Rural Trauma
THE RURAL ENVIRONMENT
While the majority of the population of the United States lives in an urban environment, 70% of the trauma deaths occur in a rural locale. “It is surprising that a disease that kills rural citizens at nearly twice the rate of urban citizens has not received more attention.”1,2 The chance of dying in a rural area from a severe injury sustained in a motor vehicle–pedestrian collision is three to four times greater than in urban areas.3 The relative risk of a rural victim dying in a motor vehicle crash is 15:1 compared with a victim of an urban crash,4 and death from motor vehicle crashes is inversely related to population density.5 In fact, death rates from all unintentional injuries combined are generally 50% greater in rural, sparsely populated counties of the western United States than they are in the densely populated northeastern counties.6,7 And pediatric deaths from injury in a rural setting are more frequent than they are in an urban setting, despite the recent increase in gunshot wounds in the urban population.8,9 Finally, autopsy studies have suggested preventable trauma death rates of 20–30% in rural populations.10–13
Not only are mortality rates higher, but outcomes in survivors based on Functional Independence Measure (FIM) scores are also worse. When fatalities are excluded, the rural to urban odds ratio of poor outcome is 1.52.14 Poor functional outcomes have also been documented in patients with traumatic brain injury sustained in rural versus urban locales. What are the reasons for these differences?15
In this chapter we will attempt to identify circumstances that make rural trauma care difficult and consider some solutions. An illustrative case will help to explain some of the unique features of trauma care outside an urban setting. A 48-year-old real estate developer was mountain biking with friends in a national forest in the Rocky Mountains. While unhelmeted, he rode ahead of the group and down a steep slope. Several minutes later his companions found him unconscious at the bottom of a ravine after he had apparently lost control of his mountain bike. One of the friends rode out for help, which arrived 45 minutes following the crash in the form of a basic life support (BLS) ambulance unit from the local ski area. The patient had to be extricated from a ravine and carried several hundred yards to the ambulance, which then had a 1-hour trip to the nearest hospital, a Level III trauma center. Communication (handheld radio) with the hospital was not possible until the ambulance exited a narrow mountain canyon about 15 minutes before arrival. His Glasgow Coma Scale (GCS) score on the scene and in the emergency department was 8. He was hemodynamically normal, but a computed tomography (CT) scan of the head showed a large epidural hematoma with >5-mm shift. No other injuries were identified. Following consultation with a neurosurgeon at the nearest Level II center (150 air miles away), a general surgeon trained in emergency limited craniotomy (and following established local protocols) drilled a burr hole and enlarged it sufficiently to permit evacuation of the clot and to control the hemorrhage. The patient was transferred directly from the operating room to a helicopter, which flew him to the neurosurgeon for a formal craniotomy. He survived with a Glasgow Outcome Score of 4 and is now independent, although no longer able to function in his former capacity.
This true scenario could have any of the following plausible variables: unaccompanied victim hours or days to discovery, less accessible to rescuers, greater distance to hospital, lack of trauma team and trained surgeon, or adverse weather preventing air transport to Level II trauma center. He might have been a hunter injured by firearm or animal, a backcountry skier caught in an avalanche, a rancher thrown from a horse, or the driver of a car on a remote rural road.
