 |
|
Study objectiveWe determine the effect of crowding on emergency department (ED) waiting room, treatment, and boarding times across multiple sites and acuity groups. MethodsThis was a retrospective cohort study that included ED visit and inpatient medicine occupancy data for a 1-year period at 4 EDs. We measured crowding at 30-minute intervals throughout each patient's ED stay. We estimated the effect of crowding on waiting room time, treatment time, and boarding time separately, using discrete-time survival analysis with time-dependent crowding measures (ie, number waiting, number being treated, number boarding, and inpatient medicine occupancy rate), controlling for patient demographic and clinical characteristics. ResultsCrowding substantially delayed patients' waiting room and boarding times but not treatment time. During the day shift, when the number boarding increased from the 50th to the 90th percentile, the adjusted median waiting room time (range 26 to 70 minutes) increased by 6% to 78% (range 33 to 82 minutes), and the adjusted median boarding time (range 250 to 626 minutes) increased by 15% to 47% (range 288 to 921 minutes), depending on the site. Crowding delayed the care of high-acuity level 2 patients at all sites. During crowded periods (ie, 90%), the adjusted median waiting room times of high-acuity level 2 patients were 3% to 35% higher than during normal periods, depending on the site and crowding measure. ConclusionUsing discrete-time survival analysis, we were able to dynamically measure crowding throughout each patient's ED visit and demonstrate its deleterious effect on the timeliness of emergency care, even for high-acuity patients. SEE EDITORIAL, P. 511. Introduction BackgroundLength of stay is an important measure of quality of care in the emergency department (ED).1, 2, 3 Several studies have found that crowding is associated with increased ED length of stay.4, 5 Rathlev et al4 observed that daily ED length of stay was positively associated with the hospital occupancy rate and number of emergency admissions at their institution. Asaro et al5 found that at their ED, crowding factors increased wait time and boarding time but not treatment time. In contrast, Schull et al6 investigated the effect of the number of low-complexity patients on the length of stay of higher-complexity patients treated at EDs in Ontario, Canada, and found that increasing numbers of low-complexity patients did not significantly lengthen the wait time or ED length of stay of higher-complexity patients. Editor's Capsule SummaryWhat is already known on this topic Emergency department (ED) crowding often results from limited hospital functional capacity and can affect patient care and outcomes. What question this study addressed How does ED crowding alter the time to initiate care, ED treatment time, and ED boarding time after admission, and are these effects similar across patient illness acuities? What this study adds to our knowledge In this 4-ED retrospective analysis of more than 225,000 visits, increased ED crowding delayed the start of care and lengthened boarding time without altering the length of the treatment phase of ED care. This pattern was observed in both high- and low-acuity patients. How this might change clinical practice These data confirm that ED crowding spares few and that sicker patients are affected, and it confirms the need for creative solutions. ImportanceMany hospital-based EDs across the country have been struggling with crowding for more than a decade.7, 8 Until recently, there were few objective measures of crowding and modest evidence of the negative effects of crowding on patient care and outcomes. However, advances in crowding measures during the past 5 years have resulted in a growing number of studies that quantify the negative consequences of crowding, especially on delays in care for time-sensitive conditions.7, 9, 10, 11, 12, 13 One of the major limitations of crowding studies conducted to date is that they measure crowding in a static way, either at a point in time (eg, patient arrival) or by averaging crowding during a specific interval (eg, a shift). However, many studies have shown that crowding is dynamic and can fluctuate substantially during the course of a patient's ED stay.14, 15, 16, 17 Goals of This InvestigationThe purpose of this study was to quantify the relationship between crowding and ED length of stay at 4 hospital EDs and to compare the effect across EDs and patient acuity levels. We measured crowding at regular intervals throughout each patient's ED stay and estimated the cumulative effect of crowding on ED waiting room time, treatment time, and boarding time, using discrete-time survival analysis, which allowed us to include time-varying crowding covariates. We stratified the analysis by shift, site, and patient acuity and determined whether crowding, adjusted for other factors, significantly delayed completion of each phase of ED care. Materials and MethodsStudy DesignThis investigation was based on a retrospective cohort design that included all ED visits to one of 4 hospital EDs during a 1-year period. Patient visit information was abstracted from the information system of each ED and combined with inpatient medicine occupancy and ED staffing data. The institutional review board of each site approved the study by expedited review. SettingThe selection of the 4 study sites was based on previous collaboration and feasibility of aggregating electronic medical record data across the different sites.14, 15 The 4 study sites are geographically dispersed; all but 1 is located in an inner city with a population of more than 500,000 residents. All of the study EDs are part of tertiary care, academic medical centers with Level I trauma centers. Three of the 4 study sites have separate pediatric EDs whose visit data were excluded. However, if a pediatric patient was treated in the study ED, the visit was retained. Table 1 displays the 4 sites by facility, staffing, and crowding characteristics. The 4 study sites differ in terms of inpatient medicine bed capacity (range 224 to 461 beds) and ED capacity (range 26 to 41 beds in main ED). The annual ED volumes across the study sites range between 50,000 and 62,000. Two of the study EDs have an observation unit and 1 has a dedicated laboratory. At 2 study sites, inpatient physicians are primarily responsible for boarders, whereas at the other 2 facilities, emergency physicians remain responsible (ED nurses continue to provide care to boarders at all sites).
