Critical Appraisal of Literature Focused on Throughput in ...



Critical Appraisal of Literature Focused on Throughput in the Emergency Department

Cheryl Miller & Lori Majeske

Ferris State University

Abstract

The following literature review provides current evidence of the challenges surrounding overcrowding in the emergency department, and possible creative solutions designed to reduce waiting times, improve patient care and satisfaction, and hospital throughput. It includes a critical appraisal of current literature evidence, and a project proposal plan for further research.

Critical Appraisal of Literature Focused on Throughput in the Emergency Department

In our observation as nurses, emergency room overcrowding and extended waiting times can negatively affect patient care, and overall patient satisfaction results. Our research goal is to implement best practice guidelines through development of a model for care that will support efficient and optimal patient care in the hospital emergency department. The following literature review supports our project proposal by providing timely and evidenced based examples of proposed interventions focused on improving patient care in the emergency department setting. Each literature source has individual merit and builds a body of evidence to support our research proposal by outlining some of the interventions currently under consideration.

Impact of Rapid Entry and Accelerated Care at Triage on Reducing Emergency Department Patient Wait Times, Lengths of Stay, and Rate of Left Without Being Seen

In their systematic review article Impact of Rapid Entry and Accelerated Care at Triage on Reducing Emergency Department Patient Wait Times, Lengths of Stay, and Rate of Left Without Being Seen (2005), Chan, Killeen, Kelly and Guss introduced a rapid entry and accelerated care at triage (REACT) process aimed specifically at reducing the number of patients who leave the emergency department prior to being evaluated by a physician.

Critical Appraisal

This is an example of a quantitative quasi-experimental research study. The title of the report explains the purpose of the study and is consistently written. The purpose of the study is prominently displayed and logical. The author’s qualifications are consistent with the study proposal as they are all licensed medical professionals. The research study results and conclusions are clearly documented, and prominently displayed on the title page of the study. These authors report a significant decrease in leave before being seen frequency, patient wait times, and overall length of stay. These research findings directly support the proposed study objectives.

Validity

Chan, Killeen, Kelly and Guss (2005) chose a quasi-experimental research study design that seeks to determine the cause and effect relationship between the applied interventions known as REACT upon ambulatory patients presenting to an urban hospital emergency department. The interventions include patient identification tracking, integrated computer interfaces, immediate placement of patients in open beds, and physician directed ancillary testing and care when no beds were available. The sample size of 36,000 is large, and includes all ambulatory patients presenting for care. As is typical with this type of study, there is no randomization of research subjects. These researchers utilized a historical comparison group within the same institutional setting, which they identified as an urban academic emergency room setting. The major strengths of utilizing a quasi-experimental approach is that it is practical to conduct, more likely to be acceptable to participants, and can be generalized to the general population (Ross-Kerr & Wood, 2011). We agree that it is an appropriate choice, as it would be difficult to randomize the applied interventions to assigned groups within this type of setting. Major weaknesses include the passage of time, and the challenge of limiting all possible extraneous variables that may affect study results (Ross-Kerr & Wood, 2011). It is important to note that Chan, Killeen, Kelly and Guss (2005) include current research evidence within their study that supports their claim that most patients leave prior to being seen by a physician due to increased waiting times. This serves to strengthen the study results as it provides proof of internal validity, a strong statistically relationship between the independent and dependent variables, as well as, clear evidence of the true cause and effect relationship identified within the study (Polit & Tatano-Beck, 2008). The authors attempted to gain research control by recording data six months prior to instituting the proposed changes, and re-measuring data again during a similar six month period, one year later in order to reduce the Hawthorne Effect. Ross-Kerr and Wood (2011) note that this important step may enhance interpretability of study findings, and helps reduce the risk of sampling bias.

