In June 2009, when the World Health Organization declared the H1N1 influenza virus a pandemic, a total of 74 countries and territories had reported laboratory-confirmed infections.1 In the United States, the Centers for Disease Control and Prevention (CDC) has estimated that during the entire season, from April 2009 through April 10, 2010, between 43 million and 89 million cases of 2009 H1N1 occurred. The CDC also estimated that 8,870 to 18,300 of these resulted in hospitalization during that same time.

The event highlighted limitations in surge capacity strategies in the United States, particularly in respiratory care. One of the biggest challenges in tackling this issue, according to experts, has been uncertainty regarding the ventilator inventory in the United States. Previously published estimates ranged from 54,000 to more than 105,000, according to Lewis Rubinson, MD, PhD, Commander, US Public Health Service, and Deputy Chief Medical Officer, National Disaster Medical System, US Department of Health and Human Services (HHS), Washington, DC. The numbers of pediatric-capable devices were completely unknown.These widely varied numbers were typically due to study limitations. Some of the previous research studies had not been led by respiratory care experts, introducing miscommunications regarding terminology and equipment. “Transport ventilators are obviously different than devices intended for noninvasive ventilation, which are different [from] full-feature ICU ventilators,” Rubinson specifies. Subsequently, many of the studies did not produce defensible results. “There was not a good way to get the number without really reaching out to institutions and asking,” Rubinson says. Without an accurate inventory to work with, disaster planners have been at a disadvantage.

“Surge capacity strategies are needed to plan for emergencies that exceed full-feature ventilator capacity, but [they] are difficult to plan with unknown numbers (for adult and pediatric populations) and unknown quality of devices (many different types, capabilities, and features),” wrote the study authors.2

To fill in this data gap, HHS collaborated with the American Association for Respiratory Care (AARC) to complete a study that would produce relevant accurate numbers. The results of this study were published last year in the journal Disaster Medicine and Public Health Preparedness.2

While the research captured an accurate picture of a defined ventilator inventory in US acute care hospitals, there are many questions still unanswered, and disaster preparedness planners will require additional research to develop adequate emergency strategies. In the meantime, the numbers from Rubinson et al have already been shared with state organizations and used in multiple ways. “We know it’s helped at a variety of levels already,” Rubinson says.

Survey Asks

To determine an accurate count of adult and pediatric mechanical ventilators in US acute care hospitals, the team developed a survey that was sent to 5,752 US acute care hospitals listed in the 2007 American Hospital Association database. The questionnaire generated an overwhelming 75% response rate.2 A total of 4,305 institutions replied, accounting for approximately 84% of US intensive care unit beds.2

Survey participants provided information on the number and types of mechanical ventilators owned by their respective medical facilities in 10 categories: full-featured, high-frequency, portable mechanical (pneumatic-driven only), portable, basic emergency medical services transport, noninvasive, CPAP-only, automatic resuscitators, neonatal-pediatric-specific, and standby (no longer used for everyday patient care but maintained and available on site).2 The survey response tool listed all mechanical ventilators and automatic resuscitators approved by the US Food and Drug Administration at the time of the initial mailing; manual resuscitators and anesthesia machines were not included.

Although there were occasional glitches, overall, the collection method worked well. “People gave the information gladly, and we are very thankful,” Rubinson says.

Survey Says

Using analysis methods described in detail within the study, Rubinson et al calculated that US hospitals own 63,066 full-feature mechanical ventilators or approximately 20.5 per 100,000 population.2 Of these, 28,811 are pediatric-capable, translating to 50.7 per 100,000 population.2 When ventilators other than full-feature devices were counted, the researchers estimated their number to be 98,738.2 The inventories varied widely across states.2

While previous studies had not been consistent, the resulting overall numbers from this effort matched some earlier estimates and did not generate much debate. The number of pediatric units, however, did catch some off guard—in a good way. “I was really surprised when the national study showed so many of these ventilators are compatible with that age group,” says Nick Kuhnley, RRT, manager of the respiratory care department at North Memorial Health Care in Robinsdale, Minn.

Rubinson was also surprised by the relatively high number of pediatric-capable devices, as well as by the fact that institutions had purchased so many automatic resuscitators. “My bias is they have a very limited role, if any, for definitive mechanical ventilation,” Rubinson says.

Focusing Scope

Some suggest that Rubinson’s team had its own bias toward full-feature mechanical ventilators. “You’re not going to need all the bells and whistles that are in a $15,000 or $20,000 ventilator to meet the needs of mass casualty ventilation. Less sophisticated ventilators can also ventilate these patients as effectively as the more sophisticated ventilators,” says Frank Rando, CRT, PA, EMTP, Arizona Health Sciences Center (Tucson); the Arizona Emergency Medicine Research Center and the Center for Public Health Preparedness at the University of Arizona (Tucson); consultant, Department of Homeland Security; and health systems preparedness and management coordinator for Allied Healthcare Products Inc in St Louis.