The 4 sites also differ substantially by staffing and crowding factors (Table 1). The difference in the median number of attending physicians or residents working per hour is 1 across the sites. However, the difference in the median number of nurses working per hour is 11 between the ED with the most versus fewest nurses. The median number boarding during a 30-minute period at one of the study EDs is twice the median number (median=10) of another ED (number=5). There is substantial variation both within and across the 4 study sites in the hourly ED occupancy rate. The variation in the inpatient medicine occupancy rate within and across the study sites is smaller but still considerable. All 235,928 ED visits that occurred at one of the 4 study sites during a 1-year period (October 1, 2006, to September 30, 2007, for 3 sites and October 1, 2005, to September 30, 2006, for 1 site) were eligible for the study. We eliminated 4% of ED visits across all 4 sites because of registration errors (N=5,986), because patients were sent elsewhere and not treated in the ED (ie, mainly labor and delivery) (N=2,537), or because patients had multiple missing (N=226) or prolonged ED times (ie, wait time >12 hours, treatment and boarding time >48 hours) (N=645). We were left with 226,534 ED visits. Methods of MeasurementThe premise of this study was that when crowding occurred, demand exceeded capacity and delays were more likely. During the course of a patient's ED stay, crowding levels fluctuate, often substantially. Thus, an analysis of crowding must take into account the time-varying nature of crowding. To capture crowding changes during each patient's length of stay, we used a “counting process approach.”18 We divided each patient's length of stay into a sequence of contiguous time periods and observed when (ie, which interval) each patient completed a particular phase of ED care. We measured 3 phases of ED care: (1) waiting room time, defined as time from registration to room placement; (2) treatment time, defined as time from room placement to disposition decision; and (3) boarding time, defined as time from disposition decision to ED-hospital transfer for patients admitted. Completion of waiting room time was measured in 15-minute intervals (eg, noon to 12:14 pm, 12:15 to 12:29 pm), whereas completion of treatment and boarding time was measured in 30-minute intervals (eg, noon to 12:29 pm, 12:30 to 12:59 pm). We measured multiple crowding factors corresponding to the 3 phases of emergency care. We divided the 1-year study period into 30-minute intervals and measured crowding every 30 minutes (starting on the hour and the half-hour) throughout the study period. Crowding measures capture the number of patients within each phase of care at any time during each 30-minute discrete interval. For example, a patient being treated during an entire 30-minute interval is counted as 1 treatment patient. A patient who is treated for 10 minutes in the interval and then boards for the remaining 20 minutes is counted as one-third treatment patient and two-thirds (ie, 20 minutes/30 minutes) boarding patient, respectively. Thus, the sum of patients during each 30-minute interval was not limited to integer values, but may be counted as a fraction. To characterize hospital crowding, we also measured the inpatient medicine occupancy rate. Because the majority of ED patients are admitted to medicine wards, we considered the inpatient medicine units to have a greater effect on ED length of stay than the overall hospital census. For 3 of the sites, the inpatient medicine occupancy rates were available hourly throughout the study period, so we used the rate closest to the previous hour (eg, noon) as the crowding value for the intervals that started on the half hour (eg, 12:30 pm to 12:59 pm). For one site, the inpatient medicine occupancy rate was measured every 6 hours, so we interpolated it linearly between each measurement period.19 To validate the linear interpolation approach we used at this one site, we linearly interpolated the hourly measures at the 3 other sites and found a strong correlation between the interpolated and observed values (r2=0.77, 0.94, and 0.97). The bottom half of Figure 1 depicts how we characterized completion of the different phases of ED care for 6 patients randomly selected from a single day at one of the sites. The first patient on the far left waited 8 minutes before entering the treatment phase, so the first circle is blackened to illustrate completion of waiting room time in the first 15-minute interval. The first patient completed treatment in 323 minutes, so treatment is depicted as completed by the 11th triangle, which is blackened. Finally, the first patient boarded for 200 minutes, and this is depicted by the seventh square, which is blackened.
