Faced with increased demands for critical care services as a result of the novel H1N1 pandemic, hospitals must prepare a surge response in an attempt to manage these needs. In discussing a surge response, the commonly used paradigm considers the factors of staff, stuff (supplies and equipment), space, and systems necessary to respond to the event (1). Several excellent resources address many of these issues in general (2–6). This article uses this general framework to discuss surge issues in the context of H1N1 and to provide specific advice for hospitals to consider.
Much of the planning for disasters has focused on emergency departments, whereas much of the pandemic planning has been focused on public health. In both situations, the importance of incorporating the inpatient areas of hospitals—particularly critical care units, which play a key role in this into planning—has often been overlooked (7, 8). However, not only are critical care units at a particularly high risk of being impacted by the return of H1N1 (9–11) but also the duration of the impact is likely to be more prolonged than in other areas of the health system.
Most intensive care units (ICUs) in Canada, and a large number in the US, operate at nearly 100% of their usual capacity (1, 12–16). Therefore, even a small increase in the number of patients requiring critical care can have a major impact on the system. In Mexico and Manitoba during the first wave of H1N1, significant stresses were placed on the critical care system (17–19). Thus, effective management strategies must be developed to respond to such situations.
H1N1 brings with it not only increased numbers of patients but also younger patients, many with severe lung injury, creating challenges similar to the management of acute respiratory distress syndrome patients who require prolonged ventilation and ICU management (20). Thus, although the impact on the ICU will mirror the increase of the community's epidemic curve at the outset, there will be a lag of weeks to months after the decline in new cases in the community before the impact on ICUs will resolve. This situation is not unique to a pandemic. Disasters such as the Rhode Island night club fire in 2003 (20–22), the London Tube bombings in 2005 (23), and Madrid train bombings in 2004 (24) all resulted in prolonged surge situations in ICU for weeks after these single-impact events. With a pandemic wave that lasts 8 or 10 wks, the impact is even more prolonged. The number of patients admitted to the ICU is dependent on the absolute number of infected individuals in the community and the severe complication rate experienced among those infected.
Determining ICU Demands in a Pandemic
To match surge capacity with the potential increase in demand during a wave of pandemic H1N1 (pH1N1) influenza, individual hospitals first estimate how many additional patients will need accommodation within the ICU. To do this, one must define four parameters: (1) what proportion of the surrounding population (i.e., those who would normally access the hospital for medical care) will be infected in the upcoming wave; (2) how much of the infected population is at risk for hospitalization; (3) what proportion of hospitalized patients will require ICU care; and (4) what is the expected length of stay in the ICU. This is a difficult task, because uncertainty exists with many of these parameters, especially in the population-based estimates. For H1N1, even in countries that have experienced a previous wave of influenza activity in the spring months, it is unclear how the virus will affect the population in subsequent waves. Some proportions may have already been infected with H1N1 and thus are immune. However, two factors make it difficult to project the number of immune individuals: changes in surveillance and testing strategies in most countries, which occurred midway through the initial wave, and the prevalence of subclinical infection with H1N1 influenza.
Seroprevalence studies, where available, may assist in answering this question. Challenges also lie in determining the proportion of people infected with H1N1 who will require hospitalization. Estimates from compiled data thus far are problematic; for example, whereas all patients hospitalized with influenza-like illness in Canada are tested for H1N1, there is no accurate denominator for the overall number of people infected with H1N1 in the country. Similarly, it is impossible to truly know the case fatality ratio of the H1N1 virus.
A number of intervening factors may alter the course and/or severity of a pandemic wave. The use of antiviral medications for early treatment and/or prophylaxis may mitigate the progression of a wave. This must be tempered with the possibility of development of antiviral resistance if the medications become used more liberally, especially for prophylaxis. Timely delivery of the pandemic virus vaccine to groups at risk for severe influenza illness also may influence the duration and severity of effect. Given that the pandemic influenza vaccine is a new product and will be administered separately from the seasonal influenza vaccine, public perception of dual vaccination and vaccine uptake must be taken into account. Finally, social distancing measures, such as school closures and cessation of mass gatherings, may result in an abrupt decline in virus transmission or a decrease the peak of H1N1 infection, forming a more protracted wave as transmission continues in a less efficient environment.