Remoteness, rugged beauty, and “nature” are powerful magnets for tourists, recreationists, and those seeking a quieter, less stressful lifestyle. Such visitors are often shocked to discover that medical services they take for granted at home are simply unavailable in a rural setting. In contrast, local residents tend to be independent, fatalistic and accepting of limitations, suspicious of outsiders, resistant to both change and regulations (helmet and seatbelt laws; gun control), and unaware of trauma as a public health problem because, in their limited experience, it is a rare event.1,16
DEFINING RURAL
Rural spaces are sparsely populated and support only one third as many physicians as do urban areas.1 Specialists and, sometimes, even primary care physicians cannot make a living in communities under a certain population threshold. Lack of physicians skilled in rapid assessment and treatment of the critically injured has a significant effect on outcomes. Conversely, the number of physicians in a given county, and particularly emergency physicians who have taken the Advanced Trauma Life Support (ATLS) course, is associated with lower death rates from trauma.17
An urban area consists of a central city and its environs with a combined population of greater than 50,000 and a population density of 1,000 or more per square mile. As far as the US Bureau of the Census is concerned, everything else is rural. Not all rural environments, however, support farming, and many are surprisingly close to cities, but separated by geographic barriers such as mountains or large bodies of water. Residents of coastal Marin County, only a few miles north of San Francisco, may have difficulty accessing a trauma center because of intervening steep mountains, narrow roads, and local regulations prohibiting noisy and disruptive helicopters. San Diego County, home of a model trauma system, reports morbidity and mortality from delayed discovery of victims of motor vehicle crashes on remote county roads.18 North Carolina is the 10th most populous state, but 29 counties in the eastern part of the state are served by a single trauma center, 12 counties have no general surgeon, 17 no orthopedic surgeon, and 23 no neurosurgeon. In only 14 of these counties can a victim of trauma reach an emergency department staffed with emergency physicians in less than 30 minutes.17
Hypothermia is an independent predictor of mortality in injured patients, with mortality rising as core temperature falls. Urban trauma centers, with more rapid discovery and transport times than rural prehospital systems, have reported a 5% incidence of trauma patients with low core temperatures.19 A Level I trauma center in North Carolina serving 29 mostly rural counties and a combined population of 1.4 million reported transport times averaging 1 hour and time from injury to definitive care of 4 hours. Their trauma registry identified 1,490 of 9,482 (16%) patients suffering from hypothermia (<36°C) on arrival. These hypothermic patients had a 14.6% mortality rate, compared with 4.5% among normothermic patients.20 Virtually all regions of our country have many areas that are sparsely populated and relatively poor in resources. In the central and northern plains, one sees the combination of few people and great distances from urban centers as nowhere else in the lower 48 states. Alaska, of course, along with some portions of the northern Rocky Mountains, is more accurately described as a frontier area (six or fewer people per square mile).
The state of Western Australia forms the western third of the country, with an area of 2.5 million km2, and a population density of 0.8 people/km2. Because of a lack of doctors and hospitals, some trauma patients may be transported in excess of 2,000 km over more than 24 hours to receive initial care. Investigators there developed an Accessibility/Remoteness Index of Australia (ARIA+) to reflect the ease or difficulty people face accessing services in nonmetropolitan areas of the country. The index is a continuous variable with values ranging from 0 (high accessibility) to 15 (high remoteness). Employing ARIA+, with Perth (the most isolated capital city in the world) as the reference point, they demonstrated that the remote rural trauma death rate is over four times the rate in major cities.21
Rural, then, may be defined in accordance with census data based on metropolitan statistical areas, in terms of geography and distance, or by virtue of limited resources. In a recent analysis of the general surgery workforce, Thompson et al.22 identified significant differences between communities with a population between 10,000 and 50,000 (large rural) and those with 2,500–10,000 residents (small, or isolated rural). Large rural towns are far more likely to have the necessary resources such as general surgery, medical and surgical subspecialties, advanced life support (ALS) ambulance services, and essential equipment to provide prompt and sophisticated trauma care.
Environmental factors are also important as “rural trauma occurs in areas where geography, population density, weather, distance, or availability of professional and institutional resources combine to isolate the patient in an environment where access to definitive care is limited.”23 An alternate, somewhat more precise definition has been proposed as follows: “… A rural trauma region would be an area in which the population served is fewer than 2500, has a population density of fewer than 50 persons per square mile, has only basic life support prehospital care, has prehospital transport times that exceed 30 minutes on average, and is lacking in subspecialty coverage for specific injuries (such as a neurosurgeon to manage the patient with head injuries).”1 In any event, though we may think we know it when we see it, it is apparent that “rural” is difficult to define.
