A 31-year-old gravida 1, para 1 woman presented at 40 weeks in the early stages of labor having received limited prenatal care at an outside clinic. Physical exam performed by the obstetrics resident was suggestive of placenta previa and an anesthesiologist was called to prepare for a cesarean section (C-section). The anesthesiologists were short-staffed and also covering the operating rooms on a different floor, but felt they could be available if needed emergently. Shortly afterward, fetal heart rate monitoring suggested fetal distress and the patient was transferred to the operating room for emergent C-section. The anesthesiologists were called again but were unavailable due to another operative emergency. After significant delay, the patient was ultimately anesthetized and underwent C-section. Unfortunately, the baby was delivered with profound neurologic abnormalities, including quadriplegia and cortical blindness.
What went wrong? While lack of an available anesthesiologist to provide urgently needed anesthetic seems the obvious conclusion, there are other factors that, had they been considered, could have helped avoid the catastrophic outcome.
When allocating health care resources, it is important to balance the goal of providing safe, efficient, effective, and timely care with available funding. This simple principle also applies to medical personnel staffing. A rational protocol for staffing is required to allocate medical personnel for delivery of optimal routine and emergency medical care. Mathematical modeling, such as analysis of Poisson distributions of events, has proved helpful in defining optimal manpower needs. For example, these techniques have been employed to assess neurosurgeon and orthopedic surgeon staffing needs for trauma centers.(1,2) Similar methods have been used to determine optimal number of beds and occupancy to minimize staffing costs in an obstetrical unit (3), as well as the minimum number of operating room nurses that should be on call for urgent procedures. After inputting manpower availability and risk tolerance, these models enumerate all possible shift combinations and identify those with the lowest cost that ensure the pre-specified level of service.
Although one might strive to ensure availability of adequate personnel in all circumstances, this level of preparedness would involve unreasonable costs in part by leading to substantial overstaffing during times of average census. Mathematical analyses cannot make the crucial decision about what level of risk should be tolerated. Is it acceptable to make manpower decisions that might lead to understaffing during two simultaneous emergencies once in 5 years? How about once in 10 years? Who establishes this level of accepted risk: physicians, insurance companies, legislators, or the courts? There are no easy answers. As in this case, however, it is inevitable that at times volume will exceed capacity and caregivers must be prepared.
The Joint Commission on Accreditation of Healthcare Organizations (JCAHO) requires staffing assessment using specific indicators such as patient complexity based on clinical service and human resource outcomes. For example, JCAHO requires that emergency departments have sufficient numbers of qualified staff to care for patients, including direct and indirect caregivers, and nurses must have the required skills to care for patients to whom they are assigned. If data such as the frequency of medication errors, incidence of wound infections, or patient satisfaction surveys reveal inadequate staffing, appropriate action must be undertaken for correction. Outcomes research is beginning to address these critical questions, and it is driving policy decisions. For example, a recent study demonstrated higher mortality rates with lower nurse-to-patient ratios (4), a finding that has led California to mandate certain minimum ratios.
In trying to ensure the availability of qualified staff at all times, larger hospitals (or hospital units, such as obstetrics) have an advantage over smaller ones. For a given risk, larger units have the ability to staff more efficiently (particularly if workers on other units are "cross trained" to cover a variety of tasks when needs arise); whereas, smaller units in remote settings are at increased risk of not being able to respond to large upticks in volume or simultaneous emergencies.
In the present case, we don’t know whether staffing analyses were ever considered, nor whether this facility commonly experienced simultaneous emergencies. Certainly, if the events presented in this case were not a rare exception, more anesthesia provider personnel should have been available.
Issues of anesthesiologist availability are complicated by a current and projected shortfall of anesthesiologists.(5-8) During the second half of the 1990s, the number of residents graduating from anesthesiology training programs decreased substantially, resulting in a workforce that has grown little in recent years. Despite an upturn in graduating anesthesia residents, there remains a projected shortfall. The supply of certified registered nurse anesthetists has also decreased slightly and is likely to decrease further.
For many hospitals, anesthesiologists are not available to staff labor and delivery separate from those available to staff the operating rooms.(9) This has been problematic in the US, as well as other countries.(10,11) In a comprehensive survey of hospitals in the State of Ohio, a complete surgical team was immediately available in 100% of Level III hospitals (in which a separate anesthesiologist is mandatory), but only 35% of Level I hospitals had an anesthesiologist immediately available.(12) This has obvious implications for the provision of a safe environment to manage potential obstetric emergencies, such as women who wish to labor after previous C-section, with the associated increased risk of uterine rupture. In fact, the American College of Obstetrics and Gynecology (ACOG) and JCAHO now mandate "immediate availability" of an anesthesiologist (and obstetrician) at hospitals where women labor after previous cesarean section. Although "immediate availability" was left undefined as to a specific amount of time or the physician’s exact whereabouts, it is generally understood to imply the anesthesiologist must be in the hospital. In smaller hospitals, changes in hospital staffing or referral patterns will be necessary to accommodate these requirements.
