, 30.0%). For statistical models, logistic regression, logistic lasso, random forest, and gradient boosting machine were used. Prediction models were developed in the derivation cohort. Model performance was assessed by area under the receiver operating characteristic curve (AUROC) and association measures in the validation cohort.

Results

Of 5145 patients transported by ambulance, including deaths in the ED and hospital transfers, 2699 (52.5%) required hospital admission. Prediction performance was higher with the addition of predictive factors, attaining the best performance with an AUROC of 0.818 (95% CI 0.792-0.839) with a machine learning model and predictive factors of age, sex, chief complaint, and vital signs. Sensitivity and specificity of this model were 0.744 (95% CI 0.716-0.773) and 0.745 (95% CI 0.709-0.776), respectively.

Conclusions

For patients transferred to EDs, we developed a well-performing hospital admission prediction model based on routinely collected prehospital information including chief complaints.

Use of Emergency Medical Services by Police

OBJECTIVES

Many police officers receive medical training for limited assessments and interventions. In most situations where medical issues arise, however, emergency medical services (EMS) are called for evaluation, treatment, and transport. Given the limited amount of information about such encounters we examined officer calls for EMS help in a single system to better describe these encounters.

METHODS

Requests for medical help from a fire-based EMS system by police in a moderate-sized city in 2014 and 2015 were identified. In this system, fire department resources are requested for initial evaluations of any medical complaint. Data were extracted from fire records including disposition, transportation from scene, type of injury or illness, and vital signs. Data analysis used descriptive statistics.

RESULTS

4,792 calls were made, representing 2.2% of all police-citizen interactions and 4.2% of all EMS calls. A total of 61.2% of calls resulted in transport to hospital. Of those, 5.6% required fire-based advanced life support; the remainder were transported by private basic life support ambulance or non-medical means. Most requests were for trauma (51.4%), followed by medical (24.7%), drug/alcohol use (17.1%), and psychiatric (6.7%). Vital signs tended to be within normal limits including 72.7% of pulses, 65.1% of systolic blood pressures, and 90.5% of respiratory rates.

CONCLUSION

Requests for EMS assistance from police were common. Most calls involved patients with normal vital signs who did not require advanced life support transport. Further research is needed to identify situations where increased officer training and change in protocols could potentially change EMS response models and improve efficiency of the system.

Early prehospital assessment of non-urgent patients and outcomes at the appropriate level of care: A prospective exploratory study

INTRODUCTION

The Ambulance Organization of Sweden provides qualified medical assessment and treatment by ambulance nurses based on patient needs regarding appropriate levels of care. A new model for patients with non-urgent medical conditions has been introduced. The main objective of this study was to examine early prehospital assessment of non-urgent patients, and its impact on the choice of the appropriate level of care.

METHODS

The study design was a 1-year, prospective study, involving an ambulance district in southwestern Sweden with a population of 78,000. Eligible patients were from18years of age, assessed as priority GREEN by Rapid Emergency Triage and Treatment System (RETTS). Ambulance nurses contacted primary care physicians on decisions on whether a patient should be transported to a primary healthcare unit or an A&E. Data was collected from electronic health records from April 2014 to July 2015. A comparison was made with a retrospective control group without consulting a physician concerning the appropriate level of care.

RESULTS

394 patients were included, 184 in the intervention group, and 210 in the control group. There were statistically significant differences in favor of the study group (p<0.001) regarding no transport, or transport and admission to an A&E. The groups did not differ significantly regarding transport to a primary care unit.

CONCLUSION

This prehospital assessment model indicates a decrease in ambulance transports to an A&E and admissions to a hospital ward. Collaboration between ambulance nurses and primary physicians affects the decision for the appropriate level of care for patients with a non-urgent condition.

Paramedic Recognition of Sepsis in the Prehospital Setting: A Prospective Observational Study

Background. Patients with sepsis benefit from early diagnosis and treatment. Accurate paramedic recognition of sepsis is important to initiate care promptly for patients who arrive by Emergency Medical Services. Methods. Prospective observational study of adult patients (age ≥ 16 years) transported by paramedics to the emergency department (ED) of a Canadian tertiary hospital. Paramedic identification of sepsis was assessed using a novel prehospital sepsis screening tool developed by the study team and compared to blind, independent documentation of ED diagnoses by attending emergency physicians (EPs). Specificity, sensitivity, accuracy, positive and negative predictive value, and likelihood ratios were calculated with 95% confidence intervals. Results. Overall, 629 patients were included in the analysis. Sepsis was identified by paramedics in 170 (27.0%) patients and by EPs in 71 (11.3%) patients. Sensitivity of paramedic sepsis identification compared to EP diagnosis was 73.2% (95% CI 61.4–83.0), while specificity was 78.8% (95% CI 75.2–82.2). The accuracy of paramedic identification of sepsis was 78.2% (492/629, 52 true positive, 440 true negative). Positive and negative predictive values were 30.6% (95% CI 23.8–38.1) and 95.9% (95% CI 93.6–97.5), respectively. Conclusion. Using a novel prehospital sepsis screening tool, paramedic recognition of sepsis had greater specificity than sensitivity with reasonable accuracy.