The study authors acknowledged this limitation, noting they also did not include devices rented by hospitals, those utilized at nursing facilities other than long-term ventilator and rehabilitation facilities, or those used by schools for respiratory care.2 And although the team could not verify that the nonresponsive hospitals had situations similar to those of the respondents, it was hypothesized that the high response rate would mitigate any response bias.2

The scope of the study was limited to generating an inventory figure; therefore, the survey was not designed to elicit information regarding device availability, and the authors acknowledge that the numbers do not provide an exact picture of the units that could be put to use immediately.2

Recognizing Limits

Some studies have estimated that the number of ventilators in use at any time could represent roughly 80% of the inventory, according to Rando. This could mean that the number of ventilators actually available for a surge represents a much smaller portion of the full inventory. In addition, there could be a geographic discrepancy in patient need and ventilator supply.2

“Recognizing this vulnerability, contingency planning for worsening of the [current] influenza pandemic or a future event has led HHS to continue to expand the number of PPC devices in the Strategic National Stockpile and support the development of novel surge mechanical ventilators.”2

Other issues, particularly the personnel and supplies required to run the devices, could also impact surge response. The Rubinson study did not collect data on ventilator circuits, humidification equipment, suction equipment, physiologic monitors, medical gas, or airway interfaces in an effort to maximize the response rate.2

A state study did and found varied response. “We did look at that in the state of Minnesota and found a lot of hospitals do not have enough supplies. An interruption in the supply chain could mean some ventilators would have to be taken out of service early because we would not have the circuit connections to use them,” Kuhnley says. North Memorial had a larger inventory than most other facilities in the state, with an average of nine circuits per ventilator. Kuhnley estimates this would likely cover a 6-week epidemic, depending on the ventilator turnover.

Adequate personnel is harder to estimate. Full-feature ventilators typically require a respiratory therapist to operate properly. “In any particular area, there may be a drop in respiratory expertise in terms of actual human resources that are able to respond to the event, because the respiratory therapists may become sick or injured, they may decide to stay at home with their families and not show up to work, [and/or] the infrastructure may be so disrupted that they may not be able to get to work,” Rando says.

This is a definite issue, Rubinson concurs. “The study was not meant to say, ‘Don’t worry. Take this off the table.’ But it was [designed] to provide credible data that the sky is not falling for a low to moderate event that is slow to occur that involves different regions at different times,” Rubinson says, citing the 2009 influenza pandemic as an example.

Taking a Snapshot

The Rubinson study was meant to produce an accurate snapshot of the full-feature ventilator inventory in US acute care hospitals at a particular point in time. “Ventilator purchases and counts are dynamic,” Rubinson says, noting they tend to change slowly due to long device life cycles and the need for acquisition capital for replacement.

Based on the numbers reflected by the survey, Rubinson et al concluded that the United States has a sufficient number of mechanical ventilators to handle an event of severity similar to that of the influenza pandemic. “The comparisons are somewhat limited, but looking at other countries that are pretty well resourced—Australia, New Zealand, Canada—people are not going without ventilators every day there,” Rubinson says. Compared to those countries with published data, the United States has far more inventory per 100,000 population.

An adequate inventory for one type of situation does not mean the inventory will be adequate for all situations, however. The need during more severe events, such as detonation of a nuclear device, could exceed the number of devices in the inventory. In addition, the logistics remain unclear and, therefore, unreliable.

Having a count has already helped planners, who previously lacked reliable inventory numbers such as those generated by the study. The Department of Health and Human Services has used the data to better understand where distribution differences exist throughout the country so that it can better support local governments. “Disease and need are not homogeneous, but if you know that you have a issue with resource distribution availability, you can get smarter about where you expect the gaps to happen,” Rubinson says.

Kuhnley intends to use the data similarly at the state level. “I would like to see a better complement of infant and pediatric ventilators in this state,” Kuhnley says, suggesting the data will provide evidence-based support for any initiative.

Filling in Gaps

More evidence would be even more helpful, and there is vast consensus that more research is needed. Possible objectives for future research could, and probably should, address supply and geographic distribution of staff and ancillary equipment, utilization rates, appropriateness of PPV use, standardization of PPV device functional capabilities, and staff training.

In addition, another snapshot of inventory, using a standard methodology, would be useful. “You actually need to have two studies like this before you can measure growth. In the state of Minnesota, we would like to do this about every 3 years,” Kuhnley says, noting previous research has indicated a growth rate of roughly 6% per year for most years.

Once a baseline number is established, as Rubinson et al have done, an important next step is to determine availability. “It varies by season and, sometimes, even by day,” Rubinson says.

Perhaps even more important, however, staffing issues need to be examined. “If there aren’t enough respiratory therapists to operate [the ventilators], who’s going to operate the rest of them?” Rando asks.

If the need for personnel or sophisticated equipment is unmet, decisions must also be made regarding adequate replacements. HHS has conducted some consensus work over the past 5 to 6 years regarding which devices should be used if the recommended devices are not available.

“Most of that information is expert opinion, which, at some point, needs to be validated. But that’s not easy,” Rubinson says, noting the challenge in studying a device that lacks the features of another. “I think most of that will have to come from systems that are already overburdened—like developing country systems—where we can learn from them because they’re already in a resource-limited environment,” Rubinson says.

Fortunately, in the United States, according to Rubinson, it is rare to hear even an anecdotal story of a patient going without ventilation. The goal is to maintain that quality during an emergency. Although many more questions demand answers, this latest accounting brings planners closer to an effective surge capacity strategy.

Renee Diiulio is a contributing writer for RT. For more information, contact [email protected].

  1. World Health Organization. What is the pandemic (H1N1) 2009 virus? Global alert and response. Available at: Accessed on January 10, 2011.
  2. Rubinson L, Vaughn F, Nelson S, et al. Mechanical ventilators in US acute care hospitals. Disaster Med Public Health Preparedness. 2010;4:199-206.