ICU Capacity Strain
May 1, 2016
Reprints
By Betty Tran, MD, MSc
Assistant Professor of Medicine, Pulmonary and Critical Care Medicine, Rush University Medical Center, Chicago
Dr. Tran reports no financial relationships relevant to this field of study.
Intensive care in the United States accounts for nearly 1% of the gross domestic product, and it is forecasted that there will be increasing demand for this type of care in the future as the population ages.1,2 Given current projections that the supply of ICU staff and beds will be constrained rather than expand to meet this increasing demand, ICUs will be faced with the challenge of continuing care delivery under conditions of increasing strain.2 Thus, there is growing interest in studying ICU capacity strain, defined as the temporally varying influence on a given ICU’s ability to provide high-quality care for patients who are or could be cared for in that ICU on any given day.3
IS THERE AN OPTIMAL MEASURE FOR ICU CAPACITY STRAIN?
A singular, optimal, validated measure of ICU capacity strain remains elusive. Theoretically, many elements can contribute to an ICU’s capacity to provide high-quality care equitably and include: the number of clinical providers available in the ICU (physicians, physician extenders, nurses, respiratory therapists, pharmacists, etc.), the efficiency of these providers, the number of available ICU beds and other fixed resources (e.g., ventilators, dialysis machines), the number of patients in need of an ICU bed (both current and new admissions), and the acuity of these patients.3
As such, studies focused on outcomes related to ICU capacity strain have used one or a combination of these elements as their primary exposures. Since no single variable is able to capture all the elements that contribute to the complexity of ICU capacity strain, each has its advantages and disadvantages. For example, multiple studies have used either number of available ICU beds, percentage bed occupancy rate, or ICU census, as these are intuitively associated with ICU capacity strain and easily measured. Disadvantages to using these measures are that they do not account for patient acuity, which can strain providers’ time and resources, or distinguish between current patients and new admissions, which usually involves more time and resource use up front.3
More complex measures of ICU capacity strain include adjustments for patient flow and acuity, although these also have disadvantages. For example, some recent studies have used the number of new admissions to an ICU in a day to account for patients who are likely to contribute to ICU capacity strain due to their consumption of increased provider time and energy; this, of course, assumes that the existing ICU census is less time-consuming and/or existing ICU patients are somehow less acute. Adjustment to ICU census can be made for patient acuity as measured via the Acute Physiology and Chronic Health Evaluation (APACHE) II score or the Mortality Prediction Model. Although these are validated scoring systems for patient severity of illness and prognosis, they are more difficult to measure given the need for medical data collection, which can be time-consuming and problematic in retrospective studies or in studies using administrative data.3
Interestingly, when physicians and nurses were surveyed on perceptions of ICU capacity strain, increased ICU census was associated with increased perceived capacity strain among both physicians and nurses.4 Among charge nurses, higher average patient acuity (assessed via APACHE II score) and higher census on the general medical wards were also associated with increased perceived capacity strain.4 Overall, there was moderate correlation between the perception ratings of physicians and nurses on the same day (intraclass correlation 0.45; 95% confidence interval [CI], 0.30-0.60).4
Although an optimal measure for ICU capacity strain has yet to be found, once developed and validated, the hope is that it can be used in future critical care research to detect the significant effects of interventions by reducing the unexplained variance in outcomes and organizational characteristics of ICUs that enable some to accommodate increasing strain better than others.3
WHAT ARE THE OUTCOMES ASSOCIATED WITH ICU CAPACITY STRAIN?
Theoretically, high demand for critical care relative to available supply on any given day could strain providers’ abilities to deliver high-quality care equally, limit the time and thought they can devote to each individual patient, and influence decisions regarding ICU admission and goals of care. Several studies have focused on outcomes due to strained ICU capacity.
