View Article PDF

Rapid response systems (RRS) were developed to improve patient safety1 and consist of an afferent arm for identification of deteriorating hospital patients and triggering of a clinical review, and an efferent arm that responds to patient deterioration by rapidly deploying a team led by a physician or a nurse with appropriate critical care expertise.2 RRS have been widely adopted throughout Australia and New Zealand to identify deteriorating patients with the aim to prevent serious adverse events such as cardiac arrest.3 Operating a RRS has financial and resource implications, although they are hard to evaluate in the context of wider hospital costs. We believe that RRC activations may reduce the overall healthcare expenditure by expediting optimal delivery of care to the deteriorating patient.

RRS are increasingly detecting patients who are actively dying and have therefore become a surrogate way of providing end-of-life care (EOLC) to such patients.4 It is also evident that not all rapid response call (RRC) activations are clinically appropriate and may not necessarily reflect a person’s wishes.5,6 Older people are particularly at risk of such avoidable RRCs as they represent a substantial proportion of the inpatient population7 and have multiple comorbidities, placing them at a higher risk of clinical deterioration.8,9 These patients may not wish to have life-prolonging measures performed but have not had the opportunity to have that adequately communicated to the treating team.10

There are significant knowledge gaps with regards to erroneously activating RRCs in patients with clearly documented treatment limitations specifying that they are not for further RRCs. Such erroneous calls are avoidable. This has the potential to cause not only distress to the patients and their relatives but could also lead to increased costs to healthcare settings. Previous studies have specifically excluded such patients who were “not for a RRC” because their needs and management were “complex”.11 In addition to contravening patients’ wishes, avoidable RRCs (ARRC) are also associated with time and financial costs. This includes obvious expenses such as equipment usage and staff wages but also the opportunity cost of removing clinical staff from attending to duties they would otherwise be performing. Despite this, no studies have examined the financial cost associated with ARRC.12

We aimed to retrospectively identify the incidence of ARRC in older people and explore the time and financial costs associated with them.

Methodology

We retrospectively reviewed the medical records of all RRC activations in people 80 years of age and over admitted to Frankston Hospital (Victoria, Australia) from 1 March 2015 to 31 August 2015 who had at least one avoidable RRC (ARRC). Frankston hospital is the major hospital within Peninsula Health serving the metropolitan and regional areas of Mornington Peninsula. Frankston hospital has 454 beds, a 15-bedded ICU and had 61,711 admissions and discharges in the last 12 months.

Patient selection

We have included patients over 80 years, as these patients are the most vulnerable and are likely to have limitations of medical treatment (LOMT).13 Hence, we chose them as the target study group.

Classification of appropriateness

A resuscitation care plan is discussed with all appropriate patients and recorded on a dedicated document (see Supplementary Figure 1). This document indicates the outcome of the discussion between doctors, patient and occasionally carers regarding the nature of appropriate interventions in the event of clinical deterioration or cardiac arrest. Any plan to limit resuscitation is discussed with senior medical staff directly responsible for the patient’s management during that hospital admission.

We defined ARRC as a RRC activation occurring despite a patient having prior documentation limiting their medical interventions to have no further RRC activations. The decision that a patient was not for further RRCs was made as a direct consequence of earlier RRCs, when resuscitation plans were revised to “treatment aimed at symptom management” and “not for further RRC”. The decision that a patient should not have any further RRCs was made by the patient or their relatives, treating team, and/or by the RRC team in conjunction with the treating team. Once it was documented that a patient should not have any further RRCs, any further RRC during the same admission was defined as avoidable.

Rapid response system (RRS) in Frankston Hospital

The RRS in Frankston Hospital was established in 1999. The rapid response team (RRT) in this hospital includes a critical care liaison nurse (CCLN) (between 8:30 and 22:00) or ICU nurse (between 22:00 and 08:30), an ICU registrar and a medical registrar. If a “Code Blue” is called, an additional ICU nurse and a coronary-care trained nurse also attend. An anaesthetist is also available if required for airway management.

Approximately 8–10 local staff members are usually present during a RRC including: 2–3 bedside nurses; ward in-charge nurses; home-team doctors (which may include a hospital medical officer or registrar, an intern and a consultant); and a patient services manager (PSM).