What evidence is there on the effectiveness of different models of delivering urgent care? A rapid review

Background

In 2013 NHS England set out its strategy for the development of an emergency and urgent care system that is more responsive to patients’ needs, improves outcomes and delivers clinically excellent and safe care. Knowledge about the current evidence base on models for provision of safe and effective urgent care, and the gaps in evidence that need to be addressed, can support this process.

Objective

The purpose of the evidence synthesis is to assess the nature and quality of the existing evidence base on delivery of emergency and urgent care services and identify gaps that require further primary research or evidence synthesis.

Data sources

MEDLINE, EMBASE, The Cochrane Library, the Cumulative Index to Nursing and Allied Health Literature (CINAHL) and the Web of Science.

Methods

We have conducted a rapid, framework-based, evidence synthesis approach. Five separate reviews linked to themes in the NHS England review were conducted. One general and five theme-specific database searches were conducted for the years 1995–2014. Relevant systematic reviews and additional primary research papers were included and narrative assessment of evidence quality was conducted for each review.

Results

The review was completed in 6 months. In total, 45 systematic reviews and 102 primary research studies have been included across all five reviews. The key findings for each review are as follows: (1) demand – there is little empirical evidence to explain increases in demand for urgent care; (2) telephone triage – overall, these services provide appropriate and safe decision-making with high patient satisfaction, but the required clinical skill mix and effectiveness in a system is unclear; (3) extended paramedic roles have been implemented in various health settings and appear to be successful at reducing the number of transports to hospital, making safe decisions about the need for transport and delivering acceptable, cost-effective care out of hospital; (4) emergency department (ED) – the evidence on co-location of general practitioner services with EDs indicates that there is potential to improve care. The attempt to summarise the evidence about wider ED operations proved to be too complex and further focused reviews are needed; and (5) there is no empirical evidence to support the design and development of urgent care networks.

Limitations

Although there is a large body of evidence on relevant interventions, much of it is weak, with only very small numbers of randomised controlled trials identified. Evidence is dominated by single-site studies, many of which were uncontrolled.

Conclusions

The evidence gaps of most relevance to the delivery of services are (1) a requirement for more detailed understanding and mapping of the characteristics of demand to inform service planning; (2) assessment of the current state of urgent care network development and evaluation of the effectiveness of different models; and (3) expanding the current evidence base on existing interventions that are viewed as central to delivery of the NHS England plan by assessing the implications of increasing interventions at scale and measuring costs and system impact. It would be prudent to develop a national picture of existing pilot projects or interventions in development to support decisions about research commissioning.

Funding

The National Institute for Health Research Health Services and Delivery Research Programme.

Can Medical Decision-making at the Scene by EMS Staff Reduce the Number of Unnecessary Ambulance Transportations, but Still Be Safe?

OBJECTIVES

The aim of this study was to evaluate the procedures adopted by the staff of the Shiraz Emergency Medical Services (EMS) and the outcome of the patients discharged from the scene over a one-year period.  

BACKGROUND

Unnecessary use of ambulances results in the overloading of EMS and the over-crowding of emergency departments. Medical assessment at the scene by EMS staff may reduce these issues. In an earlier study in Shiraz, 36% of the patients were left at home/discharged directly from the scene with or without treatment by EMS staff after consulting a physician at the dispatch center. However, there has been no evaluation of this system with regard to mortality and morbidity.  

MATERIALS AND METHODS

Retrospective data on all missions performed by the Shiraz EMS (2012-2013) were reviewed. All the patients discharged from the scene by the EMS staff on the 5th, 15th, and 25th days of each month were included. A questionnaire with nine questions was designed, and available patients/relatives were interviewed prospectively (2014; follow-up period 4-12 months).  

RESULTS

Out of 3019 cases contacted, 994 (almost 33%) replied. There were 26%-93% reductions in the complaints in all disease categories. A group of the patients left the scene at their own will. Of those who were discharged by the EMS staff at the scene, over 60% were without any complaints. Twelve out of 253 patients died after they were sent home by the EMS staff.  

CONCLUSIONS

Patients may be discharged at the scene by EMS staff and after consulting a physician. However, there is a need for a solid protocol to ensure total patient safety. This calls for a prospective study.