Brown et al reported that for increasing daily admissions in a single academic ICU, daily rounding time increased, but less time was spent on each patient, particularly new admits; newly admitted patients received 1.38 fewer minutes (95% CI, - 2.43 to - 0.33 minutes; P = 0.002) of total rounding time and 0.73 fewer minutes (95% CI, -1.42 to - 0.07 minutes; P = 0.0113) of cognitive rounding time (time spent on the patient’s assessment and plan) per additional admission.5
Most other studies have chosen mortality as a primary outcome, with mixed results. An early study at a single center in Scotland found that patients exposed to times of high ICU workload (as measured by peak occupancy, average nursing requirement per occupied bed per shift, and ratio of occupied to appropriately staffed beds) experienced three-fold higher adjusted hospital mortality (odds ratio [OR], 3.1; 95% CI, 1.9-5.0) compared to patients where the ICU workload was only moderate during their stay.6 In a larger study in Manitoba, Chrusch et al found that discharge from the ICU when there was no vacancy was associated with an increased relative risk for readmission or unexpected death within 7 days of ICU discharge after adjustment for age, diagnosis, APACHE II score, and ICU length of stay (relative risk, 1.56; 95% CI, 1.05-2.31); the implication here is that increased ICU capacity strain can affect physician decision-making and result in premature discharge from the ICU.7 Coming from a different angle, a multi-center study from France reported that compared to patients who were admitted immediately to the ICU on request, those whose admissions were delayed due to a full ICU unit had an increased incidence (although nonsignificant) of death on day 28 (OR, 1.78; 95% CI, 0.99-3.21), and day 60 (OR, 1.83; 95% CI 1.03-3.26).8 The authors concluded that ICU bed shortage could result in a lost opportunity and subsequently preventable deaths. In a similar vein, decreasing medical ICU bed availability was found to be significantly associated with increased ward cardiac arrest rates (OR, 1.25; 95% CI, 1.06-1.49), although causality in this study cannot be implied, as it could not be determined whether those who suffered cardiac arrest were evaluated for transfer to the ICU prior to their arrest.9 In the largest study to date consisting of > 260,000 patients admitted to 155 U.S. ICUs, higher standardized ICU census (to account for comparisons among ICUs of different sizes) on the day of admission was associated with an increased odds that admitted patients would die in the hospital (OR for a standardized unit increase, 1.02; 95% CI, 1.00-1.03; P = 0.02); this effect strengthened when the standardized census consisted of sicker patients (OR, 1.06; 95% CI, 1.01-1.11 for the highest decile of ICU acuity).8 Similar results were observed for the outcome of ICU mortality.10
On the other hand, several studies have not reported an association between ICU capacity strain and hospital mortality.11-15 Using the APACHE clinical information system for 200,499 patients admitted to 108 ICUs, Iwashyna et al did not find a difference in rates of hospital mortality or transfer to another hospital on increasing census days, suggesting hospitals included in the study were able to scale up their functions to meet a wide range of strain conditions.14 Using data from the Project IMPACT database of 194 ICUs in 131 hospitals, Wagner et al found that elevations in ICU capacity strain (measured by ICU census, admissions, and acuity) were associated with increased odds of ICU readmission within 72 hours, but were not associated with increased odds of in-hospital death or decreased odds of discharge.15 Although the study found that higher levels of ICU capacity strain (comparing 95th percentile vs 5th percentile of all strain variables) resulted in a 6.3 hour reduction in expected ICU length of stay (95% CI, 5.3-7.3 hours), the results of this study contrast the findings by Chrusch et al7 by suggesting that physician decision-making during times of high ICU capacity strain may actually result in better ICU efficiency rather than rationing care to the detriment of some patients.15
Studies are more consistent in their findings that increased ICU capacity strain is associated with increased risk of unplanned readmissions. As previously discussed, Chrusch et al found that increased ICU occupancy was associated with an increased risk of readmission within 7 days.7 In the Wagner et al study, for every one-unit increase in census, the odds of 72-hour readmission increased 3% (P = 0.030). Similarly, for every 10% increase in patient acuity among those already admitted to the ICU, there was a 5% increased odds of ICU readmission (P = 0.019).15 Similar findings were seen when the primary ICU capacity strain variable was ICU bed availability,9 and were even more pronounced in a study based in a neurosciences critical care unit, in which patients discharged on days with > 10 admissions had more than twice the odds of unplanned ICU readmission within 72 hours (OR, 2.34; 95% CI, 1.27-4.34) compared to those discharged on days with less than nine admissions.16
ICU capacity strain may also have an effect on end-of-life decision-making in the ICU. A multicenter study in Alberta, Canada, found that more patients had their goals of care changed from resuscitative care to comfort care when zero ICU beds were available compared with when > two ICU beds were available(14.9% vs. 8.5%; P < 0.01).13 In fact, changes in goals of care from resuscitation to comfort were 89.6% more likely when zero ICU beds were available compared to > two (95% CI, 24.9%-188.0%).13 Interestingly, hospital mortality in this study was similar for patients, regardless of the available ICU beds, implying that the available ICU beds at the time of a patient’s clinical deterioration affects processes of care, but not necessarily hospital mortality.13 Similarly, a retrospective cohort study of > 9,000 patients in the United States found that various measurements of ICU capacity strain were associated with shorter time to do-not-resuscitate orders and, subsequently, death among patients with limitations to life-sustaining therapy in the ICU. There was no association between ICU capacity strain and time to death for those without limitations in therapy.17
HOW CAN WE REDUCE THE NEGATIVE CONSEQUENCES OF ICU CAPACITY STRAIN?