End-of-life care in Peninsula Health

“Medical Futility” in the study hospital is established by the admitting senior clinicians at the time of admission to hospital in patients who are in a persistent vegetative state or who are terminally ill. This process usually takes a shared decision approach. A ‘Care of the Dying Patient’ (CDP) pathway may be implemented by the treating team consultant or multidisciplinary team decision for such patients in whom death is deemed to be inevitable. The CDP pathways are algorithms designed by the palliative care team, and are aimed at managing symptoms and ensuring patients are made comfortable. The CDP pathway is reassessed every three days if required.

Data collection and outcomes

The date, time and clinical indication for RRC activation was retrieved from CCLN database and confirmed against the patients’ electronic medical records. The database maintained by the dedicated CCLN team captures every RRC and subsequent reviews of involved patients. We extracted basic demographic data, RRC clinical trigger indication, staff in attendance, duration of RRC and equipment used from the electronic medical record.

We defined an ‘in-hours’ RRC as that occurring between the 8:30 and 17:00 and RRCs occurring outside of this time-period as ‘out-of-hours’. The time to RRC activation was calculated as the time from amended resuscitation form documentation requesting no further RRC activations to the next RRC activation.

Costings

The costings were classified into financial costs (usage of investigations, equipment and management) and opportunity costs (defined as staff duration of attendance). Other less-tangible costs (such as emotional costs) can be estimated; however, this is difficult and at times controversial. Therefore, we have limited our analysis to include only estimated financial and opportunity costs for all patients who had an ARRC. The reference year of 2015 was used to calculate the costs associated with each resource use item. All costs are presented in Australian dollars (AUD) and evaluated from the perspective of a health service provider. The medical records relating to each inappropriate RRC were reviewed and costs of an episode were summed for each individual patient based on actual resource use (Supplementary Table 1A and 1B).

Investigation, management and equipment costs

The occurrence of each investigation was recorded from the medical record and assigned a corresponding standard unit price based on the 2015 Medicare Benefit Schedule of the Australian government14 and the AMA list of Medical Services.15 See Supplementary Table 1A for a list of the common equipment and investigations costs.

Workforce costs

The hourly wages for staff were calculated based on that published in 2015 sources16–19 (See Supplementary Table 1B). Although a proportion of the nursing staff are employed on a casual basis (who are paid 25% more than a regular employee for their level of experience), we assumed all nurses to be contracted by the hospital. The doctors and nurses have varied clinical experience. We included their minimum, maximum and average wages in costing for both nurses and doctors.

Analysis

We used N (%), mean (standard deviation) and median (interquartile range) as appropriate to describe the data. We calculated the cost of each individual ARRC based on the usage of investigations, equipment and management (defined as financial costs) and staff duration of attendance (defined as opportunity costs). All statistical analysis was conducted using SPSS (IBM, v.20) statistical package.

Ethical approval

This project was reviewed and approved as audit activity by Research Governance of Peninsula Health (ref. QA/15/PH/23). Informed consent was not required as this study was a retrospective chart review that required no patient contact.

Results

A total of 1,974 patients ≥80 years of age were admitted to Frankston hospital during the study period. A total of 255 RRC occurred in 186 patients (Figure 1). Of these, a total of 25 (9.8%) RRCs that involved 18 individual patients were ARRC (mean age = 85.6 years, SD = 3.94). Table 1 describes the characteristics of patients who had a ARRC. Six patients had more than one ARRC. Most patients (88%) were admitted under a medical team with “functional decline” being the most common (25%) presenting complaint on admission.

Figure 1: Flowchart to describe that about 10% of the RRC activations were avoidable.

c

Table 1: Characteristics of avoidable rapid response call (ARRC) activations.

*Other refers to RRCs due to staff concern, including: uncontrolled or severe pain (including chest pain).
ARRC were not trigged due to other RRC triggers as listed in Supplementary Table 2.
N = number; IQR = interquartile range; UTI = urinary tract infection; COPD = chronic obstructive pulmonary disease; APO = acute pulmonary oedema; NSTEMI – non-ST elevation myocardial infarction; RRC; RR = Respiratory rate; SaO2 = blood oxygen saturation; SBP = Systolic Blood Pressure; HR = Heart Rate.

A majority (88%) of ARRCs occurred during weekdays, with similar proportions occurring in (52%) and out (48%) of business hours. The triggers for ARRC activations were: SaO2 <90% (20%); SBP <90mmHg (20%); respiratory distress (12%); tachycardia (HR >130 bpm) (12%); tachypnoea (respiratory rate >36) (8%); SBP >180 mmHg (8%); bradycardia (HR <40 bpm) (4%) and seizure (4%) (Table 1). The median duration of each RRC was 22 min (IQR 13–40 minutes).