EPIDEMIOLOGY
There are reasons why few people live in rural locations. The climate may be harsh, the terrain rugged and remote from services, the roads badly engineered and maintained, communications rudimentary, and the economy marginal. Career opportunities for the young are limited, so the young leave. As a result, significant segments of the population are elderly, poor, poorly educated, and in ill health. Population density (low) and personal income (also low) are the strongest predictors of per capita trauma death rates.17 Nearly one fourth of adults in this environment sustain some form of unintentional injury per year. The injuries are usually relatively minor, but they can be major and/or fatal. Binge drinking and depression are strongly associated comorbid factors, and suicide accounts for 10% of all rural trauma deaths.1,24 Elderly rural patients tend to start out with a lower Injury Severity Score (ISS) and are less likely to die at the scene, but have a higher complication rate and worse overall survival for comparable severity of injury. Based on data from the Major Trauma Outcome Study (MTOS), rural geriatric trauma patients fare less well than do those in an urban cohort.25 In addition, older age and lower population density independently increase vehicle-related mortality.26 A study by Wigglesworth27 compared two groups of five states each with the highest (Group 1) and lowest (Group 2) traffic death rates, respectively. Epidemiologic data from the Centers for Disease Control and Prevention (CDC) indicated that the fatality rates for falls, poisoning, drowning, fire, suffocation, homicide, and suicide conformed closely to the traffic death rates in the two groups of states. Group 1 states were rural, western, and below national averages for per capita income; Group 2 states were urban, eastern, and financially well-off.27 Overall, 60–70% of all trauma deaths occur in rural areas despite the fact that only 20–30% of the nation’s population lives in these areas.28
Unintentional blunt injury comprises about 90% of cases, largely because of the prevalence of motor vehicle crashes and the paucity of injuries from firearms. The most common causes of fatal injury are motor vehicle crashes, suicide, homicide, and falls. For these and the next 10 most frequent causes, rural death rates exceed urban rates for all but poisonings. Some of the most hazardous occupations such as mining, logging, and farming are almost exclusively rural by their very nature.29 Large animal injuries may occur on a farm or ranch in the course of daily work or in conjunction with such recreational pursuits as hunting, pleasure riding, or rodeo. Typical injuries are falls (horses), tramplings and gorings (bulls, wild game), and kicks (cows).30 Travel on rural highways entails the additional hazard of motor vehicle crashes with wild animals (elk, deer, bear, or moose). As a mature moose weighs half a tonne or more, a driver unfortunate enough to strike one risks significant injury to the brain or death.31 Fatal accidents are significantly higher for loggers (140/100,000) than they are for workers in other industries (94/100,000), and are typically the result of being struck by falling trees, limbs, or snags. Crush injury between moving logs and encounters with heavy equipment are other common mechanisms, and access to care is often a problem.32 Recreation also provides endless opportunities for serious injury and death. Particularly dangerous are four-wheeled all-terrain vehicles (ATVs). In 2004, ATV crashes were shown to result in more than 136,000 injuries and 500 deaths and one third of the deaths were in children. In Oregon, the rate of such injuries and deaths doubled between 2002 and 2005.33 Small community hospitals bear the brunt of these misadventures, particularly when situated in proximity to ski hills, wilderness areas, national parks, and seashores or lakefronts.
Most people feel safe in rural setting as the risk of violent assault and penetrating trauma is very low as noted above. The low homicide rates, however, are negated by high suicide rates, particularly among adolescents and young adults.34–36 Blunt trauma comprises 95% or more of the trauma case load at most rural community hospitals, 85% of which is minor or moderate (ISS < 10) and can be treated without the need for transfer to a trauma center. Blunt trauma is less time dependent, which is fortunate since it is far more difficult to mount a rapid response in a small community hospital. It can be subtle, however, requiring experience and a high index of suspicion to avoid missed injuries.37 Although motor vehicle crashes cause the greatest number of trauma deaths in this country and in the rest of the world, they are sporadic events in small towns and the countryside.
In essence, the greatest problems confronting rural trauma care are access to the system and lack of resources. The challenge is to devise a system, ensure access, and make the most of limited resources.
RURAL RESOURCES AND LIMITATIONS
Discovery and Access to the System
Discovery of the victim and access to appropriate care are the most important explanations for the high mortality rates of trauma victims in rural areas. When people are scarce and distances between population centers are great, the injured may be lost or misplaced, whether in the backcountry or on a remote highway.1,16,38–40 Delays of hours are common, and, occasionally, days may pass before a victim can be found. In rural systems of care, time of crash until time of arrival at the hospital is more than an hour in 30% of cases, as opposed to 7% in urban systems.41 Prolonged mean prehospital times have been reported in rural Vermont (105 minutes), upstate New York (96 minutes), northern California (55 minutes), rural Washington (48 minutes), and Georgia (40 minutes). Thus, the “golden hour” is often spent on the road and not in the hospital.1 In extreme cases, crash victims in a snow-filled roadside ditch or ravine, a hunter, or a backcountry Nordic skier may not be found until spring breakup. Retrieval is equally challenging and often relies on the special skills of search-and-rescue volunteers equipped to go into swamps and tidal flats, high mountains, or dense forests and other wilderness areas. Even when a helicopter is at hand, victims must often be moved over rough terrain by litter, watercraft, snowmobile, ATV, horse, or other conveyance to a suitable and safe landing area. Fortunately, most guides and outfitters now carry global positioning satellite (GPS) units, cellular phones, and/or handheld radios to facilitate rescues in emergency situations.
