Correspondence / American Journal of Emergency Medicine 37 (2019) 1362-1393 1379

May DD, Grubbs LM. The extent, nature, and precipitating factors of nurse assault among three groups of registered nurses in a regional medical center. J Emerg Nurs 2002;28:11-7.

Wong AH, Taylor RA, Ray JM, Bernstein SL. Physical restraint use in Adult Pa- tients Presenting to a General Emergency Department. Ann Emerg Med 2018

Table 1

injury characteristics.

Number of patients (%)

Life on the farm: A community-based study of tractor-related injuries and fatalities

Agriculture is a particularly dangerous industry when taking into ac- count occupational injury and fatality rates. In 2010, the fatality rate for agriculture was 27.9 per 100,000 workers, which is dramatically higher than the reported incidence of 3.6 deaths per 100,000 workers for all other occupations combined [1]. Tractor mishaps are a significant cause of machinery-related injuries on farms, accounting for an esti- mated 4-14% of nonfatal injuries and one-third of fatal agricultural inju- ries [2]. Safety equipment such as rollover protective structures, master shields covering the power take-off (PTO), and driveline and power input connection guards have been introduced in newer tractors in order to help prevent injury [3]. However many tractors currently in use were manufactured prior to these installments and operators may find it too expensive or cumbersome to have their equipment fitted with these devices [3]. Misuse of equipment, having children on or around these machines, and use of tractors on public roads also intro- duce an increased chance for injury or death [3]. Emergency department (ED) personnel may have little or no familiarity or knowledge about the hazards of agricultural work, including what to expect with tractor- related injuries. The purpose of this study was to characterize the inci- dence, injury characteristics, and outcomes of tractor-related injuries in West Michigan.

We conducted a retrospective cohort analysis of ED patients with a tractor-related injury seen in seven participating hospitals in West Michigan from January 2002 to August 2016. Data was collected on de- mographics, mechanisms of injury, diagnoses, treatment provided, and mortality rate. Additional farm fatalities secondary to tractor-related in- cidents were identified using data from the Michigan Fatality Assess- ment and Control Evaluation (MIFACE) research program [4]. These fatalities occurred in West Michigan during the study period but were not seen in participating hospitals. Descriptive statistics and frequency tables were used to describe the key quantitative and qualitative variables.

We evaluated 208 patients presenting to participating hospitals with 507 tractor-related injuries during the study period. The mean age was

42.0 +- 20.0 with a range of 3-86 years. Twenty-nine patients (13.9%) were b18 years of age and 11 (5.3%) were N70 years old. A total of 73 pa- tients (35.1%) had injuries classified as severe (trauma priority 1 or 2). Leading types of injury in all age groups were fractures (n = 119, 57.2%) followed by open wounds and contusions/abrasions (Table 1). The majority of fractures involved the upper or lower extremities (79.3%), followed by fractures to the spine (11.5%), pelvis (4.8%) and skull (4.4%).

Overall, the most common mechanisms of nonfatal tractor-related injuries were having an extremity pinned, caught or lacerated (n = 81, 38.9%), followed by tractor run over (n = 34, 16.4%) (Table 2). The frequencies of the injury mechanisms were different in children with

the most common mechanism being a fall/jump/ejection (n = 16, 55.2%), followed by a run over (n = 7, 24.1%). A total of 76 patients (36.9%) were admitted to the hospital (LOS 7.9 +- 7.7 days) with four subsequent fatalities (1.9%).

According to MIFACE, a total of 119 agriculture fatalities occurred in Western Michigan between 2002 and 2016, and 63 (52.9%) were tractor-related [4]. Causes of death included tractor overturn (46%), run over (16%), and road collisions (11%), among others. Documented reasons for these fatalities include a relative lack of safety training and seat belt use, fatigue and carelessness, lack of emergency preparedness, and exposure of High risk groups like children and elderly to hazardous environments.

