Impact of a standardized communication system on continuity of care between family physicians and the emergency department.

OBJECTIVE: It has been suggested that continuity of care is hampered because of the lack of communication between emergency departments (EDs) and primary care providers. A web-based, standardized communication system (SCS) that enables family physicians (FPs) to visualize information regarding their patients' ED visits was developed. This paper aims to evaluate the impact of this SCS on continuity of care. METHODS: We conducted an open, 4-period crossover, cluster-randomized controlled trial of 23 FP practices. During the intervention phase, FPs received detailed reports via SCS, while in the control phase they received mailed copies of the ED notes. Continuity of care was evaluated with a web questionnaire completed by FPs 21 days after the ED visit. The primary measures of continuity of care were knowledge of ED visit (quality and quantity), patient management and follow-up rate. RESULTS: We analyzed a total of 2022 ED visits (1048 intervention and 974 control). The intervention group received information regarding the ED visit more often (odds ratio [OR] 3.14, 95% confidence interval [CI] 2.6-3.79), found the information more useful (OR 5.1, 95% CI 3.49-7.46), possessed a better knowledge of the ED visit (OR 6.28, 95% CI 5.12-7.71), felt they could better manage patients (OR 2.46, 95% CI 2.02-2.99) and initiated actions more often following receipt of information (OR 1.62, 95% CI 1.36-1.93). However, there was no significant difference in the follow-up rate at FPs offices (OR 1.25, 95% CI 0.97-1.61). CONCLUSION: The use of SCS between an ED and FPs led to significant improvements in continuity of care by increasing the usefulness of transferred information and by improving FPs' perceived patient knowledge and patient management.

Nonurgent emergency department patient characteristics and barriers to primary care.

OBJECTIVE: Nonurgent (NU) emergency department (ED) use is at the forefront of medico-political agendas, and diversion of NU patients has been entertained as a management strategy. Before policy changes are implemented, this population should be better understood with respect to their characteristics and reasons for not presenting to primary care providers (PCPs) instead of EDs. This study compares NU with urgent and semiurgent (USU) patients and describes the NU patients' reasons for not seeking care with a PCP before presenting to the ED. METHODS: This was a secondary analysis from a cross-sectional study with sequential sampling in the EDs of five Quebec tertiary care hospitals (October 19, 1999, to May 26, 2000). Data on medical history, social support, awareness and utilization of health care, ED visits, referrals, activities of daily living, and sociodemographics were obtained. The NU group included patients with triage code 5 and the USU group included patients with triage codes 2, 3, and 4 using the Canadian Triage and Acuity Scale. Patient characteristics were structured into the Andersen behavioral model for health care utilization. RESULTS: Of 2,348 patients approached, 1,783 patients (77%) were eligible and agreed to participate. NU patients (n = 454) were younger than USU patients (n = 1,329) (mean age, 43 [SD +/- 18.1] vs. 49 [SD +/- 20.1] years). Patients in the NU group had better health (number of prior conditions, 3.1 vs. 3.9), were less likely to arrive by ambulance (5% vs. 22%), and were less often admitted from the ED (4% vs. 24%). While 70% of NU compared with 75% of USU patients were followed up by a PCP, only 22% of NU and 27% of USU patients sought PCP care before presenting to the ED. The reasons given by NU patients for not seeking PCP care were accessibility (32%), perception of need (22%), referral/follow-up to the ED (20%), familiarity with the ED (11%), trust of the ED (7%), and no reason (7%). CONCLUSIONS: NU ED patients are different from USU patients and have multiple reasons for not seeking primary care before going to the ED. This may help explain why various diversion strategies have been unsuccessful and indicate that a multifaceted approach may be better suited to this group of patients. The design of new interventions, however, will benefit from further research that clarifies the impact of NU patients on the health care system.

Impact of ambulance transportation on resource use in the emergency department.