If a hospital lacks a trauma program and leader, the response to a major trauma event tends to be disorganized as the patient will often arrive unannounced. Obvious extremity injuries may overshadow more critical internal injuries and prompt a call for an orthopedic surgeon when a trauma team led by a general surgeon is more appropriate. Even when notification occurs, the physician in the emergency department may wait to see just how badly a patient is injured, instead of mobilizing the trauma team and alerting the helicopter for interhospital transfer. The opportunity to eliminate a critical rate-limiting step is then lost. The patient proceeds from scene to litter to ambulance to local hospital and then, perhaps, on to the next higher level of care sequentially, and precious time is wasted (see Fig. 9-1).
FIGURE 9-1 Roadside crosses on rural Montana highways mark fatal crash sites. (Reproduced with permission of the Critical Illness and Trauma Foundation, Bozeman, MT.)
One of the most important steps a small hospital can take toward improving trauma care is the establishment of a trauma team. When possible, the team should be led by a general surgeon.42–44 Criteria for team activation should be established, and team members should commit to come to the patient’s bedside immediately when called. In very small hospitals lacking general surgery support, it is still possible to provide appropriate emergency care.45 The Rural Trauma Subcommittee of the American College of Surgeons Committee on Trauma conducted an informal survey of small rural hospitals and found that most could mobilize three health care providers most of the time including physician extenders, nurses, and technicians (lab, x-ray, respiratory therapy). Drawing on these resources (as well as primary care physicians, or surgeons, in slightly larger facilities) the committee’s Rural Trauma Team Development Course trains these individuals in the team approach to the initial assessment, resuscitation, and transfer to definitive care for the injured patient. This 1-day interactive course is patterned on ATLS, but is inexpensive to present and may be given in modular form. The program is coordinated through the state chair of the Committee on Trauma.46 Obstacles to this logical solution include the following: (1) medical staff reluctance to participate in trauma care; (2) fears that overtriage will place greater demands on surgeons; (3) turf wars between ambulance services, hospitals, and communities; and (4) financial incentives to treat patients locally rather than transport them. In addition, the Emergency Medical Treatment and Active Labor Act (EMTALA) may have the unintended consequence of discouraging efficient transfers within a trauma system.
MANPOWER
Physicians who practice in rural areas likely grew up in a small town, were influenced by a mentor, or have an independent streak.47 They usually have no fears of being overworked, underpaid, and unable to obtain backup or guidance. If the social and cultural deprivations bear heavily on a spouse, the sojourn will be brief. Patient volumes are insufficient to support specialty services, and generalists with little or no formal trauma training predominate. Such individuals are expected to treat a significant number of minor trauma cases as well as the occasional complex major trauma case for which they have not been trained. A study of five small hospitals in rural Washington and Idaho revealed that they averaged three patients per year with an ISS greater than 19, and each physician saw, on average, 0.6% of these patients.48 Because trauma events are sporadic and infrequent, rural physicians may develop a fear of or aversion to care of the injured. Furthermore, despite evidence to the contrary, general surgeons believe they are more likely to be sued by trauma patients.49
Rural surgeons, more so than their urban counterparts, have traditionally viewed trauma care as an integral part of their service to their communities.50 Perhaps the greatest threat to our future ability to treat the injured in rural America is the unwelcome fact that this sense of commitment is changing. Rural surgeons are, for the most part, general surgeons. And this specialty has lost its appeal among young trainees who increasingly (70–80%) choose a subspecialty and avoid embarking on rural practice. Currently, the vast majority of rural surgeons are men, while more than half of medical school graduates are women. Women now account for 25% of surgical residents, and surveys confirm that they are much more inclined to practice in an urban or suburban setting.51,52 Furthermore, declining reimbursements compounded by a heavy burden of debt accrued during medical school have led many general surgeons to insist on compensation for call. Small rural hospitals may not be able to afford underwriting emergency call. Finally, the current cadre of rural general surgeons is retiring because of advancing age or because of burnout. As they are not being replaced, the net effect is that access to definitive trauma care at the local level is rapidly disappearing. This places a greater burden on regional Level I and II trauma centers, while at the same time denying essential revenue to the small community hospital. Perhaps most distressing, loss of surgical services may lead to the closing of the only hospital in a large geographic area.53