These uncertainties apply to planning for the impact of a pandemic wave at all levels of jurisdiction. For hospital-based planning, there is the additional problem that there is no absolute way to anticipate how severely the immediate vicinity of a given hospital may be affected by an influenza wave. From the initial wave in the northern hemisphere last spring and more recent experiences in the southern hemisphere, it is clear that the H1N1 virus strikes with geographic variability—both in terms of numbers and severity (25). Making predictions on a hospital level requires knowledge of the characteristics of the surrounding population, including the age distribution, population density, and cultural make-up. In addition to this, hospitals need to consider the number of schools and long-term care facilities in proximity while planning for an influx of ill patients. Finally, a basic understanding of the extent of ICU capacity in nearby hospitals is required to account for possible diversion of patients when local resources become depleted.
Modeling Programs for Pandemic Influenza
Ideally, each province or state would independently develop tailored population-based models for ICU needs, integrating local estimates for at-risk groups, age-specific attack rates, social mixing patterns, and virus-specific features given different patterns of antiviral and vaccine use. Although this strategy is underway in some areas, it may not be practical for everyone, given the required time and expertise to perform, as well as constantly changing parameters as new data unfold.
Currently, the most widely used modeling program for early or pre-pandemic planning is the Centers for Disease Control and Prevention-endorsed FluSurge 2.0 (http://www.cdc.gov/flu/tools/flusurge.htm).Several countries, including France, the United Kingdom, the Netherlands, and New Zealand, have used FluSurge 2.0 or modified versions of the program as the basis for public health, hospital, and ICU resource planning in a pandemic (11, 26–28). FluSurge is a static model that was originally developed in 1999 by Meltzer et al (29) to describe the cost-effectiveness of vaccination in an influenza pandemic. As a static model, it does not account for the effect of virus transmission (and therefore interventions that may change the dynamics of transmission, such as the use of antivirals, antibiotics, and vaccination) or the shape of the epidemiologic curve in producing estimates of impact. Another shortcoming of the program is that it uses historic data from particular regions in the US to extrapolate risk for hospitalization. This calculation cannot be modified by users, and it may not be applicable to other situations, for example, where different age groups in the population may be disproportionately affected by the virus.
There are few user-friendly alternatives to FluSurge for rapid application in an emergent setting. StatFlu (www.s-gem.se/statflu) is a static modeling program that was recently developed in an attempt to improve some of the limitations of FluSurge (30). It allows for more assumptions to be modified according to the pandemic situation and thus provides broader, more locally applicable ranges of possible outcomes. Like FluSurge, it does not incorporate the effect of antiviral and vaccine use in the calculations. StatFlu has been adopted by the Swedish National Board of Health and Welfare to estimate hospital load attributable to a pandemic (30), but it has not yet achieved widespread acceptance.
Estimating the Impact of the H1N1 Pandemic on ICU Load
Using FluSurge 2.0 and current information about the H1N1 virus, we demonstrate how two different planning scenarios can be created in a fictional hospital in Ontario, Canada, by varying attack rates, need for ICU admission, and length of stay in ICU (Table 1). Hospital A is a tertiary care center with 388 inpatient beds and 71 ICU beds. The following series of assumptions and parameters are used:
The age-stratified population for the hospital is calculated by determining the percentage of total hospital beds in Ontario that hospital A contains and multiplying this number with age-stratified Ontario census data. Note that this process does not take into account the provincial ICU bed contribution of hospital A. Based on the Canadian experience to date, it is assumed that most H1N1 influenza patients requiring ICU care will present in moderate distress to their local emergency departments before severe respiratory deterioration occurs (20).
The attack rate is either 25% or 35%, depending on the scenario. These numbers reflect the unknown rate of population immunity to H1N1 from the initial wave in the spring.
The pandemic wave in the community surrounding hospital A is expected to last 8 wks. In reality, it is possible that the effect of the pandemic on the ICU may be protracted because of uneven accumulation of patients over time.
All ICU beds are staffed at 100% capacity throughout the pandemic wave. It is assumed that human resources are not a limiting factor.
The percentage of hospitalized patients requiring ICU care is either 25% or 35%, depending on the scenario. The first estimate is based on Ontario hospital surveillance data (approximately 20%) and the rate of ICU admission in hospital A's community during the first wave of H1N1 (approximately 25%) (31). The second parameter estimate is derived from reported rates of ICU admission in hospitalized patients in Australia during the height of their last wave, during which time the proportion of hospitalized patients in the ICU varied from 27% to >40% (32). This is coupled with concerns that more severely ill patients will present to hospital A, realizing that it is a relatively resource-rich tertiary care center.