Identification of a neurologic scale that optimizes EMS detection of older adult traumatic brain injury patients who require transport to a trauma center

OBJECTIVE

We sought to identify a scale or components of a scale that optimize detection of older adult traumatic brain injury (TBI) patients who require transport to a trauma center, regardless of mechanism.

METHODS

We assembled a consensus panel consisting of nine experts in geriatric emergency medicine, prehospital medicine, trauma surgery, geriatric medicine, and TBI, as well as prehospital providers, to evaluate the existing scales used to identify TBI. We reviewed the relevant literature and solicited group feedback to create a list of candidate scales and criteria for evaluation. Using the nominal group technique, scales were evaluated by the expert panel through an iterative process until consensus was achieved.

RESULTS

We identified 15 scales for evaluation. The panel's criteria for rating the scales included ease of administration, prehospital familiarity with scale components, feasibility of use with older adults, time to administer, and strength of evidence for their performance in the prehospital setting. After review and discussion of aggregated ratings, the panel identified the Simplified Motor Scale, GCS-Motor Component, and AVPU (alert, voice, pain, unresponsive) as the strongest scales, but determined that none meet all EMS provider and patient needs due to poor usability and lack of supportive evidence. The panel proposed that a dichotomized decision scheme that includes domains of the top-rated scales -level of alertness (alert vs. not alert) and motor function (obeys commands vs. does not obey) -may be more effective in identifying older adult TBI patients who require transport to a trauma center in the prehospital setting.

CONCLUSIONS

Existing scales to identify TBI are inadequate to detect older adult TBI patients who require transport to a trauma center. A new algorithm, derived from elements of previously established scales, has the potential to guide prehospital providers in improving the triage of older adult TBI patients, but needs further evaluation prior to use.

[Prediction of further hospital treatment for emergency patients by emergency medical service physicians]

INTRODUCTION

Prehospital assessment of illness and injury severity with the National Advisory Committee for Aeronautics (NACA) score and hospital pre-arrival notification of a patient who is likely to need intensive care unit (ICU) or intermediate care unit (IMC) admission are both common in Germany's physician-staffed emergency medical services (EMS) system.

AIM

This study aimed at comparing the prehospital evaluation of severity of disease or injuries by EMS physicians and the subsequent clinical treatment in unselected emergency department (ED) patients.

MATERIAL AND METHODS

This study involved a prospective observational analysis of patients transported to the ED of an academic level I hospital escorted by an EMS physician over a period of 6 months (February-July 2011). The physician's qualification and the patient's NACA score were documented and the EMS physician was asked to predict whether the patient would need hospital admission and, if so, to the general ward, IMC or ICU. After the ED treatment, discharge or admission, outcome and length of hospital and ICU or IMC stay were documented.

RESULTS

A total of 378 mostly non-trauma patients (88 %) treated by experienced EMS physicians could be enrolled. The number of patients discharged from the ED decreased, while the number of patients admitted to the ICU increased with higher NACA scores. Prehospital prediction of discharge or admission, IMC or ICU treatment by EMS physicians was accurate in 47 % of the patients. In 40 % of patients a lower level of care was sufficient while 12 % needed treatment on a higher level of care than that predicted by EMS physicians. Of the patients 39 % who were predicted to be discharged after ED treatment, were admitted to hospital and 48 % of patients predicted to be admitted to the IMC were admitted to the general ward. Patients predicted to be admitted to the ICU were admitted to the ICU in 75 %. Higher NACA scores were associated with increased mortality and a longer hospital IMC or ICU length of stay, but significant differences were only found between patients with NACA V versus VI scores or patients predicted to be treated on the IMC versus the ICU.

CONCLUSIONS

Prehospital NACA scores indicate the need for inpatient treatment, but neither hospital discharge or admission nor need of IMC or ICU admission after initial ED treatment could be sufficiently predicted by EMS physicians. Thus, hospital prenotification in order to predispose IMC or ICU capacities does not seem to be useful in cases where an ED can reassess admitted EMS patients.

Generalizability of a simple approach for predicting hospital admission from an emergency department

OBJECTIVES

The objective was to test the generalizability, across a range of hospital sizes and demographics, of a previously developed method for predicting and aggregating, in real time, the probabilities that emergency department (ED) patients will be admitted to a hospital inpatient unit.

METHODS

Logistic regression models were developed that estimate inpatient admission probabilities of each patient upon entering an ED. The models were based on retrospective development (n = 4,000 to 5,000 ED visits) and validation (n = 1,000 to 2,000 ED visits) data sets from four heterogeneous hospitals. Model performance was evaluated using retrospective test data sets (n = 1,000 to 2,000 ED visits). For one hospital the developed model also was applied prospectively to a test data set (n = 910 ED visits) coded by triage nurses in real time, to compare results to those from the retrospective single investigator-coded test data set.