The median time to ARRCs was within three days (IQR—one day to six days) from the time that patients were documented to be not for further RRCs. Most patients (94.4%) who had a ARRC died in hospital within three days or were transferred directly to palliative care. Twelve out of the 18 patients (66.7%) were already on CDP with syringe-driver infusions of morphine and midazolam at the time of their ARRC; however, palliative care team utilisation was seen in only in 40% of patients.

Costings

The calculated financial costs (usage of investigations, equipment and management) and estimated the opportunity costs (defined as staff duration of attendance) for all patients who had an ARRC. Table 2 describes the investigation, management and staff costs associated with ARRC. The total costs of investigations and management was AUD $2,267.01, and maximum workforce cost was AUD $3,861.55. The total financial cost of all 25 ARRC was AUD $6,128.56 (Table 3 and Supplementary Tables 3A and 3B).

Table 2: Investigation, management and workforce costs during avoidable RRC (ARRC) activations. The most common investigations, all necessary piece of equipment and staff member at the ARRC identified and costed.

*Medications included bronchodilators, steroids, analgesia (excluding opiates), antiepileptic agents, antiarryhtmic agents, diuretics, metaraminol
FBE = full blood exam; UEC = electrolytes, urea, creatinine; Coags = coagulation profile;
LFT = liver function tests; CMP = calcium magnesium phosphate; ECG = electrocardiogram;
CT = computed tomography scan; IV = intravenous.

Table 3: The doctors and nurses involved in the ARRC along with their time (in minutes) spent for each ARRC is described. The cost of investigations performed during each ARRC is included. Based on this, the maximum opportunity wages were calculated using Supplementary Table 1B.

c

*Based on maximum workforce wages. We have also calculated the cost based on minimum and average workforce wages. Refer Supplementary Tables 3A and 3B.

ARRC = Avoidable Rapid Response Call; Med = medical; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; ANUM = associate nurse unit manager; PSM = patient service manager; Cons = consultant.

Discussion

We found that approximately 10% of RRC at our hospital were avoidable, with many of these patients being subjected to non-beneficial investigations and interventions. This poses significant challenges for both the afferent and efferent limbs2 of the RRS. This was associated with a small but important cost (that includes both financial and opportunity costs). This is the first paper to explore the incidence and financial costs of ARRC in the Australian inpatient setting.

ARRC in the current study were associated with a maximum cost of approximately AUD $6,128.56 in equipment, investigations, management and staff time. Although it is a relatively small figure, it reflects only the expenditure related to a single inappropriate aspect of patients’ care over a six-month period, and at a single centre. It is also likely that this figure represents only a fraction of the proportion of healthcare expenditure that is unnecessarily spent on inappropriate interventions for patients who require, or would prefer, only symptom management and comfort care.

Most ARRCs happened within three days from the time that patients were documented to be not for further RRCs. The median time for the ARRCs was 22 minutes, with some RRCs observed to be as long as 71 minutes. The reasons for this are unclear; however, discussions with primary teams and patients’ family take time. A previous study reported that limitation of medical therapy-based RRC often took considerably longer than other RRCs.11

A majority of patients in our study who had a ARRC were admitted under medical teams. The reasons why ARRC may be more common for medical than surgical patients are unclear.

Performing basic interventions such as administering supplemental oxygen or intravenous fluids, suctioning patients’ upper airways, securing intravenous access and ordering basic investigations (ECG, arterial blood gases, chest x-ray and routine bloods)20 have become an expectation for RRCs. These interventions come at a cost to the patient/family’s quality of life and experience, staffing resources and finances. Rather than being reactive, the RRTs should consider whether such interventions and investigations are appropriate or necessary in dying patients.

A recent multicentre study reported that 29.5% of patients with a LOMT had a RRC on the same day that they died.11 Similarly we found that most ARRC were in the final days of patient’s lives. This suggests that an important trigger for ARRC may be due to the urgent need for symptom control at the end of life.21 Other potential triggers could include a lack of clear documentation or other communication between patients, family and carer providers regarding goals of care and limitations of treatment.