Core measures include average wait times, emergency room length of stay, emergency department census, and overall admission data. Outcome measures and primary data analysis was obtained by utilizing the hospital medical records database which provided time stamps measurements for overall census, patients who leave before being seen, length of stay, and wait times. Descriptive statistics were performed using a computerized statistical software package, and included use of the central tendency of the sample, and confidence intervals. These descriptive statistics further validate the results of this study. In addition, because this data is applicable and transferable to emergency practice settings, we find that it meets the criteria defined by Polit and Tatano-Beck (2008) for external validity. In addition, Chan, Killeen, Kelly, and Guss (2005) reported a significant decrease in leave before being seen frequency after application of the proposed interventions. Documentation of the research study results and conclusions are clearly, directly supporting the study objectives.

One weakness of this study, as identified by the research authors, is that some results may not be transferrable to all emergency departments. However, they note that overall improvements in registration delays and computer integration in the emergency department, could serve as a model for other emergency rooms settings. We agree. Additional strengths include a strong presentation of findings including the use of tables and graphs, use of evidence based literature that supports the overall purpose of this study and is cited appropriately throughout their review.

Usefulness in Clinical Practice

The research study results reported by Chan, Killeen, Kelly, and Guss (2005) are based upon a variety of interventions that utilize upgrades of information technology, bar coding, software integration, and changes in staff culture, and specific processes changes. As previously stated, although these results may not be transferrable to all emergency departments, overall improvements in registration delays and computer integration in the emergency department, could serve as a model for other emergency rooms settings. Adoption of these proposed changes in other health care institutions could positively affect patient safety, overall patient satisfaction, and improved access to care by allowing for better coordination of patient care services. The end result of this coordination, are partly dependent upon an effective triage system for emergency room patients, as illustrated in the following study.

Telephone consultation and triage: effects on health care use and patient satisfaction

In the systematic review by Bunn, Byrne and Kendall (2009), the use of telephone consultation as a means of determining the need for further treatment by either a General Practitioner (GP) or Emergency Department (ED) visit is studied. The review considers safety, service usage and patient satisfaction, adverse events and compared consultations by different healthcare workers.

Critical Appraisal

This study utilizes a quasi-experimental research design as described by Polit and Beck (2008). This systematic review uses several types of studies, including controlled random trials and interrupted studies. The resulting review is lengthy and describes intervention summaries as determined through the various studies. This type of review although useful, is difficult to summarize the entire content.

Validity

In order to accurately analyze data qualities in quantitative studies, several factors are involved (Polit & Tatano-Beck, 2008). The reviewers did not use a meta-analysis to present the findings, but rather a narrative summary. This is due to the heterogeneity of the study methods. The variables of this study are congruent between the introduction and methodology section of this paper. There is noted analysis error reported by the authors in at least one study. The data extraction and quality assessment was well documented. Data quality was assessed using a modified version of the EPOC checklist. Overall scores of each study were not given. Reliability methods discussions are tabulated using appropriate methods for each type of study. These include P values; slope regression lines were used for interrupted time studies. The reviewers’ document specific information regarding each study concluding the study may have had inappropriate techniques. Cautious interpretation of the findings is recommended due to the variability in interventions and lack of power of some studies. The possibility exists that not all studies could be identified, due in part to the vast terminology used. An adverse effect of validity may occur because not one study met all of the methodological criteria on the EPOC checklist. In addition, selection biases may threaten validity. Sensitivity and specificity are not specifically discussed, only that this review supports previous estimates of calls able to be handled at least fifty percent of the time.

Usefulness in Clinical Practice

Using additional resources, including advanced practice roles, to decrease patient wait times may be beneficial. The need for a communication system to aid in the determination of additional treatment can reduce ED wait times as well as increase the qualitative measure of patient satisfaction. Further research would be needed to ascertain the need for follow up treatment or a resulting decrease in quality of care if the treatment advice is not accepted. The following study outlines the impact of efficient coordination of patient care services and its impact upon providing safe and efficient patient care.

The Role and Impact of the Physician Assistant

The review by Doan, Sabhaney, Kissoon, Sheps, & Singer (2011) examines the role and impact of the Physician Assistant (PA) in relation to department efficiency and patient satisfaction. This review was not intended to determine the use of the nurse practitioner (NP), in fact, studies using NPs alone, were excluded from the study.