The average length of stay in the ICU is 10 days. This is derived from the H1N1 ICU experience in Canada to date, where the mean length of ICU admission was just >10 days (20). It is recognized, however, that there will likely be a wide range in length of ICU stay because many patients were admitted for >20 days.
As mentioned, one key parameter that cannot be modified in FluSurge is the rate of hospitalization according to the age structure of the population. This parameter estimate is built into the program using available data from previous pandemics. The observed morbidity and mortality rates from the pH1N1 to date have been significantly lower than those documented from previous pandemics. Thus, the outputs for both of these scenarios likely overestimate the true magnitude of surge needs.
These two scenarios (Table 1) may be viewed as a sensitivity analysis in light of the many uncertainties in modeling the pandemic impact. Scenario A depicts the “best case scenario,” in which the attack rate in hospital A's vicinity is 25% and the percentage of hospitalized patients requiring ICU care is 25%. Scenario B depicts the “worst case scenario,” in which the attack rate is 35% and the percentage of hospitalized patients requiring ICU care is 35%.
It is immediately clear that in both scenarios the ICU bears the heaviest impact in terms of capacity. The effect begins in the first week of the pandemic and peaks in week 5. Given the number of limitations with this modeling exercise (both in estimating the parameters and inherent to the modeling program), it is more useful to note the trends, and perhaps not the actual numbers, to guide ICU resource planning.
Surge Response: System Implementation Issues
As highlighted, no one knows how significant the ICU surge will be. In planning a response, a few simple principles should be considered. First, with H1N1, as with any other surge, the possible impacts with pH1N1 lie on a spectrum from minor day-to-day basis to a major surge resulting from overwhelming demands on the system (33–35). As a result, the methods used in the response should be scalable, and systems and processes used to respond to major surges should build on those used to respond to minor and moderate surges (36). Standalone systems should be avoided.
The first and most critical factor that must be in place to successfully respond to a surge is an functional command and control system (37). Hospitals should use an incident management system to manage their internal response (38). Ideally, the hospital will integrate into a well-coordinated, formal system of local, regional, and provincial/state incident management system structures (3, 37, 39, 40), but such formal networks are complex and take time to develop. If they are not already in place, then less formal or structured networks, often referred to as healthcare coalitions, can be established more rapidly and facilitate effective coordination (41). Similarly, in an ideal situation, formal communication networks and protocols will have been established and exercised; however, if these are lacking, then strategies as simple as regular teleconferences can be effective for information sharing and rumor control, as well as providing both peer support and expert advice to coalition members (8, 42, 43).
The second critical factor is situational awareness. For those in critical care in the H1N1 context, this amounts to having a good grasp on resource supply (staff, stuff, and space), system demands, and other evolving events within the walls of the hospital and beyond. An individual hospital will not have enough information to make accurate decisions regarding the need to shift to mass critical care or to initiate triage in a major surge, so hospitals must communicate with regional and provincial/state authorities to develop adequate situational awareness (1–6, 37, 39, 44). For higher levels of authority to develop this common information picture, individuals must cooperate by collecting and sharing data regarding resource supply and demand (i.e., number of admissions to the ICU, number of available ventilators, staffing levels, etc). Again, if a formal system is lacking, then healthcare coalitions may assist in data collection and collation that can be communicated to higher authorities.
Surge Response: Resources (Staff, Stuff, and Space)
When considering the resource issues in any surge response, the very heart of the issue is the simple concept of supply vs. demand. The response strategy must address both sides of this equation by working to increase the supply of resources as well as decrease the demand on the system. A full discussion of this topic is beyond the limitations of this piece and much advice has been published elsewhere. We focus on issues of unique concern when preparing for H1N1. The basic approach of “adapt, substitute, conserve, re-use, re-allocate” should be used to extend limited resources (5, 6, 36, 45). The Minnesota Department of Health provides an excellent example of how to apply this framework when preparing for a pandemic (46).
Stuff (supplies and equipment) are most often the rate-limiting factor in critical care services (1). Ventilator supply is a particular concern in responding to H1N1 because it is the most finite supply and does not have a substitute. Rubinson (47, 48) has written extensively about ventilator stockpiling, and a number of excellent sources provide advice on supply considerations for mass critical care (5, 6, 36, 45). One area in which guidance is not readily available is in pharmaceutical stockpiling.