The top half of Figure 1 shows that we also measured crowding variables throughout each patient's ED stay. Because completion of waiting room time was measured every 15 minutes and the crowding variables were measured every 30 minutes, we used the crowding values from the 30-minute interval that contained the 2 waiting room intervals (eg, the number of patients being treated in the interval from midnight to 12:29 am was used to reflect crowding in the midnight to 12:14 am interval, as well as the 12:15 am to 12:29 am interval). For each phase of ED care, we characterized the distribution of the probability that a particular phase of care would be completed in each interval, given that it had not been completed in previous intervals. To estimate the median time to completion of care, we calculated the probability of completing care during each successive interval, using a survivorship function. Because the survivorship function is a step function, we linearly interpolated the median time to completion of each phase of care.20 At each interval, we also estimated the effect of crowding on the probability of completion of a particular phase of ED care. In this calculation, the effect of crowding on completion of care accumulates, producing the cumulative effect of crowding on each length of stay outcome. To examine the influence of crowding across intervals with similar ED and hospital resources, we stratified the analysis by shift (8 am to 4 pm, 4 pm to midnight, and midnight to 8 am). We obtained the physician, resident, and nurse staffing schedules from each site and initially included staffing in our analyses. However, this was not necessary after stratifying by shift because staffing did not vary significantly within a shift at each site. We excluded the weekend intervals (ie, from Friday night midnight until midnight on Sunday) from the main analysis because resources are different on the weekends and crowding is less common at the study sites. However, Table E1 (available online at http://www.annemergmed.com) includes the results of the main analysis according to weekend ED visits.
Data Collection and ProcessingThis study relied primarily on ED visit and hospital census data. For each ED visit, the following data elements were extracted from each site's ED information system: (1) date and time of registration; (2) date and time of room placement; (3) date and time of initial contact with physician or midlevel provider, if available; (4) date and time of disposition decision; (5) date and time of disposition or transfer to inpatient ward if admitted; (6) disposition status; (7) demographic characteristics (age, sex, insurance status); (8) triage level; (9) mode of arrival; and (10) chief complaint. All of the sites use their ED information system for operational monitoring and improvement purposes and have processes in place to ensure data accuracy. The ED information system at each site is different (2 have different commercial products and 2 developed their own system). Despite the different systems, the majority of data were documented and stored in a similar way, which facilitated analysis across the 4 sites. For example, at each site, arrival time is documented by either a triage nurse or dedicated registration clerk immediately after the patient presents to the ED. The time of transition from waiting to a treatment bed is documented by bed assignment in each ED information system by a nurse or technician. If bed assignment was missing, we used time to initial contact with emergency physician or midlevel provider. At one site, we used the disposition time (completed for both discharged and admitted patients by a physician) to determine the time the physician decided to admit the patient, whereas at the other 3 sites, we used an admission decision order data field completed by physicians for admitted patients. All 4 sites use the Emergency Severity Index to triage patients. The Emergency Severity Index is a 5-level triage scale that prioritizes patients according to their severity of illness and anticipated number of resources needed. The index differentiates patients in terms of urgency: high-acuity patients who should be treated first (Emergency Severity Index levels 1 and 2) from those less urgent (Emergency Severity Index levels 3 to 5). Among those less urgent, Emergency Severity Index levels 3, 4, and 5 are distinguished according to predicted resources needed to make a disposition.21, 22, 23 Each site's ED information system has a standard list of chief complaints, with the option of free text. To standardize the chief complaints across the 4 sites, we classified each chief complaint according to the Reason for Visit Classification System used by the National Hospital Ambulatory Medical Care Survey.24 From the above data, we calculated each patient's waiting room time, treatment time, and boarding time. Across the 4 sites, 7% of visits were missing a data element needed to calculate either the waiting room time or boarding time. For these visits, we imputed waiting room time and boarding time according to acuity, chief complaint, and other timing information that was available. Primary Data AnalysisAll analyses were conducted in R, version 2.7 (available at http://www.R-project.org). The following analyses were conducted separately at each of the EDs. For readers who are interested in our analytical methods, please see Figure E1 (available online at http://www.annemergmed.com), which includes the commands we used and the types of output generated.