Enhanced services such as effective management of physical, psychological and spiritual distress; timely and sensitive communication about appropriate goals of patient care; alignment of treatment with patient preferences; attention to families’ needs and concerns; planning for care transitions; and support for clinicians at the end of life could help reduce the number of ARRC and potentially improve patient and carer’s experiences.21–26

A recent review observed that LOMT discussions occurred more commonly than resuscitation interventions such as endotracheal intubation. This review also highlighted the fact that none of the studies reported the frequency of palliative care interventions implemented by the MET.25,27 There is good evidence that formal involvement of palliative care at the end-of-life improves patient and carer outcomes,23,24 however, such services appear to be underutilised.4,22 Likewise, patients who had ARRC activations were less likely to have palliative care consultations to those who did not.22 The reasons for this are unclear and warrant further investigation.

We believe that a multidisciplinary team approach should be adopted in managing these patients, with timely referral to palliative care teams. The concept of a palliative emergency team (PET) warrants further investigation. This may present a more cost-effective and appropriate alternative to RRCs for patients on end-of-life pathways, once the educational and behavior shaping components are optimised.

Even though clinicians should not be discouraged from activating RRCs, ARRC should be avoided to maximise comfort and dignity in the last days of life.

Demystifying definitions and terminology such as ‘palliative care’, ‘not for resuscitation’, ‘advance care planning’ and ‘advance health directive’ is prudent to aid communication and standardising practice.28

Although most clinicians working in acute facilities will participate in providing end-of-life care to patients, it appears that the management of this care varies due to knowledge gaps, skill and experience among the clinicians caring for the patient.29,30 Active education, using simulation and role playing will be beneficial for all healthcare professionals.

By reducing the number of ARRC, there may be some financial support available to fund these under-utilised strategies (eg, additional resources in palliative care).4,22

Strengths and limitations

To our knowledge, this is the first study to directly evaluate the incidence and the costs associated with ARRC. A strength of this study was, given the small sample size, we were able to micro-cost individual times and associated financial costs of each individual ARRC. However, the study also has a few limitations. The small sample size at a single site and retrospective design limits broader generalisation of our results. Given the retrospective design of our study, we are unable to exactly identify the reasons why ARRC occurred, despite the documentation to confirm the patients should only be treated symptomatically and a higher incidence of such calls in patients admitted for medical conditions. Some patients had more than one ARRC and the reasons for this remain unclear due to the retrospective nature of our study. We relied upon documentation to identify ARRC. This study was unable to quantify other costs associated with ARRC such as emotional and physical discomfort of the patient and their relatives’, and the negative impact on the care of other patients. It was also not possible to retrospectively investigate the specific wishes of patients and their families using available medical records.

Conclusion

Avoidable RRC (ARRC) occur in 10% of RRCs for hospital inpatients aged over 80 years, and are associated with potentially increased time and financial costs. Further research is required to understand the triggers to these ARRC and to study effectiveness of new strategies designed to reduce the burden of ARRC on patients, carers and hospital staff.

Supplementary Figure 1: Example of a Peninsula Health Resuscitation Plan defining avoidable rapid response calls (ARRC).  

c

We defined ARRC as RRC activations despite clear documentation confirming not for further RRCs. Patients were included only if a resuscitation plan document was completed with the Patient is for treatment aimed at “Symptom Management” box ticked and additional hand-written documentation of “not for further MET calls”.

Supplementary Table 1A and 1B. Items and item costs used to calculate financial cost of rapid response calls (RRCs). Table 1A: Most common blood and radiological investigations. Table 1B: Costing of workforce who were involved in the RRC.

*/**AMA List of Medical Services and Fees, Medical Benefit Scheme.

*Dorevitch pathology (Peninsula Health pathology provider).

**MIA radiology.

IV = intravenous; CT = computed tomography scan; ECG = electrocardiogram.

Supplementary Table 1B: Costing of workforce who attend and look after the RRC. Wages are represented as entry level (minimum) and maximum.

CCLN = Critical Care Liaison Nurse.

Doctors wages based on Award as of December 2015.

http://admin.anfvic.asn.au/multiversions/2776/FileName/RN_RM_PubSect2014.pdf (assessed 18 January 2017).

RRC = Rapid Response Call; ICU = intensive care unit; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; PSM = patient service manager; Cons = consultant.

Supplementary Table 3B: The doctors and nurses involved in the RRC along with their time (in minutes) spent for each RRC is described. The cost of investigations performed during each RRC is included. Based on this, the average workforce wages calculated using Supplementary Table 1B.

c

RRC = Rapid Response Call; ICU = intensive care unit; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; PSM = patient service manager; Cons = consultant.