The clinical guidelines for medications necessary in a mass critical care event tend to focus on antiviral therapies for pandemics or disease-specific antidotes for a bioterrorism attack. Although vaccines, antimicrobials, and antidotes are keys to a contingency plan, based on the available H1N1 information to date, the emergency plan should include estimations of a wide range of critical care drug therapies, including those to support mechanical ventilation. Hospital pharmacy departments typically maintain varied reserves of drug therapies based on factors such as formulary status, frequency of use as dictated by the institution's patient population, shelf life, and associated acquisition costs. Typically, it has been recommended to stockpile agents necessary for up to 10 days into a disaster (5, 6). Critical care medication resource planning for a pandemic should assume resource scarcity because of the number of weeks impacted by the pandemic and the potentially limited human resources required to support surge manufacturing. The traditional just-in-time supply chains may be difficult to maintain. Thus, pharmacists should be involved in determining the institution's essential agents and in facilitating pre-negotiated agreements with manufacturers to help avoid delays in stock acquisition.
Numerous individual drug therapies are used in the ICU during routine, complex, or labor-intensive (e.g., dialysis, tight glycemic control) or expensive (e.g., activated protein C) operations. Because supplies and staff to support complex drug therapies are likely to be impacted during a pandemic, we suggest developing a narrow critical care formulary of essential agents that offer some potential benefit and are relatively affordable. Despite the numerous uncertainties, in 2007 we attempted to determine the types and quantities of essential items for our ICU for a potential avian flu pandemic. The Appendix outlines the thought processes we applied to this activity. Numerous amounts of knowledge were encountered in researching and developing our list. Few resources suggested the types of agents to stock, and only one demonstrated how to estimate quantities (Radonovich L). Perhaps the most important limitation of this approach is the reliance on assumptions, such as the severity of disease and the rate of ICU-related complications.
Staff will likely be the second most likely resource to become scarce during the H1N1 pandemic because of a combination of factors, including illness, fatigue, fear, and caregiver duties (particularly if school closures occur). Hospitals should have plans to redeploy staff to support areas such as critical care, general medicine, emergency, and occupational health departments. These plans should address physician, nursing, and allied healthcare worker staff resources. In addition to redeploying staff, the manner in which they are organized may need to shift to care team models (5, 6, 49). Further, clinician time efficiencies will be needed; physician rounds, particularly in teaching hospitals, are a very inefficient use of time, with many residents idly observing. Respiratory therapists may be in particularly short supply. The XTREME project provides guidance on augmenting respiratory therapists during a disaster and includes a just-in-time training module, which is available on DVD (50).
Space is least likely to be a limiting factor in critical care capacity. The Task Force on Mass Critical Care has provided specific guidance for expanding critical care capacity (5). However, one issue not addressed is the management of satellite critical care units, which are often geographically distant from the main ICU and present difficulties in coordinating patient flow and supervising non-ICU staff brought in to help. One solution is to group the sicker or more complex patients with H1N1 together in the main ICU and use the satellite units to manage more routine postoperative and less complex patients. The use of an administrative intensivist whose role it is to manage all patient flow and liaise with senior administration creates a single point of contact for critical care and allows those working in the ICU to focus on patient management.
Preparing to respond to a surge of patients associated with the H1N1 pandemic presents some unique challenges. It is best to develop systems that build on the systems in place rather than trying to develop and implement completely new systems in the middle of a disaster. The plans created should form a component of the standard responses to use in future surges. As demonstrated by the discussion on modeling the potential demands, hospital surge scenarios may range from very minor to fairly severe. Hospitals must recognize the need for plans to be scalable and flexible. Because of these uncertainties, hospitals need to rely heavily on their system of command and control, along with effective communication with external partners, to allow them to scale the response appropriately.