First, we examined the frequency distribution of ED wait time, treatment time, and boarding time. Because all 3 outcomes were positively skewed, we focused the analysis on the median rather than the mean time. Second, we conducted a bivariate analysis and examined the relationship between each covariate and the median time in each of the 3 phases of ED care. The following categories of covariates were examined: (1) crowding measures (number of patients waiting, number of patients being treated, number of boarders, and inpatient medicine occupancy rate); (2) patient demographics (age, sex, insurance status); and (3) clinical characteristics (chief complaint, mode of arrival, and acuity level). For the time-dependent variables (ie, crowding measures), we examined the proportion of subjects who completed their care by various levels of crowding. For the time-invariant variables, we compared the median time to completion for different categories for each variable. Third, for each site and shift we used discrete-time survival analysis to estimate the effect of crowding separately on the probability of completing waiting room time, treatment time, and boarding time, adjusted for demographic factors, clinical characteristics, and variation in follow-up time.25 For each phase of ED care, we included only crowding factors that would affect that phase of care or a future phase of ED care. For example, in the treatment time model, we included the number of patients being treated, the number of boarders, and the inpatient medicine occupancy rate, but we did not include the number of patients in the waiting room because subjects in the treatment phase should not be affected by patients who are “behind them” in the ED process. Similarly, in the boarding time model, we included the number of boarders and the inpatient medicine occupancy rate, but we excluded the number of patients in the waiting room and the number of patients being treated because those patients are not yet competing with the boarders for an inpatient bed. To allow for the probability of completion of a phase of care changing over time, we included a natural cubic spline of time with 2 df for the wait time models, 3 df for the treatment time models, and 4 df for the boarding time models. The number of degrees of freedom selected were based on the log likelihood ratio test.26 The discrete-time survival analysis models estimate the cumulative probability of completing a particular phase of care, and we used these cumulative probability distributions to linearly interpolate the median predicted waiting room time, treatment time, and boarding time for the study sample. Fourth, to evaluate the effect of each crowding variable on length of stay, we calculated the median completion times of care from the fitted models under 2 different scenarios: (1) when all crowding measures were at the 50th percentile; and (2) when each crowding measure was at the 90th percentile, holding the other crowding measures at their median values and the other covariates constant. We did not estimate the effect of all crowding factors at the 90th percentile simultaneously because this rarely occurred during the 1-year study period. To illustrate the overall effect of crowding, we calculated the average median waiting room, treatment, and boarding times across the 4 sites by different percentiles of each crowding factor, holding the other crowding factors at 50%. Sensitivity AnalysesTo examine the validity of our imputation methods, we compared the visits with missing data to those with no missing data at each site by patient, clinical, or temporal factors. Because there were no large differences between the 2 groups, we reran our models without the imputed data and compared the results to the results we obtained with the models that included the imputed data. Because there were no substantial differences, the results shown include the imputed data (see Table E2 [available online at http://www.annemergmed.com] for main results without imputed data).
ResultsThe patient populations of the 4 study sites differ substantially from one another, particularly in terms of insurance status, admission rate, and acuity level (Table 2). Site A treats more patients who are uninsured (37%) compared with the other sites (range 5% to 21%). Site D admits proportionately fewer patients (19%) than the other EDs (range 23% to 30%). Site D also treats fewer highly acute patients (ie, Emergency Severity Index levels 1 and 2) (13%) compared with the other sites (range 25% to 41%). Despite the differences in demographic characteristics and acuity level, the unadjusted median ED length of stay for patients across the 4 sites was similar (Table 1).
Table 3 displays the sites' unadjusted median waiting room times, treatment times, and boarding times by patient and clinical characteristics for those visits that occurred during the week. At all 4 sites, at least 50% of ED patients admitted wait at least 3 hours before transfer to an inpatient bed. The clinical characteristics are associated with larger differences in ED length of stay outcomes compared with the patient characteristics. For example, the median waiting room time for patients who walk in is more than double the waiting room time of patients who arrive by ambulance. ED length of stay is longest for acuity level 2 and 3 patients.
Adjusting for differences in patient and clinical characteristics, crowding factors consistently delayed waiting room time and boarding times at all of the sites; however, the magnitude of the effect varied by site. The results based on the weekday data presented in Table 4 are similar to results obtained on the weekend data included in Table E1 (available online at http://www.annemergmed.com).
The crowding factor associated with the longest waiting room times was the number of patients in the waiting room (Table 4). For example, during the day shift, an increase in the number of waiting room patients from 50% to 90% was associated with an increase in waiting room times of 44% to 158%, depending on the site. The number of patients being treated and the number of boarders also significantly increased patients' waiting room time. Crowding affected the waiting room time of the EDs, with the shortest median waiting room times more than that of the EDs with longer median waiting room times. For example, at site B during the day shift, the adjusted median waiting room time increased from 26 minutes to 54 minutes (109% increase) when the number of patients in the waiting room was at 90% (11 patients) compared with 50% (4 patients). In the boarding time models, the number of boarders and the inpatient medicine occupancy rate both had independent, negative effects on boarding time. During the day shift, an increase in the number of boarders was associated with an increase in boarding time of 15% to 47%, depending on the site. Crowding had the greatest effect on the boarding times of patients at site C during the overnight shift (ie, midnight to 8 am). The median boarding time increased from 347 minutes to 967 minutes (179% increase) when the number of boarders increased from the median (12 boarders) to 90% (20 boarders) at that site. In general, crowding factors also had a negative effect on ED treatment times across the sites, but the increases were relatively small. For example, during the day shift, the biggest increase (8%) in median treatment time (20 minutes) was at site A when the number of patients being treated was high (ie, 37 patients) compared with normal (ie, 29 patients). Figure 2 displays the strong effect of the different crowding factors on ED length of stay averaged over all 4 sites during the day shift. As observed with the site-specific data, the delay in care is most marked for waiting room time and boarding time. The impact of the number of patients boarding and the inpatient medicine occupancy rate is similar for the waiting room time and boarding time.