Our study supports facts previously found in that various pro- cesses may lead to a tractor-related injury, and these mechanisms vary by age and impact on the severity of the resulting injury [1]. Although fractures and soft-tissue extremity wounds are com- monly reported, in our study approximately one-third of patients had severe traumatic injuries. ED physicians must be prepared to handle a wide variety of injuries in tractor-related incidents. A par- ticularly disturbing aspect of this study is the frequency of injuries to children, who make up a substantial portion of rural farm staff. It is common for children to work in agriculture at a young age, espe- cially since they are protected by Federal labor laws implemented by the US Department of Labor to a lesser extent than children in other areas of the workforce. For example, laws declare it hazard- ous for children under 16 years of age to operate a tractor and other heavy machinery on farms, but in other realms, individuals must be 18 or older to operate hazardous equipment [5]. Investiga- tion of unsafe working conditions is often completed by the Office of Safety and Health Administration (OSHA), of which there are both Federal and State offices [6]. Reports to these offices must be taken seriously and investigated thoroughly. Ensuring proper training and oversight for agricultural workers is a must. It would also be prudent to ensure heavy machinery such as tractors either

Table 2

Tractor related injuries in West Michigan 2002-2016.

1380 Correspondence / American Journal of Emergency Medicine 37 (2019) 1362-1393

come equipped or are retrofitted with the most current safety equipment since rollover deaths have been documented since the 1920s [7]. Finally, training and information for ED doctors, first re- sponders, and other care givers about the hazards and possible out- comes of tractor injuries, especially in Rural healthcare systems, could lead to faster Response times and more comprehensive care [7].

Daniel Vryhof¹ Todd Chassee¹ Matt Singh¹

Spectrum Health - Michigan State University Emergency Medicine Residency Program, Grand Rapids, MI, United States

Helen DeVos Children’s Hospital, Grand Rapids, MI, United States

Lindsey Ouellette²

Helen DeVos Children’s Hospital, Grand Rapids, MI, United States

Jeffrey Jones* Spectrum Health - Michigan State University Emergency Medicine Residency Program, Grand Rapids, MI, United States

Michigan State University College of Human Medicine, Department of Emergency Medicine, Grand Rapids, MI, United States

*Corresponding author at: 15 Michigan St NE Suite 701, Grand Rapids,

MI 49503, United States of America.

E-mail address: [email protected].

13 November 2018

https://doi.org/10.1016/j.ajem.2018.12.035

References

1. Jawa RS, Young DH, Stothert JC, et al. Farm machinery injuries: the 15-year experi- ence at an urban joint trauma center system in rural state. J Agromedicine 2013;18: 98-106.
2. Swanton AR, Young TL, Leinenkugel K, et al. Nonfatal tractor-related injuries present- ing to a state trauma system. J Safety Res 2015;53:97-102.
3. Jennissen C, Mathiasen R. An introduction to traumatic injuries and environmental

   hazards on the farm: part I tractors - what the heck is a PTO? J Rural Emerg Med 2015;2(1):12-25.

   Michigan State University College of Human Medicine, Occupational and Environmental Medicine. Work related fatalities. http://www.oem.msu.edu/index. php/work-related-injuries/work-related-fatalities. [accessed 17 October 2018].

4. National Farm Worker Ministry. Children in the fields. http://nfwm.org/resources/ children-in-the-fields/. [accessed 22 October 2018].
5. United States Department of Labor. Occupational Safety and Health Administration. https://www.osha.gov/about.html. [accessed 4 December 2018].
6. Myers JR, Hendricks KJ. Agricultural tractor overturn deaths: assessment of trends and risk factors. Am J Ind Med 2010;53:662-72.