1.Christian MD, Devereaux AV, Dichter JR, et al: Definitive care for the critically ill during a disaster: Current capabilities and limitations: From a Task Force for Mass Critical Care summit meeting, January 26–27, 2007, Chicago, IL. Chest
2008; 133(5 Suppl):8S–17S
2.Committee on Guidance for Establishing Standards of Care for Use in Disaster Situations: Guidance for Establishing Crisis Standards of Care for Use in Disaster Situations: A Letter Report. Washington, DC, Institute of Medicine, National Academy of Sciences, 2009
3.Devereaux A, Christian MD, Dichter JR, et al: Summary of suggestions from the Task Force for Mass Critical Care summit, January 26–27, 2007. Chest
2008; 133(5 Suppl):1S–7S
4.Devereaux AV, Dichter JR, Christian MD, et al: Definitive care for the critically ill during a disaster: A framework for allocation of scarce resources in mass critical care: From a Task Force for Mass Critical Care summit meeting, January 26–27, 2007, Chicago, IL. Chest
2008; 133(5 Suppl):51S–66S
5.Rubinson L, Hick JL, Curtis JR, et al: Definitive care for the critically ill during a disaster: medical resources for surge capacity: From a Task Force for Mass Critical Care summit meeting, January 26–27, 2007, Chicago, IL. Chest
2008; 133(5 Suppl):32S–50S
6.Rubinson L, Hick JL, Hanfling DG, et al: Definitive care for the critically ill during a disaster: A framework for optimizing critical care surge capacity: From a Task Force for Mass Critical Care summit meeting, January 26–27, 2007, Chicago, IL. Chest
2008; 133(5 Suppl):18S–31S
7.Eggertson L: Critical care doctors want escalated pandemic planning. CMAJ
8.Hawryluck L, Lapinsky S, Stewart T: Clinical review: SARS—lessons in disaster management. Crit Care
9.Rello J, Rodriguez A, Ibanez P, et al: Intensive care adult patients with severe respiratory failure caused by Influenza A (H1N1)v in Spain. Crit Care
10.Lister P, Reynolds F, Parslow R, et al: Swine-origin influenza virus H1N1, seasonal influenza virus, and critical illness in children. Lancet
11.Ercole A, Taylor BL, Rhodes A, et al: Modelling the impact of an influenza A/H1N1 pandemic on critical care demand from early pathogenicity data: The case for sentinel reporting. Anaesthesia
12.Brilli RJ, Spevetz A, Branson RD, et al: Critical care delivery in the intensive care unit: Defining clinical roles and the best practice model. Crit Care Med
13.Halpern NA, Pastores SM, Greenstein RJ: Critical care medicine in the United States 1985–2000: An analysis of bed numbers, use, and costs. Crit Care Med
14.Halpern NA, Pastores SM, Thaler HT, et al:Changes in critical care beds and occupancy in the United States 1985–2000: Differences attributable to hospital size. Crit Care Med
15.Houser SH, Houser HW: Are we preparing health services administration students to respond to bioterrorism and mass casualty management? J Health Adm Educ
16.Needham DM, Bronskill SE, Calinawan JR, et al: Projected incidence of mechanical ventilation in Ontario to 2026: Preparing for the aging baby boomers. Crit Care Med
17.Chowell G, Bertozzi SM, Colchero MA, et al: Severe respiratory disease concurrent with the circulation of H1N1 influenza. N Engl J Med
18.Kermode-Scott B: Canada has world's highest rate of confirmed cases of A/H1N1, with Aboriginal people hardest hit. BMJ
19.Perez-Padilla R, de lR-Z, Ponce de LS, et al: Pneumonia and respiratory failure from swine-origin influenza A (H1N1) in Mexico. N Engl J Med
20.Kumar A, Zarychanski R, Pinto R, et al: Critically ill patients with 2009 influenza A (H1N1) infection in Canada. JAMA
21.Dacey MJ: Tragedy and response- the Rhode Island nightclub fire. N Engl J Med
22.Mahoney EJ, Harrington DT, Biffl WL, et al: Lessons learned from a nightclub fire: Institutional disaster preparedness. J Trauma
23.Aylwin CJ, Konig TC, Brennan NW, et al: Reduction in critical mortality in urban mass casualty incidents: Analysis of triage, surge, and resource use after the London bombings on July 7, 2005. Lancet
24.Gutierrez de Ceballos JP, Fuentes FT, Diaz DP, et al: Casualties treated at the closest hospital in the Madrid, March 11, terrorist bombings. Crit Care Med
26.Doyle A, Bonmarin I, Levy-Bruhl D, et al: Influenza pandemic preparedness in France: Modelling the impact of interventions. J Epidemiol Community Health
27.Nap RE, Andriessen MP, Meessen NE, et al: Pandemic influenza and hospital resources. Emerg Infect Dis
28.Wilson N, Mansoor O, Baker M: Estimating the impact of the next influenza pandemic on population health and health sector capacity in New Zealand. N Z Med J
29.Meltzer MI, Cox NJ, Fukuda K: The economic impact of pandemic influenza in the United States: priorities for intervention. Emerg Infect Dis
33.Barbisch DF, Koenig KL: Understanding surge capacity: Essential elements. Acad Emerg Med
34.Jenkins JL, O'Connor RE, Cone DC: Differentiating large-scale surge versus daily surge. Acad Emerg Med
35.Kaji A, Koenig KL, Bey T: Surge capacity for healthcare systems: A conceptual framework. Acad Emerg Med