Summary

Abstract

Rapid response calls (RRCs) are designed to appropriately manage clinical deterioration. However, not all RRCs are appropriate due to medical futility or the patient s wishes. Incidence and costs associated with avoidable-RRC (ARRC) remain underexplored.

Aim

The aim of this study was to describe the incidence and costs of ARRC activations in older patients.

Method

We retrospectively reviewed RRCs in patients aged 580 years over six months. We defined ARRC as RRC activations despite clear documentation confirming not for further RRCs. Data on investigations, equipment and management of each ARRC were analysed. We then micro-costed each ARRC using standard references.

Results

Ten percent (25/255) of RRCs were ARRC (mean age 85.6 years) with most patients (88%) admitted under medical teams. Median duration of ARRC was 22 minutes (IQR 7-38 minutes). Palliative care services were underutilised (40%). Most patients (94.4%) died soon after ARRC. The costs for investigations, equipment and management was AUD $2,267.01, opportunity costs were AUD $3,861.55, with a grand total cost of AUD $6,128.56.

Conclusion

ARRC, noted in 10% of RRCs, are associated with increased time and financial costs. Further research is required to better understand ARRC triggers to reduce the burden of ARRC on patients, carers and hospital staff.

Author Information

- Ashwin Subramaniam, Intensivist, Intensive Care, Frankston Hospital, Australia; Cameron Green, Intensive Care, Frankston Hospital, Australia; Mohammad Omair, Department of Medicine, Frankston Hospital, Australia; Lionel Soh, Department of

Acknowledgements

We gratefully acknowledge the assistance of Dr Christopher Moran (Geriatrician) for his valuable suggestions in preparing this manuscript. We would also like to thank the Critical Care Liaison Nurse service (in particular, Jo Molloy and Naomi Pratt) of Peninsula Health for the use of their patient databases, and for their commitment to maintaining the rapid response system.

Correspondence

Dr Ashwin Subramaniam, Intensivist, Intensive Care, Frankston Hospital, Hastings Road, Frankston 3199, Australia.

Correspondence Email

catchdrash@yahoo.com

Competing Interests

Nil.