Crowding affected patients differently, depending on their acuity level (Table 5). The highest acuity patients, Emergency Severity Index level 1 patients, were not affected by crowding. Crowding had a strong negative effect on high-acuity level 2 patients and level 3 patients. For example, the adjusted median waiting room time for high-acuity level 2 patients increased by 13% to 29%, depending on the site, when the number of patients in the waiting room increased from 50% to 90%. For level 3 patients, the same increase in the number of waiting room patients was associated with a 24% to 68% increase in adjusted median waiting room time. The increase in the adjusted median boarding time of high-acuity level 2 patients (range 12% to 83%) and level 3 patients (range 12% to 72%) was similar when the number of boarders increased from 50% to 90% at the different sites (Table 5).
LimitationsThe results of this study must be interpreted in the context of the following limitations. First, this study did not attempt to connect the different phases of care and determine how the completion of care in one phase affects the completion of care in another. Second, our models did not include all potential covariates that affect ED length of stay, such as orders for diagnostic tests, treatments, or specialty consultations. Although overall treatment times were not significantly delayed by crowding factors, future research should address how crowding affects different diagnostic and treatment services. Third, the definitions of different ED phases of care and crowding measures were based on somewhat different operational data, depending on the site. Although this is not ideal, the benefits of multicenter participation outweighed the weaknesses of variation in operational data and the results were consistent across the sites. Fourth, the inpatient medicine occupancy rate was collected every 6 hours at one site rather than hourly at the other sites. However, we developed site-specific models, we linearly interpolated the hourly measures at one site, and the results of inpatient medicine occupancy on ED length of stay were consistent across the sites. Using the inpatient medicine occupancy rate rather than the overall hospital occupancy rate may limit the generalizability of the study. However, as previously found by Rathlev et al,4 hospital occupancy can also be used to measure crowding; the magnitude of the effect may be smaller or larger than the result we found, depending on the facility and patient population. Fifth, we used the same calendar period for the analysis at all 4 sites, but one site's analysis is based on data provided to the data coordinating center 1 year earlier than the other 3 sites. We do not believe the results would have been any different at this site if the data used were from the same year as the other 3 sites. Sixth, to better understand how hospital operations affect ED length of stay, it would have been advantageous to have had information on non-ED patients who are competing with ED patients for hospital resources (eg, elective admissions, diagnostic test requests, specialty consultations). Seventh, a small number of pediatric patients treated at the four EDs were left in the analysis, but, given the small proportion, we do not believe it substantially affected the results. Eighth, we did not conduct any site-specific evaluations of the accuracy of the study data. However, the sites have been using all of the data we relied on in this study for operational reporting purposes, so the general accuracy of the study data should be relatively high. Finally, the study sites were all EDs that are part of tertiary care hospitals with Level I trauma centers; they are not a representative sample of hospital EDs in the United States. Regardless, we expect that crowding will negatively affect ED length of stay at other types of EDs if they experience frequent periods of crowding, as did the EDs in this study. DiscussionTo our knowledge, this is the first study to measure crowding dynamically and to examine its effect on ED length of stay. We found that crowding was associated with substantial delays in ED length of stay across 4 ED sites. Moreover, crowding prolonged the ED length of stay of high-acuity level 2 patients. Output factors, such as the number of patients boarding and the inpatient medicine occupancy rate, were associated with large delays in ED care. Crowding negatively affected ED patients' waiting room time and boarding time but not treatment time. Some sites were more sensitive to crowding than others. The results of our study suggest that crowding has a negative effect on the timeliness of the delivery of emergency care; however, the magnitude of the effect varied by the crowding measure, the phase of ED care, patient characteristics, and site. During patients' ED stay, the factor most consistently associated with delays in ED care was the number of boarders in the ED. Not only did the number of boarders significantly increase ED patients' boarding time but it also substantially delayed their waiting room time. After adjusting for the number of boarders in the ED, the inpatient medicine occupancy rate was also associated with significant delays in boarding time across the 4 sites. There is a widespread misconception