In many communities, a nurse practitioner or autonomous physician’s assistant is the town “doctor.” Their educational background and experience, as well as that of their supervising physician, rarely includes exposure to sufficient cases of major injury. Ambulance workers are predominantly volunteers trained in advanced first aid, emergency medical technician (EMT)-Basic, or, at most, EMT-I level, usually at their own expense. Their experience with trauma is also very limited, though trauma may be one third to one half of the case load in some rural ambulance services.54
Outside the academic medical center it is uncommon for resident staff or independent practitioners to stay in-house after hours. In hospitals with less than 30 inpatient beds it is unusual to have emergency physicians. In many small communities, a nurse or physician’s assistant is the only professional at the hospital on nights and weekends. The doctor may be at home or out of town and may or may not be willing to come in if called for a trauma emergency. If the doctor does come in, he or she will conduct the resuscitation with minimal assistance and limited equipment. Despite these shortcomings, there usually is no other choice since the next hospital may be many miles distant and may be no better equipped. It is for reasons such as these that it is so important for rural physicians, physician assistants, and nurses to take or audit the ATLS course and to become involved in the regional trauma system.
An effective emergency medical service (EMS) program is vital for proper trauma care. In rural areas, the configuration of such systems varies and may include fire department–based, hospital-based, or freestanding entities. Personnel may be volunteers, salaried, or partially subsidized. Most are trained to the EMT-Basic level, which permits noninvasive interventions to reduce the morbidity and mortality associated with acute, out-of-hospital medical and trauma emergencies.55 Skills and capabilities may be enhanced with the addition of certain modules, under the guidance of their medical director. Some rural communities have personnel trained at higher levels of care.56 Specific trauma training (i.e., Prehospital Trauma Life Support [PHTLS]) may be challenging to conduct in rural areas for lack of instructors, but should be supported and encouraged. The nomenclature for the various levels of training is in transition, which may be clarified by the publication of a new scope of practice document in 2007.57 Currently, the primary challenges to rural EMS are maintenance of skills in a low-volume environment and dealing with collapse of infrastructure as a result of an aging volunteer workforce that is not being replaced. One proposed solution is to upgrade volunteer EMT-Bs to paramedics, employ their new skills as an adjunct to a broader community health program, and pay them.58,59
Aeromedical and ground transport systems that furnish critical care are becoming more common; however, their availability lags behind in many rural areas. In some locations direct scene responses may be available while in others rendezvous with such units is more practical. Thoughtful incorporation of all resources into a regionalized response system for time-sensitive, life-threatening conditions (high-risk obstetrics, stroke, and STEMI as well as trauma) is beginning to evolve. Surgical leadership into the evolution of such systems is essential to ensuring the needs of the injured patient are not overshadowed by other acute conditions.
COMMUNICATIONS
The original EMS legislation and subsequent funding bills recognized the need for effective and reliable communication between field and hospital. Availability of funds to improve communications infrastructure following the 9/11 bombings has improved radio coverage in many metropolitan areas. Paradoxically, those same systems have in some instances resulted in poorer rather than better coverage in rural areas due to terrain and distance issues. Skilled dispatchers are hard to find in small towns. Physicians and nurses at the hospital may be unfamiliar with and wary of communications equipment, and, accordingly, reluctant to talk with field personnel to provide medical guidance. Cellular telephones have improved prehospital provider-to-physician dialogue in many areas. In some areas with appropriate infrastructure and networks, telemedicine technology permits audio, video, and data transmission from field to hospital. Many 9-1-1 systems have upgraded to E9-1-1 (associates caller’s telephone number with a physical address). Even as rural areas are beginning to catch up in E9-1-1 availability, next generation NG9-1-1 is beginning to be deployed. NG9-1-1 is a network of systems that enables the transmission of voice, data, video, and text from various types of communication devices to a public service answering point. It makes that information actionable so that it can be moved into interconnected emergency responder networks. With the explosion of cell phone use that is not attached to a specific address, the FCC is mandating that cell phone services develop the ability to provide the latitude and longitude coordinates of the calling handset, accurate to within 50–300 m, to any Public Safety Answering Point (PSAP). Ultimately, cell phones will be able to transmit images and data.