  1. Lee A, Bishop G, Hillman KM, et al. The Medical Emergency Team. Anaesth Intensive Care. 1995; 23(2):183–6.
  2. Jones DA, DeVita MA, Bellomo R. Rapid-response teams. N Engl J Med. 2011; 365(2):139–46.
  3. Hillman KM, Chen J, Jones D. Rapid response systems. Med J Aust. 2014; 201(9):519–21.
  4. Hillman K. Dying safely. Int J Qual Health Care. 2010; 22(5):339–40.
  5. Hillman K, Chen J, Cretikos M, et al. Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. Lancet. 2005; 365(9477):2091–7.
  6. Buist MD, Moore GE, Bernard SA, et al. Effects of a medical emergency team on reduction of incidence of and mortality from unexpected cardiac arrests in hospital: preliminary study. BMJ. 2002; 324(7334):387–90.
  7. Ruiz M, Bottle A, Long S, et al. Multi-Morbidity in Hospitalised Older Patients: Who Are the Complex Elderly? PLoS One. 2015; 10(12):e0145372.
  8. Karlamangla A, Tinetti M, Guralnik J, et al. Comorbidity in older adults: nosology of impairment, diseases, and conditions. J Gerontol A Biol Sci Med Sci. 2007; 62(3):296–300.
  9. Guralnik JM. Assessing the impact of comorbidity in the older population. Ann Epidemiol. 1996; 6(5):376–80.
  10. Parr JD, Zhang B, Nilsson ME, et al. The influence of age on the likelihood of receiving end-of-life care consistent with patient treatment preferences. J Palliat Med. 2010; 13(6):719–26.
  11. Jones DA, Bagshaw SM, Barrett J, et al. The role of the medical emergency team in end-of-life care: a multicenter, prospective, observational study. Crit Care Med. 2012; 40(1):98–103.
  12. Subramaniam A, Botha J, Tiruvoipati R. The limitations in implementing and operating a rapid response system. Intern Med J. 2016; 46(10):1139–45.
  13. Smith RL, Hayashi VN, Lee YI, et al. The medical emergency team call: a sentinel event that triggers goals of care discussion. Crit Care Med, 2014; 42(2):322–327.
  14. Medicare Benefits Schedule Book. In: Ageing AGDoHa, editor.: Canberra; 2015.
  15. AMA. List of Medical Services and Fees: Australian Meical Association Limited; 2015.
  16. Nurses And Midwives (Victorian Public Health Sector) (Single Interest Employers) Enterprise Agreement 2012–2016, (2014).
  17. AMA. Rates of Pay: Doctors in Training 2015 [18/01/2017]. Available from: http://amavic.com.au/page/Doctors_in_Training/What_is_the_AMA_doing_for_you/Rates_of_Pay_Doctors_in_Training/
  18. Federation AN. Quick Reference Pay Rates for Registered Nurses and Midwives 2012–2016 Agreement - Public Sector 2014.
  19. AMA Victoria - Victorian Public Health Sector Medical Specialists Enterprise Agreement 2013, (2013).
  20. Ranji SR, Auerbach AD, Hurd CJ, et al. Effects of rapid response systems on clinical outcomes: systematic review and meta-analysis. J Hosp Med. 2007; 2(6):422–32.
  21. Qaseem A, Snow V, Shekelle P, et al. Evidence-based interventions to improve the palliative care of pain, dyspnea, and depression at the end of life: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2008; 148(2):141–6.
  22. Downar J, Barua R, Rodin D, et al. Changes in end of life care 5 years after the introduction of a rapid response team: a multicentre retrospective study. Resuscitation. 2013; 84(10):1339–44.
  23. Lorenz KA, Lynn J, Dy SM, et al. Evidence for improving palliative care at the end of life: a systematic review. Ann Intern Med. 2008; 148(2):147–59.
  24. Lynn J, Schall MW, Milne C, et al. Quality improvements in end of life care: insights from two collaboratives. Jt Comm J Qual Improv. 2000; 26(5):254–67.
  25. Tan LH, Delaney A. Medical emergency teams and end-of-life care: a systematic review. Crit Care Resusc. 2014; 16(1):62–8.
  26. Aslakson RA, Curtis JR, Nelson JE. The changing role of palliative care in the ICU. Crit Care Med. 2014; 42(11):2418–28.
  27. Vazquez R, Gheorghe C, Grigoriyan A, et al. Enhanced end-of-life care associated with deploying a rapid response team: a pilot study. J Hosp Med. 2009; 4(7):449–52.
  28. Care ACoSaQiH. safety and quality of end-of-life care in acute hospitals: a background paper: ACSQHC, 2013. Sydney, Australia2013. p. 52.
  29. Parish K, Glaetzer K, Grbich C, et al. Dying for attention: palliative care in the acute setting. Aust J Adv Nurs. 2006; 24(2):21–5.
  30. Le BH, Watt JN. Care of the dying in Australia’s busiest hospital: benefits of palliative care consultation and methods to enhance access. J Palliat Med. 2010; 13(7):855–60.
  31. Federation AN. Quick Reference Pay Rates Registered Nurses and Midwives 2012–2016 Agreement -PUBLIC SECTOR 2014 [18/01/2017]. Available from: http://admin.anfvic.asn.au/multiversions/2776/FileName/RN_RM_PubSect2014.pdf

For the PDF of this article, contact
communications@nzma.org.nz

View Article PDF

Rapid response systems (RRS) were developed to improve patient safety1 and consist of an afferent arm for identification of deteriorating hospital patients and triggering of a clinical review, and an efferent arm that responds to patient deterioration by rapidly deploying a team led by a physician or a nurse with appropriate critical care expertise.2 RRS have been widely adopted throughout Australia and New Zealand to identify deteriorating patients with the aim to prevent serious adverse events such as cardiac arrest.3 Operating a RRS has financial and resource implications, although they are hard to evaluate in the context of wider hospital costs. We believe that RRC activations may reduce the overall healthcare expenditure by expediting optimal delivery of care to the deteriorating patient.