Recognizing the outcomes associated with ICU capacity strain is important, but next steps will be to determine how to reduce the negative consequences of strain. One approach is to improve average length of stay in the ICU, which can subsequently improve throughput and quality of care. Several well-known interventions include spontaneous breathing trials, daily interruption of sedation, and early physical and occupational therapy. If applied successfully and consistently, they can significantly affect an ICU census.18 If findings from these interventional trials are applied to a hypothetical ICU, for example, “wake up and breathe” protocols (i.e., paired daily interruption of sedation and spontaneous breathing trial) could allow ICUs to care for 847 more patients per year by decreasing average ICU length of stay, without the need to build additional beds or increase capacity.18
Another consideration would be to identify certain structural characteristics in the ICU that may have an effect on how each one responds to capacity strain and evaluate their outcomes. In other words, what types of ICUs adapt better to increasing strain without sacrificing quality of care, and can their organizational qualities be applied to other, less flexible ICUs? For example, Gabler et al reported that standardized census in the ICU was associated with increased odds of in-hospital and ICU death overall, but also found that this effect was greater in ICUs with closed physician staffing models compared to those with open physician staffing models (OR, 1.07; 95% CI, 1.02-1.12 vs. OR, 1.01; 95% CI, 0.99-1.03, respectively).10 The authors hypothesized that this finding could be explained by the observation that during periods of high capacity strain, less time is allocated to patients in a closed staffing system with a limited number of providers as opposed to being more widely distributed in an open system. Indeed, others have confirmed that during times of high strain, less time is devoted on rounds per patients.5
It is worth mentioning, however, that the relationship between patient volume and mortality is not as straightforward as one may think. In a study of 39 trauma centers, a highly specialized field, Marcin et al reported that higher monthly trauma admission volume was associated with an increased risk of 30-day readmission for elderly trauma patients, and higher annual trauma volume was significantly associated with an increased risk of readmission for nonelderly adult trauma patients.19 This is despite the finding that elderly trauma patients treated at centers with high annual trauma volume had lower mortality (OR for each 100 admissions, 0.79; 95% CI, 0.71-0.87).19 Similarly, a multicenter study of European ICUs found that a higher ICU occupancy rate was associated with increased hospital mortality (OR, 1.324; 95% CI, 1.133-1.548), but annually, hospital mortality for all admitted ICU patients decreased by 3.4% for every five extra patients treated per bed per year in the same ICU, and by 17% for every five extra high-risk patients (Simplified Acute Physiological Score II above median for those staying longer than 47 hours).20
Findings from all these studies initially appear contradictory, especially in light of prior data showing that high-intensity staffing models reduce hospital and ICU mortality and length of stay.21 Although closed ICUs provide high-quality care under average conditions, they may experience more adverse outcomes under high-demand conditions. As such, they may need additional strategies and policies to balance workload demand and supply compared to ICUs with different staffing models. Although regionalization of care can result in increased capacity strain for a particular center at certain times with possibly untoward consequences, it can still improve patient survival overall by concentrating patients in centers where there are concentrated resources and specialized expertise. In other words, between-hospital volume and within-hospital volume are separate and distinct hospital-level characteristics that can affect patient care differently. Further efforts to reduce the effects of ICU capacity strain and/or enhance an ICU’s ability to adapt during periods of strain may need to focus on intensivist-directed ICUs in high-volume hospitals.
SUMMARY
ICU capacity strain results from an imbalance between demand for ICU high-quality care and supply. As such, it’s disconcerting to consider that the outcome for a patient admitted to the ICU is related not only to the patient’s own severity of illness, but also dependent on the state of the ICU itself in terms of other admissions, acuity, and census. There is still much work to be done with regard to defining a valid and optimal measure for ICU capacity strain. The hope, however, is that development of a valid metric will facilitate further outcomes research in terms of identifying effective ICU structural characteristics that can withstand strain, improving the transparency and equity with which critical care is rationed during times of strain, and studying the effects of interventions that could improve outcomes of ICU capacity strain.