Communication between the local hospital and regional trauma center may also be difficult and unrewarding. Local practitioners often complain of unpleasant encounters with flight crews, emergency room personnel at the receiving hospital, and surgical staff on the trauma service. Attending surgeons are infrequently available for telephone consultation, despite the fact that they are, in essence, being referred a patient by a colleague. Feedback and constructive criticism regarding transferred patients are frequently sought by referring physicians, but not often attainable. In addition, they may receive mixed messages including criticism for overtriage on the one hand to holding onto a patient too long on the other.
The net effect of problems in communication is that the rural practitioner ends up functioning in a relative vacuum, receiving little advance warning from the field, limited help at the hospital, and negative or no feedback from the regional center.
TRANSPORT
Evacuation of rural trauma victims is generally accomplished by surface conveyances. If the victim is inaccessible to an ambulance, various methods, all of them slow, may be employed to convey the patient to a road. The ambulance may then need to negotiate a sequence of roads from unsurfaced or gravel to county or state highway. Even the latter may be narrow, winding, and poorly maintained. Most often the destination is the nearest hospital, which will vary in its capabilities, and may be many miles distant. Response times, which include travel from the dispatch site to scene, extrication or retrieval, packaging, and travel to the hospital, are sometimes measured in hours, not minutes, as previously noted.
Ambulance services may be freestanding or, in some instances, an integral part of the local fire department. Funding may be through a special ambulance district or as part of the county budget, jealously guarded by county commissioners.37 Frequently, because of limited funding, the ambulance service may employ aging although lovingly maintained vehicles, which are limited in number. In Vermont, it is estimated that the average local ambulance is unavailable 15% of the time.1 Surveys of state EMS directors in 2000 and 2004 indicated that the greatest need for rural services is the adequate recruitment and retention of staff. In the same surveys, 24/7 coverage rose from the 22nd to the 2nd most important rural EMS issue. Response time rose from 20th to 5th. If an ambulance is in service on a call or out of service for maintenance, the next call might have to be answered by a crew in a neighboring district through a mutual aid agreement.60 Multiple incidents or victims can easily overwhelm the transport system.
Helicopters can be used both for scene rescue and for interhospital transport. Ideally, evacuation from the scene of injury directly to the trauma center should afford the patient the best opportunity for recovery. Due to reimbursement changes, there has been a proliferation and associated overutilization of helicopter services in some areas. In the urban environment, ground ALS has actually been shown to be preferable for relatively short distances, since it takes time to prepare the aircraft for flight. With flight times above 15 minutes, helicopters gain the advantage.61 In Fresno County, California, a study of ground versus helicopter transport in a relatively flat, nonmountainous area served by one Level I trauma center concluded that, within 10 miles of the hospital, ground transport yielded the shortest 9-1-1–hospital interval. Beyond that distance, the simultaneous dispatch of ground and air transport was the most efficient as ground personnel could extricate and resuscitate in advance of the arrival of the helicopter. For surface transports of more than 45 miles, helicopter was faster even if dispatched after the ground unit.62
In the rural environment, provided the scene is within the range of aircraft without the need to refuel, direct transport may be worthwhile if the time to the local hospital by ground ambulance is greater than that of the helicopter flight.63 If not, surface transport is preferable.64 A helicopter may also be invaluable in wilderness rescue if a suitable landing site can be assured. The downside is that such aircraft are expensive ($900,000–2,200,000 start-up; $500,000–2,000,000 annual maintenance), hazardous (fatal accident rate 4.7/100,000 hours),65 and have a limited range. They are also sensitive to weather and altitude and are not always available. Although newer models are roomier, it is still difficult to examine, monitor, and resuscitate unstable patients while airborne. Finally, their effectiveness is open to question. In one study of scene (18.8%) and interhospital (79.5%) transports, the most severely injured patients (17%) died en route or shortly after arrival at the medical center, while 55% had relatively trivial injuries that did not require the use of the rotorcraft.66