   Refining reflex urine culture testing in the ED

   Screening for urinary tract infections is often performed by uri- nalysis (UA) which measures specific urine elements such as ni- trites, white blood cells , and Leukocyte esterase (LE). In our Emergency Department (ED), a positive UA, defined as positive nitrite, or N3 WBC, or positive LE, will automatically generate a urine culture to identify and quantify bacteriuria for specimens or- dered as a UA with reflex to culture. Our institution’s policy is that antibiotic treatment will be provided for patients with urine cul- ture results yielding 60,000 colony-forming units (cfu) per millili- ter (ml). Culture results of >=60,000 cfu per ml are reviewed by an ED physician to determine the course of action and treatment. Al- though treatment is based on chart review, there is inclination to treat positive UA and culture results >=60,000 cfu per ml. The Infec- tious Disease Society of America states there is no measureable benefit to screen for asymptomatic bacteriuria for most adults

   N18 years of age, unless atypical symptoms or patient characteris- tics are present [1-4].

   UA with reflex to culture is a common laboratory order that was initially designed to reduce the number of urine cultures and anti- biotic treatment for patients with negative UA results [5-7]. How- ever, reflexing specimens to urine culture based on positive UA does not include indications for UTI in the reflex algorithm [8,9]. Urine cultures ordered without clinical indication can lead to inap- propriate Antimicrobial treatment [10-14]. It is recommended that EDs examine automatic reflexes to culture based on UA results [15].

   In an urban tertiary care hospital in Illinois, USA from January 2016-December 2016, 23,722 UA specimens from the ED were or- dered as a UA with reflex to culture. Of the 23,722 specimens, 8721 were cultured, 2701/8721 (31%) were positive, defined as bacteri- uria >=60,000 cfu per mL or positive for yeast, and 6020/8721 (69%) were negative, defined as no growth or non-diagnostic growth (b60,000 cfu per ml).

   This project was approved by the local institutional review board as non-human subjects research. Through cause mapping, the project team discovered that 1) the high number of negative urine cultures was linked to UA with reflex to culture being ordered without an indica- tion, 2) UA with reflex to culture was pre-checked in several ED order sets, and 3) reflex to culture thresholds were less sensitive compared to other institutions. Additionally, not all physicians were clear on guidelines of when it was appropriate to order a UA with reflex to cul- ture in the ED.

   Based on completion of an impact to effort matrix and failure modes and effects analysis, the project team decided to update the ED abdominal pain and psychiatric pre-checked order sets, which had the highest negative urine culture rate, from pre- checked UA with reflex to culture to pre-checked UA, with urine culture available but unchecked. Education on when to screen for and treat a UTI, based on Infectious Disease Society of America guidelines, was provided at Emergency Medicine department meetings to ED attending physicians, administration and leader- ship, and midlevel providers [1-4]. A hardcopy of the information was given to attendees and an email was sent to all ED providers. The reflex to culture threshold was changed from N3 WBC to N4 WBC. UA specimens positive for nitrite or LE continued to reflex to urine culture. By incorporating the changes into the process of ordering urine cultures, the team believed the results would be sustainable long-term.

   Data was analyzed 5 months pre- and post-intervention. All statistics were performed in Minitab Statistical Software Version

   18.1. A p-value of b0.05 was considered significant. The number of UA with reflex to culture orders per day was reduced from 92 or- ders per day to 49 orders per day (mean) post-intervention (t-test, p b 0.001) (Fig. 1A). The number of UA orders per day without re- flex was significantly increased from 3 orders per day to 41 orders per day (median) post-intervention (Mann-Whitney U Test, p b 0.001) (Fig. 1B). The number of negative urine cultures per day was significantly reduced from 13 cultures per day to 6 cultures per day (median), a 54% reduction in negative urine cultures post-intervention with an estimated cost savings of $71,350 per year based on direct costs of urine cultures (Mann-Whitney U Test, p b 0.001) (Fig. 2).

   ED patient volumes remained stable throughout the pre- and post- intervention time periods. There were no significant changes in the num- ber of 72-hour returns to the ED and 30-day readmissions with diagnosis of UTI, sepsis, bacteremia, or pyelonephritis post-intervention (Fig. 3A-B). Providing education on when it is appropriate to screen for and treat a UTI, updating pre-checked order sets, and increasing reflex to culture thresholds led to a decreased number of negative urine cultures, reducing the high cost of quality waste from inappropriate urine culture testing in the ED. There is substantial opportunity for decreasing costs and