OBJECTIVE: To determine how ambulance transportation is associated with resource use in the emergency department (ED). METHODS: A retrospective administrative database review of patient visits to a Montreal tertiary care hospital ED in one year (April 2000-March 2001). Measures of resource use included ED length of stay, admission to the hospital, and whether consultations and radiology/imaging tests (excluding plain-film x-rays) were ordered from the ED. RESULTS: During the study period, 39,674 patients made 59,142 visits to the ED. Ambulance transportation was used for 15.6% of these ED visits. Compared with non-ambulance visits, ambulance visits were more likely to be made by older patients (mean age: 68 vs. 47 years), to be made by females (59% vs. 55%), to have a greater triage urgency score (mean on 1-5 scale, with 1 most urgent: 2.7 vs. 3.9), and to occur after office hours, 5 PM to 9 AM (47% vs. 43%). Ambulance visits were also more likely than non-ambulance visits to result in: a longer length of stay (mean: 13.3 hours [95% CI = 13.0 to 13.6] vs. 5.9 [95% CI = 5.8 to 6.0]), hospital admission (40% vs. 10%) (odds ratio [OR]: 5.94 [95% CI = 5.59 to 6.33]), consultations (56% vs. 20%) (OR: 5.15 [95% = 4.86 to 5.45]), and radiology/imaging tests (20% vs. 12%) (OR: 1.93 [95% CI = 1.81 to 2.07]). In multivariate models that adjusted for the effects of age, gender, triage urgency, and temporal factors, ambulance transportation maintained its association with greater resource use. CONCLUSIONS: This preliminary study indicates that patients arriving at the ED by ambulance use significantly more resources than their walk-in counterparts.

Development of the Canadian Emergency Department Diagnosis Shortlist.

OBJECTIVE: Managers of emergency departments (EDs), governments and researchers would benefit from reliable data sets that characterize use of EDs. Although Canadian ED lists for chief complaints and triage acuity exist, no such list exists for diagnosis classification. This study was aimed at developing a standardized Canadian Emergency Department Diagnosis Shortlist (CED-DxS), as a subset of the full International Classification of Diseases, 10th revision, with Canadian Enhancement (ICD-10-CA). METHODS: Emergency physicians from across Canada participated in the revision of the ICD-10-CA through 2 rounds of the modified Delphi method. We randomly assigned chapters from the ICD-10-CA (approximately 3000 diagnoses) to reviewers, who rated the importance of including each diagnosis in the ED-specific diagnosis list. If 80% or more of the reviewers agreed on the importance of a diagnosis, it was retained for the final revision. The retained diagnoses were further aggregated and adjusted, thus creating the CED-DxS. RESULTS: Of the 83 reviewers, 76% were emergency medicine (EM)-trained physicians with an average of 12 years of experience in EM, and 92% were affiliated with a university teaching hospital. The modified Delphi process and further adjustments resulted in the creation of the CED-DxS, containing 837 items. The chapter with the largest number of retained diagnoses was injury and poisoning (n = 292), followed by gastrointestinal (n = 59), musculoskeletal (n = 55) and infectious disease (n = 42). Chapters with the lowest number retained were neoplasm (n = 18) and pregnancy (n = 12). CONCLUSION: We report the creation of the uniform CED-DxS, tailored for Canadian EDs. The addition of ED diagnoses to existing standardized parameters for the ED will contribute to homogeneity of data across the country.

A workshop to improve workflow efficiency in emergency medicine.

OBJECTIVE: The emergency department (ED) environment requires physicians to focus on workflow efficiency (WFE) and manage ED throughput. We sought to determine whether an interactive workshop could be designed and favourably perceived by emergency physicians and residents as a means to improve their self-assessed WFE skills. METHODS: The authors designed a 4-station workshop to simulate key components of ED throughput. These included resource management in 1) acute care, 2) minor care, 3) charting and 4) communication skills and patient sign-overs. Anonymous surveys were completed after each workshop using 5-point Likert scales and qualitative responses. Qualitative data encompassed participants' past WFE training experiences and perspectives on the current workshop. Data were analyzed using descriptive statistics. The workshops were administered on 2 separate occasions to different groups of physicians. The first occasion was primarily for residents and the second session was only for practising physicians. RESULTS: A total of 22 residents and 24 practising physicians participated. Evaluations were completed by 45 of 46 participants. Ratings of "definitely helpful" or "helpful" as noted for each station were received by 37 of 44 respondents for the sign-over and communication station, by 37 of 44 for the minor care station, by 41 of 44 for the acute care station and by 33 of 43 for the effective charting station. Among all participants, 42 of 45 reported that they felt the overall workshop experience was "helpful" or "definitely helpful." CONCLUSION: ED management "flow skills" are valued yet undertaught. A flow workshop designed to improve self-perceived WFE skills yields positive evaluations. Teaching this competency in a workshop setting is both feasible and appreciated by participants. Similar efforts should be considered for inclusion in both graduate and continuing medical education curricula.