|
Article Excerpt
While there is considerable evidence in the case of hospitals and surgeons that high volume is associated with better patient outcomes across a variety of medical conditions (Luft, Hunt, and Maerki 1987; Tu, Austin, and Chan 2001; Birkmeyer et al. 2002, 2003; Halm, Lee, and Chassin 2002; Elixhauser, Steiner, and Fraser 2003; Gandjour, Bannenberg, and Lauterbach 2003; Huckman and Pisano 2006; Marcin et al. 2007), to date no study has examined this relationship in a prehospital setting.
Emergency medical services (EMS) networks respond to, stabilize, and transport trauma patients involved in situations, such as automobile accidents, injuries from falls, and violence. EMS is a crucial part of the health care system and the public's emergency medical safety net. However, more than three decades after Congress passed the Emergency Medical Services Systems Act of 1973, trauma victims are treated by EMS with little or no evidence that the care they receive is optimal (Delbridge et al. 1998).
There is little research addressing the effectiveness of EMS practice (Pointer 2001) and there are even fewer studies examining the effect of experience accumulation, as proxied either by certification levels (Pollock, Brown, and Dunn 1997) or tenure (Soo et al. 1999) on performance. Our study is the first to use detailed data on experience accumulation in addition to certification levels and tenure to document the performance gains from higher volume.
Learning by doing is one channel through which skills are acquired and therefore provides a potential rationale for the observed positive correlation between volume and outcome. Under this mechanism, more acquired experience improves performance. Nevertheless, the causality may run in the opposite direction, through a selective referrals channel, whereby better performers command greater demand for their services (Luft, Hunt, and Maerki 1987).When using retrospective data to study volume-outcome relations for surgeons, the selective referrals problem is difficult to overcome credibly. Unlike settings which involve the choice of provider (e.g., elective surgery), the unpredictable nature of EMS does not lend itself to selective referrals. Ambulance units are dispatched based on proximity and not on reputation, and trauma victims are not offered the choice of emergency care providers on scene.
In this paper, we focus on trauma-related ambulance runs to study the effect of volume and experience accumulation on performance. The evidence regarding the appropriateness of prehospital interventions on trauma victims is mixed. However, the importance of getting the patient to definitive care as soon as possible, allowing only for the performance of essential procedures, is widely accepted in EMS and prehospital time markers are frequently used process measures. Shorter out-of-hospital time intervals represent a potentially important factor in survival, especially in the case of trauma incidents (Sampalis et al. 1993; Feero et al. 1995; Nichol et al. 1996; IOM 2000; Blackwell and Kaufman 2002; Carr et al. 2006). For trauma patients, the first hour of care (also referred to as the "golden hour") is usually considered critical with some evidence that when the time from the incident to hospital treatment is within this critical first hour, the patient's likelihood of survival is increased (Rawlinson and Crews 2003). In trauma care, the goal is to get seriously injured patients into the operating room of a trauma center with an experienced team of appropriately specialized trauma surgeons as fast as possible. Therefore, it is not surprising that dispatching emphasizes ambulance proximity and EMS contracts often include response standards.
We hypothesize that paramedic learning will lead to shorter out-of-hospital times. In particular, as paramedics become more proficient in diagnosing patients' acuity, identifying the appropriate procedures, mastering protocols and techniques, gaining familiarity with local roads and traffic conditions, and identifying quicker routes to scene and to the hospital, the shorter is the time spent out of hospital, and, in particular, the time spent at the scene.
Using the universe of trauma-related ambulance runs in Mississippi between 1991 and 2005, we find that greater volume is robustly related to improved paramedic performance. In specifications that include both individual- and firm-level measures of recent volume, we find small productivity spillovers. In addition, we find that the benefits of recent volume accrue differentially across tenure groups, favoring paramedics with above median tenure.
Data limitations have led most researchers to focus on the effect of recent experience (e.g., volume in the past quarter or year) on performance. (1) This is appropriate only if past experience rapidly loses its relevance. Alternatively, if past experience matters, high turnover rates may affect overall prehospital performance as paramedic human capital accumulation is subject to more frequent interruptions. The richness of our data allows us to follow 1,728 paramedics (85 percent of all paramedics in our data) for whom we observe their full histories in the profession. We construct measures of their tenure and cumulative experience as well as of recently acquired experience. Hence, by evaluating the contribution of cumulative experience, we assess the value of paramedic retention in performance units.
The benefit of greater experience operates both through recent and past experiences. This has important implications for retention: retaining the typical paramedic in our sample for an additional quarter is associated with a reduction in out-of-hospital time of approximately 4 minutes. Moreover, 3 years after replacement of an experienced paramedic by a new one, the loss in out-of-hospital time from turnover is between 1 minute and 90 seconds depending on functional form.
Finally, we find evidence that individual experience operates not only on mean performance but also on the upper quantiles of the paramedic performance distribution, which can be interpreted as greater experience leading to better standardization with the severity of trauma.
BACKGROUND
The EMS Workforce
The three national standard levels of training for Emergency Medical Technicians (EMT) are EMT-Basic (EMT-B), EMT-Intermediate (EMT-I), and EMT-Paramedic (EMT-P). (2) In addition, Mississippi requires operators of ambulance vehicles to be EMS-Driver certified (EMS-D), by participating in a training program in operation of emergency vehicles. Licenses for EMTs must be renewed, typically every 2-3 years.
A combination of factors renders the hiring and retention of EMTs problematic, including concerns regarding personal safety, stressful working conditions, irregular hours, excessive training and requirements, limited mobility, and low wages (Institute of Medicine 2000).
Paramedics face a difficult, often hazardous, work environment. They are exposed to potentially infectious bodily fluids, for instance through contact with contaminated needlesticks, and to the hepatitis B virus (Delbridge et al. 1998). Moreover, they are frequently exposed to the threat of violence, incur injuries associated with lifting or falling, and face oncoming traffic at the scene of motor vehicle crashes (Franks, Kocher, and Chapman 2004). Occupational fatality rates for EMTs are comparable to those of police and fire personnel (Maguire et al. 2002). (3)
In 2001, the Journal of Emergency Medical Services conducted a salary survey consisting of 371 EMS organizations. The survey found salaries to be about 10 percent lower in the Southeastern Census region, which includes Mississippi, compared with the national average. Tenure for paramedics was 7.3 years in private ambulance companies compared with 10.8 years in fire-based EMS agencies. Finally the average yearly turnover rate for private ambulance companies was 18 percent, as compared with 6 percent for fire-based agencies.
EMS Performance
Some studies have found evidence that prehospital advanced life support (ALS) interventions reduce the severity of injuries, lower mortality rates in the first 24 hours, improve outcomes from multiple trauma, and improve survival among patients with blunt trauma (Isenberg and Bissell 2005). Nichol et al. (1996) perform a meta-analysis and find shorter out-of-hospital time to be related to lower mortality. However, overall, the medical literature provides mixed evidence regarding the effectiveness of out-of-hospital management of trauma victims (Carr et al. 2006; Delbridge and March 2007).
Rural trauma patients, on average, face considerably lengthier transport to definitive care (Isenberg and Bissell 2005). Studies focusing on rural areas have found...
|