Preparation for emergency care involves more than establishing the optimal number of providers. It requires training, organization, and adept communication among health care workers. The use of high-fidelity patient simulators for training health care professionals has increased in recent years. Simulators have been widely adapted for use in anesthesia crisis resource management, and these methods are rapidly spreading to other medical disciplines.(13) For example, teaching modules for crisis management recently have been established for radiologists. Participants manage clinical scenarios on lifelike computer-controlled manikins within realistic clinical environments. A large component of simulator-based training is experiential and involves learning how to communicate clearly and directly, with the goals of improving situational management and team effectiveness skills. This sort of multidisciplinary team training may not prevent understaffing at a moment of crisis, but can give providers a set of tools to make their responses more rational, productive, and ultimately safer.
In this case, what was told to the anesthesiologist by the obstetrician? Was the situation communicated as unstable, urgent, or a problem that 'might' happen? Did the anesthesiologist convey how likely or not someone could be available emergently? Were contingency plans considered? The anesthesiologist conceivably could have called in another anesthesiologist given the pending potential to be acutely understaffed. In the setting of no serious vaginal bleeding and the absence of a worrisome ("non-reassuring") fetal heart rate tracing, the obstetrician could have temporized by tocolytic administration and volume replacement, pending availability of an anesthesiologist.
Extraordinary circumstances often require uncommon responses, including employment of skills and knowledge rarely used. Given the situation of 'fetal distress' (non-reassuring fetal heart rate pattern) suggestive of developing fetal asphyxia, proceeding with emergent cesarean section using local, infiltration anesthesia by the obstetrician was indicated. Regretfully, local anesthesia techniques for cesarean section are rarely taught to obstetrical residents; in fact, most obstetricians have never seen one performed. This seems most unfortunate in a world where an anesthesia provider cannot always be immediately available for all obstetric emergencies.
- Mathematical modeling can facilitate careful planning for medical staffing requirements, which should include defined and rational protocols for primary and back-up responders, particularly for emergency care. Optimal care must be balanced with the reality of limited resources.
- Preparation for emergency care requires training, organization, and proficient communication. Expert and professional interpersonal and communication skills are essential for the effective exchange of information required to provide optimum patient care and for the most advantageous collaboration among health care workers. Simulators and teamwork training can facilitate such learning.
- Extraordinary circumstances often require skills and knowledge rarely used. However, they must be learned and may benefit from being rehearsed (much like annual CPR training) for those rare times when they may become essential.
Mark A. Rosen, MD Professor and Vice Chairman of Anesthesia Professor of Obstetrics, Gynecology, and Reproductive Sciences Director of Obstetrical Anesthesia Director of the Anesthesia Residency Training Program University of California, San Francisco
1. Lucas CE, Dombi GW, Crilly RJ, Ledgerwood AM, Pingyang U, Vlahos A. Neurosurgical trauma call: use of a mathematical simulation program to define manpower needs. J Trauma. 1997;42:818-24.[ go to PubMed ]
2. Lucas CE, Middleton JD, Coscia RL, et al. Simulation program for optimal orthopedic call: a modeling system for orthopedic surgical trauma call. J Trauma. 1998;44:687-90.[ go to PubMed ]
3. Dexter F, Marcario A. Optimal number of beds and occupancy to minimize staffing costs in an obstetrical unit? Can J Anaesth. 2001;48:295-301.[ go to PubMed ]
4. Aiken LH, Clarke SP, Sloane DM, Sochalski J, Silber JH. Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. JAMA. 2002;288:1987-93.[ go to PubMed ]
5. Schubert A, Eckhout G, Tremper K. An updated view of the national anesthesia personnel shortfall. Anesth Analg. 2003;96:207-14.[ go to PubMed ]
6. Byrick RJ, Craig D, Carli F. A physcian workforce planning model applied to Canadian anesthesiology: assessment of needs. Can J Anaesth. 2002;49:663-70.[ go to PubMed ]
7. Craig D, Byrick R, Carli F. A physcian workforce planning model applied to Canadian anesthesiology: planning the future supply of anesthesiologists. Can J Anaesth. 2002;49:671-7.[ go to PubMed ]
8. Rowland RG, Wofford DA. Are you prepared for a shortage of anesthesia providers? Healthc Financ Manage. 2003;57:66-70.[ go to PubMed ]
9. Hawkins JL, Gibbs CP, Orleans M, Martin-Salvaj G, Beaty B. Obstetric anesthesia work force survey, 1981 versus 1992. Anesthesiology. 1997;87:135-43.[ go to PubMed ]
10. Halpern S, Watson-MacDonell JW. Optimizing obstetrical suite staffing: it’s more than mathematicals. Can J Anaesth. 2001;48:219-21.[ go to PubMed ]
11. Qureshi AM, Stevens M, Plaat F. Survey of anaesthetic support staff in obstetric units in England and Wales. Anaesthesia. 2003;58:578-82.[ go to PubMed ]
12. Lavin JP, DiPasquale L, Crane S, Stewart J. A state-wide assessment of the obstetric, anesthesia, and operative team personnel who are available to manage the labors and deliveries and to treat the complications of women who attempt vaginal birth after cesarean delivery. Am J Obstet Gynecol. 2002;187:611-14.[ go to PubMed ]
13. Sica GT, Barron DM, Blum R, Frenna TH, Raemer DB. Computerized realistic simulation: a teaching module for crisis management in radiology. AJR Am J Roentgenol. 1999;172:301-4.[ go to PubMed ]