36.Hick JL, Barbera JA, Kelen GD: Refining surge capacity: Conventional, contingency, and crisis capacity. Disaster Med Public Health Prep
2009; 3(2 Suppl):S59–S67
37.Christian MD, Kollek D, Schwartz B: Emergency preparedness: What every health care worker needs to know. Can J Emerg Med
38.Barbera JA, Macintyre AG: Medical and health incident management (MaHIM) system: A comprehensive functional system description for mass casualty medical and health incident management. Washington, DC: Institute for Crisis, Disaster, and Risk Management, The George Washington University, October 2002
39.Burkle FM Jr, Hsu EB, Loehr M, et al: Definition and functions of health unified command and emergency operations centers for large-scale bioevent disasters within the existing ICS. Disaster Med Public Health Prep
40.The CNA Corporation. Medical Surge Capacity and Capability: A management system for integrating medical and health resources during large-scale emergencies. Department of Health and Human Services, August 2004
41.Courtney B, Toner E, Waldhorn R, et al: Healthcare coalitions: The new foundation for national healthcare preparedness and response for catastrophic health emergencies. Biosecur Bioterror
42.Booth CM, Stewart TE: Communication in the Toronto critical care community: Important lessons learned during SARS. Crit Care
43.Booth CM, Stewart TE: Severe acute respiratory syndrome and critical care medicine: The Toronto experience. Crit Care Med
2005; 33(1 Suppl):S53–S60
44.Christian MD, Hawryluck L, Wax RS, et al: Development of a triage protocol for critical care during an influenza pandemic. CMAJ
45.Hick JL, Hanfling D, Burstein JL, et al: Health care facility and community strategies for patient care surge capacity. Ann Emerg Med
47.Daugherty EL, Branson R, Rubinson L: Mass casualty respiratory failure. Curr Opin Crit Care
48.Rubinson L, Branson RD, Pesik N, et al: Positive-pressure ventilation equipment for mass casualty respiratory failure. Biosecur Bioterror
50.Hanley ME, Bogdan GM: Mechanical ventilation in mass casualty scenarios. Augmenting staff: project XTREME. Respir Care
ICU Prescription List
The following information is a summary of methodologies and assumptions used to estimate the drug stock quantities that need to be available to manage mechanically ventilated ICU patients for a 4-wk time block during an influenza pandemic.
Materials and Methods
A computerized literature search of MEDLINE, EMBASE, International Pharmaceutical Abstracts, the Cochrane Central Register of Controlled Trials and Science Citation Index databases, and Internet search engines Google and Google Scholar was undertaken between the time period 1950 through September 2009. In anticipation of limited published resources, no exclusion criteria were specified for the search (i.e., any controlled study, uncontrolled study, case series, or review article would be reviewed). References of identified articles were reviewed manually for additional references not identified by the computerized search.
The systematic search did not identify any article describing medication resource planning, nor any paper providing guidance on the estimation of medication needed for mechanically ventilated ICU patients in the event of a pandemic. Identified articles only provided details on antiviral medications.
An attempt to identify any reference providing information on resource or medication planning for any critical care disaster was undertaken (e.g., bioterrorism). Again, identified articles only provided detail on stocking specific antidotes.
The search was extended to contacting individuals who were thought to have experience in disaster planning, such as North American Critical Care Pharmacy Network and Department of National Defense—Pharmacy Division. Again, little was identified.
Therefore, the basic principles of critical care were applied to estimate essential drug therapies and potential quantities. Calculations were extrapolated from the current utilization in our 16-bed medical-surgical ICU and our drug consumption during the winter months when pneumonia and septic shock are common admission diagnoses. Table 1 summarizes the assumptions to calculate 4-wk supplies. For indications in which multiple therapy options are available, the agent with the most favorable drug properties was selected (e.g., kinetics, dynamics, dosing interval, dilution requirements, etc). Dosages used for calculations were based on mean published doses from criterion standard references (Micromedix, American Hospital Formulary Service, Compendium of Pharmaceuticals and Specialties). For agents that are weight-based, a mean weight of 80 kg was used. The calculations focus on agents related specifically to critical care. No calculations have been made to estimate utilization of noncritical care drugs, such as those consumed by patients at home (e.g., antihypertensive, antidepressants, etc).