among the public, policymakers, and the lay press that crowding is primarily caused by large influxes of patient arrivals to the ED, particularly among the uninsured or the poor27; however, the results of this study and others clearly show that output factors are associated with the biggest delays in emergency care.4, 5, 28, 29 Among the ED patients admitted during weekdays, 58% experienced a boarding time that was longer than their treatment time (range across sites 45% to 78%). Patients boarding in the ED for prolonged periods negatively affect patient flow in the ED. Each boarder is equivalent to the loss of 1 ED treatment space and, more important, the ED staff time associated with patient care, family communication, and documentation requirements for an inpatient. In this study, during normal periods, the median number of boarders in the study EDs ranged from 5 to 13, which represents a reduction in main ED bed capacity of 16% to 33%, depending on the study site. When the bed capacity of an ED is routinely reduced by boarders, it signifies a critical problem elsewhere in the system. In this case, a boarding problem most likely reflects a resource constraint (ie, inadequate inpatient bed capacity) or a policy constraint (eg, bed management policies, financial incentives to prioritize elective admissions over emergency ones). The results of this study suggest that we must take a system-wide approach to solve crowding; we cannot limit our analyses and changes to the ED. Proportionately, crowding affected ED waiting room time the most across the 4 sites. ED waiting room time is the shortest of the 3 phases of emergency care, so one could misconstrue that decreasing waiting room time is the least important. However, the ED is the front door of the hospital to almost half of all patients admitted (42%).17 It is during the waiting room time that ED patients and their families form their first impressions of the facility. A number of studies have found that prolonged wait time is associated with patient dissatisfaction and an unwillingness to return to the same facility or to recommend it to others.30, 31, 32, 33 Thus, it is critical that EDs keep waiting room times to a minimum (experts recommend no more than 15 minutes without an explanation for the delay) and develop strategies (eg, frequent communication about delays, providing information on how the ED works, offering comforts in the waiting area) to make patients' waiting room time as short and pleasant as possible.34, 35, 36 Crowding had little effect on ED treatment time across all of the study sites, regardless of the shift. These results suggest that ED providers base their clinical decisions on patient and clinical (ie, results of tests) characteristics rather than system factors. Asaro et al5 also found at their institution that treatment time was strongly influenced by patient demographic and clinical characteristics and not crowding factors. Similarly, Forster et al37 reported that ED admission rates did not vary significantly by hospital occupancy level. These findings imply that any substantial reductions in the ED treatment phase will be related to improvement initiatives or technological advances in the diagnostic, treatment, or discharge process, rather than better management of patient flow.38, 39 Triage did not adequately protect severely ill patients from delays in care during periods of crowding. Across all of the sites, the wait time and boarding time of high-acuity level 2 patients increased substantially during crowded periods. These results are consistent with those of other studies that have found that crowding causes delays in emergency care for patients with time-sensitive conditions.11, 12, 13 It may be that EDs need to consider alternative prioritization of patients such as the See-and-Treat approach being adopted in the United Kingdom that divides patients into those likely to be discharged versus admitted40, 41, 42 or a manufacturing strategy that categorize patients by their expected treatment rather than reported symptoms or acuity.43 There was substantial variation in how crowding affected the study EDs. A sample size of 4 facilities is too small to draw any conclusions about how structural, procedural, or cultural characteristics may contribute to crowding, but the following qualitative observations are worth noting. First, the waiting room time during normal periods at 2 of the EDs was approximately equivalent to the waiting room time during crowded periods at the other study sites. These long waiting room times during normal periods most likely reflect inefficient operations or inadequate capacity. Second, the 2 EDs with the shortest waiting room times during normal periods were disproportionately affected by crowding. At these 2 EDs, the number of boarders was associated with bigger increases in boarding time (23% to 179% across the shifts) compared with that of the other EDs (5% to 18% across the shifts). Third, the ED with the worst boarding problem had the most treatment spaces in the main ED but lacked an observation unit. Finally, the 2 sites where crowding affected the boarding times the most relied on inpatient physicians to be responsible for ED boarders. These results suggest that individual organizations face different