Critical Appraisal

This study used non randomized studies and is considered a quantitative methodology demonstrating before and after results of implementing the use of PAs to increase patient satisfaction and reduce wait times resulting in efficiency in the emergency department. The authors of this review are medical professionals, and consulted experts if disagreements were found in the data. The review is clearly explained in the title and is consistent throughout the review. Results are clearly documented and evident in the abstract. According to the findings of this review, there is limited evidence as to whether PAs improve the workflow or are cost effective.

Validity

Congruence of variables associated with this study are found consistently throughout the review. The studies do not delineate any specific type of measurement used for operational definitions. The review does address a level of agreement using Kappa statistic and Quality Assessment Tool for Quantitative Studies Method. The Kappa score is 0.83, with 1.00 being the highest, differing opinions exist but some concensus to 0.60 minimally with 0.75 and above as excellent (Polit & Tatano-Beck, 2008, p. 456). The factor analysis used in this study attempts to deduce the variables of prevalence, guideline and policies, roles and task assignments, quality of care, use, ED flow and costs, and physician attitudes. The last variable includes a cross-sectional survey of senior ED residents acting as patients. These patients determined their level of satisfaction based on the severity of their supposed illness and willingness to be treated by a PA. The methodological quality of these studies is rated as weak with low responses. Two tables are included in the study succinctly displaying study methods and outcomes measures with results. These tables compare the factors of clinical decision and patient management and procedure performance between physicians and PAs. The generalizability of the data is limited. The methodological quality is reported as weak to moderate, with specific qualities listed. These include small number of participants studies, and comparison baselines not adequately adjusted.

Usefulness in Clinical Practice

The use of advanced practice professionals in an emergency department setting can increase patient throughput, thereby increasing patient satisfaction scores and lead to a positive budgetary outcome by decreasing overhead expenses of physician or resident salaries and increasing volume of visits annually. If current trends continue with associated indigent care on the rise, patients may still be able to receive treatment in a timely manner while reducing associated issues seen with co-morbidities in this patient population. The role of nursing in emergency care settings also impacts overall efficiency and throughput, as illustrated in the following research study.

Fast Track to Efficiency

In her journal article Fast Track to Efficiency (2011), Handley summarizes how a new nursing role, an acute medicine assessment nurse, was created to improve patient care and reduce the amount of time that patients spend in the emergency department.

Critical Appraisal

The title is clearly written and explains the general nature of the topic to be discussed. The author is a freelance writer who interviewed nursing staff and physicians regarding their role in conducting this research and their results. A list was compiled of the top sixteen medical conditions requiring treatment, and nurses developed new pathways designed to improve patient triage and throughput. Unstable patients were directed to the emergency department, while stable patients were directed to the medical assessment unit. Because of these nursing initiatives, patient wait times are reported as being reduced from two hours to twenty minutes.

Validity

This journal article is a secondary source document, as described by Polit and Tatano-Beck (2008). It is interesting to read, and current. The overall study results sound promising, and may offer some insight into the benefits of early triaging of patients. However, it does not provide information regarding methodology utilized or concrete use of data collection. The use of this information is therefore limited.

Usefulness in Clinical Practice

This study may offer insight for health care professionals who are engaged in process improvement initiates by offering potential creative options for triaging emergency room patients, and serve as an impetus for encouraging further research in this area. Engaging emergency room staff in this process can be beneficial as presented in the following study focused on throughput efficiency.

Rapid Assessment and Disposition; Applying LEAN in the Emergency Department

In the journal article Rapid Assessment and Disposition: Applying LEAN in the Emergency Department (2009), Eller explains how they instituted LEAN principles within their metropolitan adult acute care hospital emergency department utilizing a quantitative quasi-experimental study design.

Critical Appraisal

The title is clearly written, describes the major goals of rapid assessment and disposition, and is prominently displayed. The research question proposed by Eller (2009) is clear written, logical, and aligned with their final research study results. The purpose of the study is explained in the abstract, which outlines the main goals as reducing emergency department length of stay, reducing diversion, and reducing the number of patients who leave before being evaluated by a physician. This author’s qualifications support and enhance this research study, as she is a bachelor’s prepared licensed registered nurse, clinical educator, and project lead for LEAN implementation in the emergency department. Study results appear in the abstract portion of this journal article using appropriate statistical data. Eller (2009) reports that application of the LEAN principles had a substantial impact on reducing average length of stay, diversion, and overall number of patients who left prior to being treated.

