Reduction of opioid prescribing through the sharing of individual physician opioid prescribing practices.

BACKGROUND: Drug overdoses are the most common cause of accidental death in the United States, with the majority being attributed to opioids. High per capita opioid prescribing is correlated with higher rates of opioid abuse and death. We aimed to determine the impact of sharing individual prescribing data on the rates of opioid prescriptions written for patients discharged from the emergency department (ED). METHODS: This was a pre-post intervention at a single community ED. We compared opioid prescriptions written on patient discharge before and after an intervention consisting of sharing individual and comparison prescribing data. Clinicians at or over one standard deviation above the mean were notified via standard template electronic communication. RESULTS: For each period, we reported the median number of monthly prescriptions written by each clinician, accounting for the total number of patient discharges. The pre-intervention median was 12.5 prescriptions per 100 patient discharges (IQR 10-19) compared to 9 (IQR 6-11) in the post-intervention period (p < 0.001). This represents a 28% reduction in the overall rate of opioid prescriptions written per patient discharged. Using interrupted time series analysis for monthly rates, this was associated with a reduction in opioid prescriptions, showing a decrease of almost 9 prescriptions for every 100 discharges over the 6 months of the study (p = 0.032). CONCLUSION: Our study demonstrates the sharing of individual opioid prescribing data was associated with a reduction in opioid prescribing at a single institution.

New media use by patients who are homeless: the potential of mHealth to build connectivity.

BACKGROUND: Patients experiencing homelessness represent a disproportionate share of emergency department (ED) visits due to poor access to primary care and high levels of unmet health care needs. This is in part due to the difficulty of communicating and following up with patients who are experiencing homelessness. OBJECTIVE: To determine the prevalence and types of "new media" use among ED patients who experience homelessness. METHODS: This was a cross-sectional observational study with sequential enrolling of patients from three emergency departments 24/7 for 6 weeks. In total, 5788 ED patients were enrolled, of whom 249 experienced homelessness. Analyses included descriptive statistics, and unadjusted and adjusted odds ratios. RESULTS: 70.7% (176/249) of patients experiencing homelessness own cell phones compared to 85.90% (4758/5539) of patients in stable housing (P=.001) with the former more likely to own Androids, 70% (53/76) versus 43.89% (1064/2424), and the latter more likely to have iPhones, 44.55% (1080/2424) versus 17% (13/76) (P=.001). There is no significant difference in new media use, modality, or frequency for both groups; however, there is a difference in contract plan with 50.02% (2380/4758) of stably housed patients having unlimited minutes versus 37.5% (66/176) of homeless patients. 19.78% (941/4758) of patients in stable housing have pay-as-you-go plans versus 33.0% (58/176) of homeless patients (P=.001). Patients experiencing homelessness are more likely to want health information on alcohol/substance abuse, mental health, domestic violence, pregnancy and smoking cessation. CONCLUSIONS: This study is unique in its characterization of new media ownership and use among ED patients experiencing homelessness. New media is a powerful tool to connect patients experiencing homelessness to health care.

A simulation-based biodefense and disaster preparedness curriculum for internal medicine residents.

AIMS: Disaster and bioterrorism preparedness is poorly integrated into the curricula of internal medicine residency programs. Given that victims may present to a variety of healthcare venues, including primary care practices, inpatient hospital wards, and intensive care units, we developed a curriculum to address this need. METHODS: The curriculum consisted of four didactic sessions with supplemental readings covering biologic, chemical, and radiologic agents, as well as public health infrastructure. All 30 internal medicine resident participants also underwent a four hour training seminar at a high fidelity human simulation center. Instruction included the use of personal protective equipment (PPE)and participation in simulated scenarios utilizing technologically sophisticated mannequins with monitoring and interactive capability. Sessions were videotaped, reviewed with participants, and followed by self-evaluation and constructive feedback. RESULTS: Compared to a control group of residents who did not undergo training, the participants' level of knowledge was significantly better, with mean objective test scores of 66.8%+/-11.8% SD vs. 50%+/-13.1% SD, p < 0.0001. Although there was a trend toward increasing knowledge with increasing level of training in the control group, this difference was not significant. Subjective preparedness was also significantly better in the intervention group (p < 0.0001). Objective improvements were not maintained after one year. CONCLUSIONS: In this pilot study, a disaster-preparedness curriculum including simulation-based training had a positive effect on residents' knowledge base and ability to respond to disaster. However, this effect had diminished after one year, indicating the need for reinforcement at regular intervals.

Variation in opioid prescribing patterns between ED providers.

Abuse of opioid prescription drugs has become an epidemic across the developed world. Despite the fact that emergency physicians overall account for a small proportion of total opioids prescribed, the number of prescriptions has risen dramatically in the past decade and, to some degree, contributes to the available supply of opioids in the community, some of which are diverted for non-medical use. Since successfully reducing opioid prescribing on the individual level first requires knowledge of current prescribing patterns, we sought to determine to what extent variation exists in opioid prescribing patterns at our institution. This was a single-institution observational study at a community hospital with an annual ED volume of 47,000 visits. We determined the number of prescriptions written by each provider, both total number and accounting for the number of patients seen. Our primary outcome measure was the level of variation at the physician level for number of prescriptions written per patient. We also identified the mean number of pills written per prescription. We analyzed data from November 13, 2014 through July 31, 2015 for 21 full-time providers. There were a total of 2211 prescriptions for opioids written over this time period for a total of 17,382 patients seen. On a per-patient basis, the rate of opioid prescriptions written per patient during this period was 127 per 1000 visits (95 % CI 122-132). There was a variation on the individual provider level, with rates ranging from 33 per to 332 per 1000 visits. There was also substantial variation by provider in the number of pills written per prescription with coefficient of variation (standard deviation divided by mean) averaged over different opioids ranging from 16 to 40 %. There was significant variation in opioid prescribing patterns at the individual physician level, even when accounting for the number of patients seen.