RRS are increasingly detecting patients who are actively dying and have therefore become a surrogate way of providing end-of-life care (EOLC) to such patients.4 It is also evident that not all rapid response call (RRC) activations are clinically appropriate and may not necessarily reflect a person’s wishes.5,6 Older people are particularly at risk of such avoidable RRCs as they represent a substantial proportion of the inpatient population7 and have multiple comorbidities, placing them at a higher risk of clinical deterioration.8,9 These patients may not wish to have life-prolonging measures performed but have not had the opportunity to have that adequately communicated to the treating team.10

There are significant knowledge gaps with regards to erroneously activating RRCs in patients with clearly documented treatment limitations specifying that they are not for further RRCs. Such erroneous calls are avoidable. This has the potential to cause not only distress to the patients and their relatives but could also lead to increased costs to healthcare settings. Previous studies have specifically excluded such patients who were “not for a RRC” because their needs and management were “complex”.11 In addition to contravening patients’ wishes, avoidable RRCs (ARRC) are also associated with time and financial costs. This includes obvious expenses such as equipment usage and staff wages but also the opportunity cost of removing clinical staff from attending to duties they would otherwise be performing. Despite this, no studies have examined the financial cost associated with ARRC.12

We aimed to retrospectively identify the incidence of ARRC in older people and explore the time and financial costs associated with them.

Methodology

We retrospectively reviewed the medical records of all RRC activations in people 80 years of age and over admitted to Frankston Hospital (Victoria, Australia) from 1 March 2015 to 31 August 2015 who had at least one avoidable RRC (ARRC). Frankston hospital is the major hospital within Peninsula Health serving the metropolitan and regional areas of Mornington Peninsula. Frankston hospital has 454 beds, a 15-bedded ICU and had 61,711 admissions and discharges in the last 12 months.

Patient selection

We have included patients over 80 years, as these patients are the most vulnerable and are likely to have limitations of medical treatment (LOMT).13 Hence, we chose them as the target study group.

Classification of appropriateness

A resuscitation care plan is discussed with all appropriate patients and recorded on a dedicated document (see Supplementary Figure 1). This document indicates the outcome of the discussion between doctors, patient and occasionally carers regarding the nature of appropriate interventions in the event of clinical deterioration or cardiac arrest. Any plan to limit resuscitation is discussed with senior medical staff directly responsible for the patient’s management during that hospital admission.

We defined ARRC as a RRC activation occurring despite a patient having prior documentation limiting their medical interventions to have no further RRC activations. The decision that a patient was not for further RRCs was made as a direct consequence of earlier RRCs, when resuscitation plans were revised to “treatment aimed at symptom management” and “not for further RRC”. The decision that a patient should not have any further RRCs was made by the patient or their relatives, treating team, and/or by the RRC team in conjunction with the treating team. Once it was documented that a patient should not have any further RRCs, any further RRC during the same admission was defined as avoidable.

Rapid response system (RRS) in Frankston Hospital

The RRS in Frankston Hospital was established in 1999. The rapid response team (RRT) in this hospital includes a critical care liaison nurse (CCLN) (between 8:30 and 22:00) or ICU nurse (between 22:00 and 08:30), an ICU registrar and a medical registrar. If a “Code Blue” is called, an additional ICU nurse and a coronary-care trained nurse also attend. An anaesthetist is also available if required for airway management.

Approximately 8–10 local staff members are usually present during a RRC including: 2–3 bedside nurses; ward in-charge nurses; home-team doctors (which may include a hospital medical officer or registrar, an intern and a consultant); and a patient services manager (PSM).

End-of-life care in Peninsula Health

“Medical Futility” in the study hospital is established by the admitting senior clinicians at the time of admission to hospital in patients who are in a persistent vegetative state or who are terminally ill. This process usually takes a shared decision approach. A ‘Care of the Dying Patient’ (CDP) pathway may be implemented by the treating team consultant or multidisciplinary team decision for such patients in whom death is deemed to be inevitable. The CDP pathways are algorithms designed by the palliative care team, and are aimed at managing symptoms and ensuring patients are made comfortable. The CDP pathway is reassessed every three days if required.

Data collection and outcomes

The date, time and clinical indication for RRC activation was retrieved from CCLN database and confirmed against the patients’ electronic medical records. The database maintained by the dedicated CCLN team captures every RRC and subsequent reviews of involved patients. We extracted basic demographic data, RRC clinical trigger indication, staff in attendance, duration of RRC and equipment used from the electronic medical record.

We defined an ‘in-hours’ RRC as that occurring between the 8:30 and 17:00 and RRCs occurring outside of this time-period as ‘out-of-hours’. The time to RRC activation was calculated as the time from amended resuscitation form documentation requesting no further RRC activations to the next RRC activation.