Validity

The identified research design utilized by Eller (2009) is a quasi-experimental research approach, which seeks to determine what relationship applying LEAN principles will have upon average length of stay, diversion, and patients who leave prior to being evaluated by a physician. The interventions utilized include education, workplace organization, and standardization to reduce process variances in registration, triage, patient placement, physician assessment, and testing results. The sample size is large consisting of all patients presenting for care in the emergency department averaging a reported 32,000 annually. This is adequate. Randomization was not utilized in this research study, which is consistent, and appropriate for this type of setting. Baseline historical data was compiled in 2007 utilizing internal hospital metrics that included emergency department length of stay, diversion time, and the number of patients who left before receiving treatment. Data collection and measurement results are presented in graphic form, and in paragraph form utilizing percentages. Eller (2009) ranked baseline results by comparing the data to a study by the Centers for Disease Control (CDC), which adds internal validity, and supports the goals for this study. In addition, the use an emergency severity index (ESI) was used in the data collection process. This is an international standardized measurement, which provides additional credibility to the study data. The following outcome data was monitored and reported on a daily and monthly basis: emergency department length of stay, percentage of patients who left prior to being treated, total hospital diversion, number of radiology patients, length of stay, and emergency department volume. Eller (2009) reported sustained reductions in three of the above measurements at six months, one year, and two years, which further validates the study results. Study results are presented in table form, and adequately discussed. The data methods are described sufficiently, and a thorough analysis was done through comparison of baseline data and post intervention data results. Research results are suggestive of a positive correlation between LEAN initiatives and a reduction in emergency department length of stay, percentage of patients who left prior to being treated, and total hospital diversions. However, there are no study limitations identified by the author, which clearly weakens the study results. Polit and Tatano-Beck (2008) state that researchers need to understand possible study limitations, take precautions to either counteract them, or consider them when interpreting the data. Limitations common to this type of study includes the Hawthorne Effect and the passage of time (Polit & Tatano-Beck, 2008). Eller (2009) cites all relevant sources appropriately in the research study.

Usefulness in Clinical Practice

The research study results reported by Eller (2009) are current, evidence based, well written and appear to be applicable to other emergency department settings. The LEAN initiatives discussed appear to offer value to understanding the issues associated with overcrowding in the emergency department, and possible solutions for consideration. These results may serve as a guideline for future research studies focused on reducing overcrowding in the hospital emergency department. Overall treatment time and measures of emergency room overcrowding are further reviewed in the following study.

The Effect of Triage Diagnostic Standing Orders on Emergency Department Treatment Time

The journal article in the February 2011 volume of the Annals of Emergency Medicine discuss a retrospective nested cohort study of patients receiving treatment at an academic emergency department between January 2007 and August 2009 (Retezar, Bessman, Ding, Zeger, & McCarthy, 2011). This study attempts to determine if initiating standing orders for four common diagnosis decreases the length of stay. The study concludes a substantial reduction in length of stay while further evaluation is warranted to include additional diagnoses.

Critical Appraisal

This nested cohort study analyzing data retrospectively ascertains the answer to the proposed question “Do triage standing orders decrease patient length of stay defined as room placement through disposition?” The study question was clearly explained in the abstract of the article and presented clearly throughout. The study authors are colleagues in an academic based emergency department and are all medical professionals. The conclusions are clearly documented with pertinent summary clearly accessible. The emergency department described in the study had an annual volume of approximately 57,000 visits with 33 acute care rooms and 8 urgent care rooms.