The Prevalence and Characteristics of Emergency Medicine Patient Use of New Media.

BACKGROUND: Little is known about "new media" use, defined as media content created or consumed on demand on an electronic device, by patients in emergency department (ED) settings. The application of this technology has the potential to enhance health care beyond the index visit. OBJECTIVE: The objectives are to determine the prevalence and characteristics of ED patients' use of new media and to then define and identify the potential of new media to transcend health care barriers and improve the public's health. METHODS: Face-to-face, cross-sectional surveys in Spanish and English were given to 5,994 patients who were sequentially enrolled from July 12 to August 30, 2012. Data were collected from across a Southern Connecticut health care system's 3 high-volume EDs for 24 hours a day, 7 days a week for 6 weeks. The EDs were part of an urban academic teaching hospital, an urban community hospital, and an academic affiliate hospital. RESULTS: A total of 5,994 (89% response rate) ED patients reported identical ownership of cell phones (85%, P<.001) and smartphones (51%, P<.001) that were used for calling (99%, P<.001). The older the patient, however, the less likely it was that the patient used the phone for texting (96% vs 16%, P<.001). Income was positively associated with smartphone ownership (P<.001) and the use of health apps (P>.05) and personal health records (P<.001). Ownership of iPhones compared to Android phones were similar (44% vs 45%, P<.05). Race and ethnicity played a significant role in texting and smartphone ownership, with Hispanics reporting the highest rates of 79% and 56%, respectively, followed by black non-Hispanics at 77% and 54%, respectively, and white non-Hispanics at 65% and 42%, respectively (P<.05). CONCLUSIONS: There is a critical mass of ED patients who use new media. Older persons are less comfortable texting and using smartphone apps. Income status has a positive relationship with smartphone ownership and use of smartphone apps. Regardless of income, however, texting and ownership of smartphones was highest for Latinos and black non-Latinos. These findings have implications for expanding health care beyond the ED visit through the use of cell phones, smartphones, texting, the Internet, and health care apps to improve the health of the public.

Evaluation of a teaching laboratory using a cadaver model for tube thoracostomy(1).

A prior study evaluated the efficacy of a dog laboratory to teach residents chest tube thoracostomy. This study evaluated a similarly structured program using human cadavers. A prospective repeat measure study of chest tube thoracostomy placement training was performed in a university laboratory setting using human cadavers. Ten Emergency Medicine residents were given a written pretest, followed by training. Resident attempts were then timed. The following day, a repeat test was administered. Three weeks later, a third written post-test was conducted. The written test scores improved for every participant. Mean times for procedure completion improved from 86 sec to 34 sec during the first session, and remained stable over 4 attempts from 30 sec to 32 sec during the second session. This approach to teaching clinical procedures should be considered for Emergency Medicine residency programs and for continuing education courses that emphasize procedural skills.

Cardiopulmonary resuscitation using the cardio vent device in a resuscitation model.

To compare the "Bellows on Sternum Resuscitation" (BSR) device that permits simultaneous compression and ventilation by one rescuer with two person cardiopulmonary resuscitation (CPR) with bag-valve-mask (BVM) ventilation in a single blind crossover study performed in the laboratory setting. Tidal volume and compression depth were recorded continuously during 12-min CPR sessions with the BSR device and two person CPR. Six CPR instructors performed a total of 1,894 ventilations and 10,532 compressions in 3 separate 12-min sessions. Mean tidal volume (MTV) and compression rate (CR) with the BSR device differed significantly from CPR with the BVM group (1242 mL vs. 1065 mL, respectively, p = 0.0018 and 63.2 compressions per minute (cpm) vs. 81.3 cpm, respectively, p = 0.0076). Error in compression depth (ECD) rate of 9.78% was observed with the BSR device compared to 8.49% with BMV CPR (p = 0.1815). Error rate was significantly greater during the second half of CPR sessions for both BSR and BVM groups. It is concluded that one-person CPR with the BSR device is equivalent to two-person CPR with BVM in all measured parameters except for CR. Both groups exhibited greater error rate in CPR performance in the latter half of 12-min CPR sessions.

A comparison of emergency department versus inhospital chest pain observation units.

US hospitals use observation units (OUs) for safe and cost-effective management of low-risk to moderate-risk patients presenting to the ED with chest pain. This study retrospectively compared the utility of an ED observation unit (EDOU) with an inhospital observation unit (IHOU) for chest pain at the same institution. A 5-month period during which patients with chest pain were admitted to the EDOU was compared with a 5-month period during which patients with chest pain were admitted to the IHOU. During the 5-month EDOU period, 440 (36.9%) of 1190 patients with chest pain presenting to the ED were admitted for observation. During the IHOU period, 973 (69.3%) of 1404 patients with chest pain presenting to the ED were admitted for observation (P<.0001). Fewer patients with chest pain were converted to full inpatient admission from the EDOU, 35 (7.9%) of 440, when compared with the IHOU, 187 (19.2%) of 973 (P<.0001). Mean cost for each patient was US $889.87 (95% CI 862.8-916.9) versus US $1039.70 (95% CI 991.7-1087.7) for each IHOU patient. We conclude that the EDOUs are more cost-effective than IHOUs for management of low-risk to moderate-risk patients with chest pain.