problems, so each facility should conduct a local assessment to determine the best solutions to its crowding problem. In addition, if EDs in the United States collected a uniform minimum data set with standard data definitions, we could more easily make across-facility comparisons and identify best practices. In summary, we have used a discrete-time survival analysis approach to demonstrate the negative effect that crowding has on the timely delivery of emergency care, even among high-acuity patients. As the evidence of the deleterious effects of crowding on patient care and outcomes continues to accumulate, the federal government may need to step in as it did in the United Kingdom and set a standard of care that requires 98% of all ED patients be treated and discharged, admitted, or transferred within 4 hours of presentation. It is our belief that without major changes in our health care delivery system, such as those undertaken in the United Kingdom, EDs in the United States will continue to struggle with crowding and its negative effect on patient care.40 The authors would like to thank Allison Orlina, JD, for her help obtaining the inpatient medicine occupancy rate data for one of the study sites. References1. 1. Measuring crowding: time for a paradigm shift. Acad Emerg Med. 2006;13:459–461. CrossRef 2. 2 Optimizing Patient Flow: Moving Patients Smoothly Through Acute Care Settings. Cambridge, MA: Institute for Healthcare Improvement; 2003;IHI Innovation Series White Paper. 3. 3. The Joint Commission requirements. http://www.jcrinc.com/Joint-Commission-RequirementsAccessed January 2, 2009. 4. 4 Time series analysis of variables associated with daily mean emergency department length of stay. Ann Emerg Med. 2007;49:265–271. Abstract | Full Text | Full-Text PDF (275 KB) | CrossRef 5. 5. The impact of input and output factors on emergency department throughput. Acad Emerg Med. 2007;14:235–242. CrossRef 6. 6. The effect of low-complexity patients on emergency department waiting times. Ann Emerg Med. 2007;49:257–264. Abstract | Full Text | Full-Text PDF (366 KB) | CrossRef 7. 7. Hospital-Based Emergency Care: At the Breaking Point. Washington, DC: National Academies Press; 2006;. 8. 8. Hospital Emergency Departments: Crowded Conditions Vary Among Hospitals and Communities. Washington, DC: United States General Accounting Office; 2003;GAO-03-460 [March 2003]. 9. 9 The effect of emergency department crowding on clinically oriented outcomes. Acad Emerg Med. 2008;15:1–10. CrossRef 10. 10. Systematic review of emergency department crowding: causes, effects and solutions. Ann Emerg Med. 2008;52:126–136. Abstract | Full Text | Full-Text PDF (209 KB) | CrossRef 11. 11. Emergency department crowding is associated with poor care for patients with severe pain. Ann Emerg Med. 2008;51:1–5. Abstract | Full Text | Full-Text PDF (87 KB) | CrossRef 12. 12 The impact of emergency department crowding measures on time to antibiotics for patients with community-acquired pneumonia. Ann Emerg Med. 2007;50:510–516. Abstract | Full Text | Full-Text PDF (109 KB) | CrossRef 13. 13 Effect of emergency department crowding on time to antibiotics in patients admitted with community-acquired pneumonia. Ann Emerg Med. 2007;50:501–509. Abstract | Full Text | Full-Text PDF (176 KB) | CrossRef 14. 14 The challenge of predicting demand for emergency department services. Acad Emerg Med. 2008;15:337–346. CrossRef 15. 15 The emergency department occupancy rate: a simple measure of emergency department crowding?. Ann Emerg Med. 2007;51:15–24. Abstract | Full Text | Full-Text PDF (460 KB) 16. 16. Advanced statistics: developing a formal model of emergency department census and defining operational efficiency. Acad Emerg Med. 2007;14:799–809. CrossRef 17. 17. National Hospital Discharge Survey 2006. Hyattsville, MD: National Center for Health Statistics; 2008;. 18. 18. Survival Analysis: Techniques for Censored and Truncated Data. 2nd ed.. New York, NY: Springer-Verlag; 2009;. 19. 19. Attrition in longitudinal studies: how to deal with missing data. J Clin Epidemiol. 2002;55:329–337. Abstract | Full Text | Full-Text PDF (97 KB) | CrossRef 20. 20. Applied Longitudinal Data Analysis: Modeling Change and Event Occurrence. New York, NY: Oxford University Press; 2003;. 21. 21 The Emergency Severity Index triage algorithm version 2 is reliable and valid. Acad Emerg Med. 2003;10:1070–1080. MEDLINE | CrossRef 22. 22 Reliability and validity of a new five-level triage instrument. Acad Emerg Med. 2000;7:236–242. MEDLINE | CrossRef 23. 23 Implementation and refinement of the Emergency Severity Index. Acad Emerg Med. 2001;8:170–176. MEDLINE | CrossRef 24. 24. A reason for visit classification for ambulatory care. Vital Health Stat 2. 1979;78:1–63. MEDLINE 25. 25. Generalized Linear Models. 2nd ed.. London, England: Chapman & Hall/CRC; 1989;. 26. 26. The Elements of Statistical Learning: Data Mining, Inference and Prediction. New York, NY: Springer-Verlag; 2001;. 27. 27 Are the uninsured responsible for the increase in emergency department visits in the United States?. Ann Emerg Med. 2008;52:108–115. Abstract | Full Text | Full-Text PDF (725 KB) | CrossRef 28. 