Validity

There is congruence in the variables assessed throughout the review. The data extracted was both from an electronic source and from the retrospective review of charts revealing data that was scanned after disposition. A limited amount of data was retrieved using manual methods. To ascertain the validity of the elements, a random sample was compared. The results show a high reliability of the factor. The study participants were categorized into three groups, a propensity score calculating a likelihood score was noted. This score assisted the researchers to retain variables if the regression coefficient was less than .05. This measurement describes the specificity and sensitivity (Polit & Tatano-Beck, 2008, p. 471). Analysis comparisons using quintiles are described. A succinct table displaying the distribution of variables according to placement within the three participant groups is beneficial to readers. Limitations of the research are presented. These limitations include patients not randomized to treatment assignments, the triage nurse determined discretion of order usage, and the groups could not be balanced on unmeasured confounders such as presence of co-morbidities. The volume of compliance of use of orders was done by volume not individually. The magnitude of the effect may be reduced. The generalizability of the study is also limited, since it was used at only one facility. An appropriateness of order sets was not evaluated, nor was the effect on further orders needed. Morbidity and mortality were not determined based on effect of standing orders used.

Usefulness in Clinical Practice

This research could assist in increasing efficiencies in an emergency department setting because of their time saving ability. In addition, determining which standing orders to initiate can control factors associated with overall patient satisfaction. However, cost assessments will still be a crucial element, as will the effect of unnecessary testing. Efficient order entry has the ability to positively impact emergency room throughput, as illustrated in the following study.

Impact of computerized physician order entry on ED patient length of stay

Length of stay in an emergency department setting can assist in the measurement of practice efficiency and patient satisfaction. Computerized Patient Order Entry (CPOE) as studied by Retezar, Bessman, Ding, Zeger, and McCarthy, (2011) attempts to determine a favorable association with this additional factor, as it has shown much promise in similar critical care settings.

Critical Appraisal

The study is current, being published this year. The researchers could not determine a similar study involving the factors of emergency departments and CPOE. Similar studies were found associating treatment for patients receiving thrombolysis. The hypothesis was clearly written and discussed thorughout the review. The design uses written order entry as a pretest variable, with computerized orders as post test.

Validity

The research method chosen for the study was a before and after design. This is considered a quantitative design study, more commonly referred to as pre-test/post-test design (Polit & Tatano-Beck, 2008, p. 261). The incongruence between variables conceptualized and operationalized is noted. Nurse staffing hours were not taken into consideration. Other unmeasured covariates are likely to affect patient length of stay. The onset of computerized order entry took place between the time frames of pre and post test. This may assist in assuring any validity concerns with user education are not applicable. Student t test compared variables with multiariable linear regression analysis testing for association between the variables. Adjustments for variables were noted. A table succinctly describes the affect of pre and post test results by patient demographic and clinical covariates. The independent inverse association is shown between the variable of discharged and patient admitted. This association confirms the reliability of the data (Polit & Tatano-Beck, 2008, p. 196). The limitations of this study are revealed. The generalizability may be limited as this study took place at only one facility. The baseline length of stay was comparatively long. The data set as previously stated, did not include nurse staffing hours, but physician hours were considered. There are system-, patient- and provider-related variables not measured including patient acuity, seasonal factors and test processing times. Average daily census was lower in post test than in pretest, which may confound relationships. Many other factors affecting emergency department efficiencies were not measured such as quality of care or error rates.

Usefulness in Clinical Practice

Interventions designed to decrease length of stay in an emergency department setting can attribute to other favorable affects. Increasing patient satisfaction, decreasing the number of patients leaving without being seen and decreasing ambulance diversion are concluded. Multiple methods can be utilized for research analsysis as demonstrated by the following study.

Emptying the Corridors of Shame:Organizational Lessions From England’s 4-Hour Emergency Throughput Target

England’s National Health Service mandated patients need to be treated or discharged home within four hours of arrival in 2005 (Weber, Mason, Carter, & Hew, 2011). This target has been gradually instituted at hospitals, receiving financial rewards for meeting thresholds. The meeting of these targets is the responsibility of the administration of the facility. The improvement of clinical outcomes is also a benefit, but some are concerned with the unintentional consequence of facilities reporting false information. It is noted that change is difficult and the qualitative study attempts to provide insight to facilities meeting the imposed target.