RESULTS

The prediction models for each hospital performed reasonably well and typically involved just a few simple-to-collect variables, which differed for each hospital. Areas under receiver operating characteristic curves (AUC) ranged from 0.80 to 0.89, R(2) correlation coefficients between predicted and actual daily admissions ranged from 0.58 to 0.90, and Hosmer-Lemeshow goodness-of-fit statistics of model accuracy had p > 0.01 with one exception. Data coded prospectively by triage nurses produced comparable results.

CONCLUSIONS

The accuracy of regression models to predict ED patient admission likelihood was shown to be generalizable across hospitals of different sizes, populations, and administrative structures. Each hospital used a unique combination of predictive factors that may reflect these differences. This approach performed equally well when hospital staff coded patient data in real time versus the research team retrospectively.

Predicting emergency department inpatient admissions to improve same-day patient flow

OBJECTIVES

The objectives were to evaluate three models that use information gathered during triage to predict, in real time, the number of emergency department (ED) patients who subsequently will be admitted to a hospital inpatient unit (IU) and to introduce a new methodology for implementing these predictions in the hospital setting.

METHODS

Three simple methods were compared for predicting hospital admission at ED triage: expert opinion, naïve Bayes conditional probability, and a generalized linear regression model with a logit link function (logit-linear). Two months of data were gathered from the Boston VA Healthcare System's 13-bed ED, which receives approximately 1,100 patients per month. Triage nurses were asked to estimate the likelihood that each of 767 triaged patients from that 2-month period would be admitted after their ED treatment, by placing them into one of six categories ranging from low to high likelihood. Logit-linear regression and naïve Bayes models also were developed using retrospective data and used to estimate admission probabilities for each patient who entered the ED within a 2-month time frame, during triage hours (1,160 patients). Predictors considered included patient age, primary complaint, provider, designation (ED or fast track), arrival mode, and urgency level (emergency severity index assigned at triage).

RESULTS

Of the three methods considered, logit-linear regression performed the best in predicting total bed need, with a receiver operating characteristic (ROC) area under the curve (AUC) of 0.887, an R(2) of 0.58, an average estimation error of 0.19 beds per day, and on average roughly 3.5 hours before peak demand occurred. Significant predictors were patient age, primary complaint, bed type designation, and arrival mode (p < 0.0001 for all factors). The naïve Bayesian model had similar positive predictive value, with an AUC of 0.841 and an R(2) of 0.58, but with average difference in total bed need of approximately 2.08 per day. Triage nurse expert opinion also had some predictive capability, with an R(2) of 0.52 and an average difference in total bed need of 1.87 per day.

CONCLUSIONS

Simple probability models can reasonably predict ED-to-IU patient volumes based on basic data gathered at triage. This predictive information could be used for improved real-time bed management, patient flow, and discharge processes. Both statistical models were reasonably accurate, using only a minimal number of readily available independent variables.

Patient safety in emergency medical services: a systematic review of the literature

BACKGROUND

Preventable harm from medical care has been extensively documented in the inpatient setting. Emergency medical services (EMS) providers care for patients in dynamic and challenging environments; prehospital emergency care is a field that represents an area of high risk for errors and harm, but has received relatively little attention in the patient safety literature.

OBJECTIVE

To identify the threats to patient safety unique to the EMS environment and interventions that mitigate those threats, we completed a systematic review of the literature.

METHODS

We searched MEDLINE, EMBASE, and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) for combinations of key EMS and patient safety terms composed by a pan-canadian expert panel using a year limit of 1999 to 2011. We excluded commentaries, opinions, letters, abstracts, and non-english publications. Two investigators performed an independent hierarchical screening of titles, abstracts, and full-text articles blinded to source. We used the kappa statistic to examine interrater agreement. Any differences were resolved by consensus.

RESULTS

We retrieved 5,959 titles, and 88 publications met the inclusion criteria and were categorized into seven themes: adverse events and medication errors (22 articles), clinical judgment (13), communication (6), ground vehicle safety (9), aircraft safety (6), interfacility transport (16), and intubation (16). Two articles were randomized controlled trials; the remainder were systematic reviews, prospective observational studies, retrospective database/chart reviews, qualitative interviews, or surveys. The kappa statistics for titles, abstracts, and full-text articles were 0.65, 0.79, and 0.87, respectively, for the first search and 0.60, 0.74, and 0.85 for the second.

CONCLUSIONS

We found a paucity of scientific literature exploring patient safety in EMS. Research is needed to improve our understanding of problem magnitude and threats to patient safety and to guide interventions.