REFERENCES
- Halpern NA, Pastores SM. Critical care medicine in the United States 2000-2005: An analysis of bed numbers, occupancy rates, payer mix, and costs. Crit Care Med 2010;38:65-71.
- Angus DC, Kelley MA, Schmitz RJ, et al. Current and projected workforce requirements for care of the critically ill and patients with pulmonary disease: Can we meet the requirements of an aging population? JAMA 2000;284:2762-2770.
- Halpern SD. ICU capacity strain and the quality and allocation of critical care. Curr Opin Crit Care 2011;17:648-657.
- Kerlin MP, Harhay MO, Vranas KC, et al. Objective factors associated with physicians’ and nurses’ perceptions of intensive care unit capacity strain. Ann Am Thorac Soc 2014;11:167-172.
- Brown SE, Rey MM, Pardo D, et al. The allocation of intensivists’ rounding time under conditions of intensive care unit capacity strain. Am J Respir Crit Care Med 2014;190:831-834.
- Tarnow-Mordi WO, Hau C, Warden A, Shearer AJ. Hospital mortality in relation to staff workload: A 4-year study in an adult intensive care unit. Lancet 2000;356:185-189.
- Chrusch CA, Olafson KP, McMillan PM, et al. High occupancy increases the risk of early death or readmission after transfer from intensive care. Crit Care Med 2009;37:2753-2758.
- Robert R, Reignier J, Tournoux-Facon C, et al. Refusal of intensive care unit admission due to a full unit: Impact on mortality. Am J Respir Crit Care Med 2012;185:1081-1087.
- Town JA, Churpek MM, Yuen TC, et al. Relationship between ICU bed availability, ICU readmission, and cardiac arrest in the general wards. Crit Care Med 2014;42:2037-2041.
- Gabler NB, Ratcliffe SJ, Wagner J, et al. Mortality among patients admitted to strained intensive care units. Am J Respir Crit Care Med 2013;188:800-806.
- Strauss MJ, LoGerfo JP, Yeltatzie JA, et al. Rationing of intensive care unit services: An everyday occurrence. JAMA 1986;255:1143-1146.
- Harrison DA, Lertsithichai P, Brady AR, et al. Winter excess mortality in intensive care in the UK: An analysis of outcome adjusted for patient case mix and unit workload. Intensive Care Med 2004;30:1900-1907.
- Stelfox HT, Hemmelgarn BR, Bagshaw SM, et al. Intensive care unit bed availability and outcomes for hospitalized patients with sudden clinical deterioration. Arch Intern Med 2012;172:467-474.
- Iwashyna TJ, Kramer AA, Kahn JM. Intensive care unit occupancy and patient outcomes. Crit Care Med 2009;37:1545-1557.
- Wagner J, Gabler NB, Ratcliffe SJ, et al. Outcomes among patients discharged from busy intensive care units. Ann Intern Med 2013;159;447-455.
- Baker DR, Pronovost PJ, Morlock LL, et al. Patient flow variability and unplanned readmissions to an intensive care unit. Crit Care Med 2009;37:2882-2887.
- Hua M, Halpern SD, Gabler NB, Wunsch H. Effect of ICU strain on timing of limitations in life-sustaining therapy and on death. Intensive Care Med 2016 Feb 9 [Epub ahead of print].
- Howell MD. Managing ICU throughput and understanding ICU census. Curr Opin Crit Care 2011;17:626-633.
- Marcin JP, Romano PS. Impact of between-hospital volume and within-hospital volume on mortality and readmission rates for trauma patients in California. Crit Care Med 2004;32:1477-1483.
- Iapichino G, Gattinoni L, Radrizzani D, et al. Volume of activity and occupancy rate in intensive care units. Association with mortality. Intensive Care Med 2004;30:290-297.
- Pronovost PJ, Angus DC, Dorman T, et al. Physician staffing patterns and clinical outcomes in critically ill patients: A systematic review. JAMA 2002;288:2151-2162.
ICUs are faced with the challenge of continuing care delivery under conditions of increasing strain that's tough to get a handle on.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.