Validity

A mixed methods study over a three month period was employed. The participants were categorized into three groups, low, mid and high level performers as determined by published results from the third quarter in the previous year. Sample size was determined by saturation, that is samples are obtained until no new information emerged from data analysis (Polit & Tatano-Beck, 2008, p. 357). Methods using both face-to-face and telephone interviews were conducted. The questions were open-ended and semistructured, the use of open-ended questions can result in a richer understanding of the topic of interest (Polit & Tatano-Beck, 2008, p. 415). The data was categorized into themese by the researchers, this use of data collection can then quantify the qualitative information (Polit & Tatano-Beck, 2008, p. 374). After interviewing coding by two investigators, interrater reliability improved after discussion between the two investigators. Further concurrence of the findings were reached by using consultants who were not part of the study and viewed blinded identity information. A table is presented demonstrating results between the variables using themes. Validity may be decreased as this study took place at only one point in time. Further validity may be biased due to the response bias of interviewees only being pooled from the respondents of the quantitative portion that occurred prior to this study. The investigators also discuss other qualitative studies exploring similar variables and enhance their analysis by comparing effects. This can assist with validating findings objectively (Polit & Tatano-Beck, 2008, p. 133)

Uefulness in Clincal Practice

The overall result of the study concludes that achieving a target is largely impacted on the implementation of the change. This information is essential to researchers as it can assist in developing hypotheses and determining types of research needed.

Conclusion and Plan

Based upon our review of these research studies, the interventions of rapid and efficient triage of patients, phone consultation, nursing and provider roles, streamlining of information technology, standing orders, and computerized order entry may have a positive impact upon overcrowding and patient throughput in the emergency department. In addition, as these studies illustrate, the identified problem of overcrowding in the emergency department is complex, and requires an in-depth analysis. Therefore, further research investigation of other possible interventions is warranted, so that solutions can be forthcoming.

Our research proposal and plan includes developing a patient care model to reduce emergency department length of stay and increase overall patient satisfaction. The use of several methods will need to be diligently researched, with considerations for value added processes and economic strategies. Senior administration will be consulted to determine the extent of our proposal. Knowledge gained from this study will be evaluated for both short and long-term impact upon patient throughput, reduced length of stay, and overall patient satisfaction measures.

References

Bunn, F., Byrne, G., & Kendall, S. (2009). Telephone consultation and triage: effects on health care use and patient satisfaction. The Cochrane Collaboration , DOI: 10.1002/14651858.CD004180.pub2.

Chan, T., Killeen, J., Kelly, D., & Guss, D. (2005). Impact of rapid entry and accelerated care at triage on reducing emergency department patient wait times, lengths of stay, and rate of left without being seen. Annals of Emergency Medicine, 491-497.

Doan, Q., Sabhaney, V., Kissoon, N., Sheps, S., & Singer, J. (2011). A systematic review: the role and impact of the physician assistant in the emergency department. Emergency Medicine Australia , 7-15 doi:10.1111/j.1742-6723.2010.01368.x.

Eller, A., (2009). Rapid Disposition: Applying lean in the emergency department. Journal for Healthcare Quality, 33(3), 17-22.

Handley, A., (2011). Fast Track to Efficiency. Nursing Standard, 25(20), 18-19.

Polit, D. F. & Tatano-Beck, C. (2008). Nursing research generating and assessing evidence for nursing practice (8th ed.) Philadelphia, PA: Lippincott Williams & Wilkins.

Retezar, R., Bessman, E., Ding, R., Zeger, S., & McCarthy, M. (2011). The effect of triage diagnostic standing orders on emergency department treatment time. Annals of Emergency Medicine , 89-99.

Ross-Kerr, J. C., & Wood, M. J. (2011). Basic steps in planning nursing research (7th ed.) Sudbury, MA: Jones & Bartlett Publishers.

Spalding, S., Mayer, P., Ginde, A., Lowenstein, S., & Yaron, M. (2011). Impact of computerized physician order entry on ED patient length of stay. American Journal of Emergency Medicine , 201-211.

Weber, E., Mason, S., Carter, A., & Hew, R. (2011). Emptying the corridors of shame: organizational lessons from england's 4-hour emergency throughput target. Annals of Emergency Medicine , 79-88.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download