28 Adding more beds to the emergency department or reducing admitted patient boarding times: which has a more significant influence on emergency department congestion?. Ann Emerg Med. 2009;53:575–585. Abstract | Full Text | Full-Text PDF (588 KB) | CrossRef 29. 29. Access block causes emergency department overcrowding and ambulance diversion in Perth, Western Australia. Emerg Med J. 2005;22:351–354. CrossRef 30. 30. Keys to patient satisfaction in the emergency department: results of a multiple facility study. Hosp Health Serv Adm. 1996;41:515–532. MEDLINE 31. 31 Determinants of patient satisfaction and willingness to return with emergency care. Ann Emerg Med. 2000;35:426–434. Abstract | Full Text | Full-Text PDF (841 KB) | CrossRef 32. 32. Validating a model of patient satisfaction with emergency care. Ann Emerg Med. 2001;38:527–532. Abstract | Full Text | Full-Text PDF (79 KB) | CrossRef 33. 33. Patient satisfaction in the emergency department: a review of the literature and implications for practice. J Emerg Med. 2004;26:13–26. Abstract | Full Text | Full-Text PDF (138 KB) | CrossRef 34. 34. Total Customer Satisfaction: A Comprehensive Approach for Health Care Providers. San Francisco, CA: Jossey-Bass; 1999;. 35. 35. Managing patient waits in hospital emergency departments. Health Care Manag. 2002;21:46–61. 36. 36. Managing the clinic wait: an important quality of care challenge. J Nurs Care Qual. 1998;13:11–20. MEDLINE 37. 37 The effect of hospital occupancy on emergency department length of stay and patient disposition. Acad Emerg Med. 2003;10:127–133. MEDLINE | CrossRef 38. 38 Factors associated with longer ED lengths of stay. Am J Emerg Med. 2007;25:643–650. Abstract | Full Text | Full-Text PDF (143 KB) | CrossRef 39. 39 Decreasing lab turnaround time improves emergency department throughput and decreases emergency medical services diversion: a simulation model. Acad Emerg Med. 2008;15:1130–1135. CrossRef 40. 40Alberti G. Emergency care ten years on: reforming emergency care [Department of Health]. Accessed December 30, 2008. 41. 41. See and treat: a management driven method of achieving targets or a tool for better patient care? (one size does not fit all). Emerg Med J. 2003;20:118. MEDLINE | CrossRef 42. 42 Assessing the impact of streaming in a regional emergency department. Emerg Med Australas. 2008;20:221–227. 43. 43. Designing the accident and emergency system: lessons from manufacturing. Emerg Med J. 2003;20:126–130. MEDLINE | CrossRef a Department of Emergency Medicine, Johns Hopkins University School of Medicine, Baltimore, MD b Department of Biostatistics, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD c Department of Emergency Medicine, University of Michigan School of Medicine, Ann Arbor, MI d Department of Emergency Medicine, George Washington University School of Medicine, Washington, DC e Department of Biomedical Informatics and the Department of Emergency Medicine, Vanderbilt University Medical Center, Nashville, TN
Provide feedback on this article at the journal's Web site, www.annemergmed.com. Supervising editor: Donald M. Yealy, MD Author contributions: All of the authors were involved in the study concept and design, drafting of the article, and critical revision of the article for important intellectual content. The objectives, data collection protocol, review of the analysis, and findings were discussed by teleconference calls with all of the investigators. MLM, JSD, JL, and DA were responsible for acquiring the data from their sites and obtaining institutional review board approval. RD performed the data analysis under the supervision of MLM and SLZ. However, all of the authors had input into the variables considered for the analysis and how it was conducted. MLM drafted the article, and all authors contributed substantially to its revision. MLM takes responsibility for the paper as a whole. Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article that might create any potential conflict of interest. The authors have stated that no such relationships exist. See the Manuscript Submission Agreement in this issue for examples of specific conflicts covered by this statement. Earn CME Credit: Continuing Medical Education is available for this article at: www.ACEP-EMedHome.com. Publication date: Available online May 6, 2009. PII: S0196-0644(09)00239-X doi:10.1016/j.annemergmed.2009.03.006 © 2009 American College of Emergency Physicians. Published by Elsevier Inc. All rights reserved.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
FULL TEXT
Crowding Delays Treatment and Lengthens Emergency Department Length of Stay, Even Among High-Acuity Patients
00239-X/journalimage?img=PIIS019606440900239X.gr1.lrg.gif&fig=fig1&kwhquery=&issn=0196-0644&ishighres=false&allhighres=false&free=yes','journalimage');)
00239-X/journalimage?img=PIIS019606440900239X.gre1a.lrg.gif&fig=fige1&kwhquery=&issn=0196-0644&ishighres=false&allhighres=false&free=yes','journalimage');)
00239-X/journalimage?img=PIIS019606440900239X.gre1b.lrg.gif&fig=fige1&kwhquery=&issn=0196-0644&ishighres=false&allhighres=false&free=yes','journalimage');)
00239-X/journalimage?img=PIIS019606440900239X.gre1c.lrg.gif&fig=fige1&kwhquery=&issn=0196-0644&ishighres=false&allhighres=false&free=yes','journalimage');)
00239-X/journalimage?img=PIIS019606440900239X.gr2.lrg.gif&fig=fig2&kwhquery=&issn=0196-0644&ishighres=false&allhighres=false&free=yes','journalimage');)