Costings

The costings were classified into financial costs (usage of investigations, equipment and management) and opportunity costs (defined as staff duration of attendance). Other less-tangible costs (such as emotional costs) can be estimated; however, this is difficult and at times controversial. Therefore, we have limited our analysis to include only estimated financial and opportunity costs for all patients who had an ARRC. The reference year of 2015 was used to calculate the costs associated with each resource use item. All costs are presented in Australian dollars (AUD) and evaluated from the perspective of a health service provider. The medical records relating to each inappropriate RRC were reviewed and costs of an episode were summed for each individual patient based on actual resource use (Supplementary Table 1A and 1B).

Investigation, management and equipment costs

The occurrence of each investigation was recorded from the medical record and assigned a corresponding standard unit price based on the 2015 Medicare Benefit Schedule of the Australian government14 and the AMA list of Medical Services.15 See Supplementary Table 1A for a list of the common equipment and investigations costs.

Workforce costs

The hourly wages for staff were calculated based on that published in 2015 sources16–19 (See Supplementary Table 1B). Although a proportion of the nursing staff are employed on a casual basis (who are paid 25% more than a regular employee for their level of experience), we assumed all nurses to be contracted by the hospital. The doctors and nurses have varied clinical experience. We included their minimum, maximum and average wages in costing for both nurses and doctors.

Analysis

We used N (%), mean (standard deviation) and median (interquartile range) as appropriate to describe the data. We calculated the cost of each individual ARRC based on the usage of investigations, equipment and management (defined as financial costs) and staff duration of attendance (defined as opportunity costs). All statistical analysis was conducted using SPSS (IBM, v.20) statistical package.

Ethical approval

This project was reviewed and approved as audit activity by Research Governance of Peninsula Health (ref. QA/15/PH/23). Informed consent was not required as this study was a retrospective chart review that required no patient contact.

Results

A total of 1,974 patients ≥80 years of age were admitted to Frankston hospital during the study period. A total of 255 RRC occurred in 186 patients (Figure 1). Of these, a total of 25 (9.8%) RRCs that involved 18 individual patients were ARRC (mean age = 85.6 years, SD = 3.94). Table 1 describes the characteristics of patients who had a ARRC. Six patients had more than one ARRC. Most patients (88%) were admitted under a medical team with “functional decline” being the most common (25%) presenting complaint on admission.

Figure 1: Flowchart to describe that about 10% of the RRC activations were avoidable.

c

Table 1: Characteristics of avoidable rapid response call (ARRC) activations.

*Other refers to RRCs due to staff concern, including: uncontrolled or severe pain (including chest pain).
ARRC were not trigged due to other RRC triggers as listed in Supplementary Table 2.
N = number; IQR = interquartile range; UTI = urinary tract infection; COPD = chronic obstructive pulmonary disease; APO = acute pulmonary oedema; NSTEMI – non-ST elevation myocardial infarction; RRC; RR = Respiratory rate; SaO2 = blood oxygen saturation; SBP = Systolic Blood Pressure; HR = Heart Rate.

A majority (88%) of ARRCs occurred during weekdays, with similar proportions occurring in (52%) and out (48%) of business hours. The triggers for ARRC activations were: SaO2 <90% (20%); SBP <90mmHg (20%); respiratory distress (12%); tachycardia (HR >130 bpm) (12%); tachypnoea (respiratory rate >36) (8%); SBP >180 mmHg (8%); bradycardia (HR <40 bpm) (4%) and seizure (4%) (Table 1). The median duration of each RRC was 22 min (IQR 13–40 minutes).

The median time to ARRCs was within three days (IQR—one day to six days) from the time that patients were documented to be not for further RRCs. Most patients (94.4%) who had a ARRC died in hospital within three days or were transferred directly to palliative care. Twelve out of the 18 patients (66.7%) were already on CDP with syringe-driver infusions of morphine and midazolam at the time of their ARRC; however, palliative care team utilisation was seen in only in 40% of patients.

Costings

The calculated financial costs (usage of investigations, equipment and management) and estimated the opportunity costs (defined as staff duration of attendance) for all patients who had an ARRC. Table 2 describes the investigation, management and staff costs associated with ARRC. The total costs of investigations and management was AUD $2,267.01, and maximum workforce cost was AUD $3,861.55. The total financial cost of all 25 ARRC was AUD $6,128.56 (Table 3 and Supplementary Tables 3A and 3B).

Table 2: Investigation, management and workforce costs during avoidable RRC (ARRC) activations. The most common investigations, all necessary piece of equipment and staff member at the ARRC identified and costed.

*Medications included bronchodilators, steroids, analgesia (excluding opiates), antiepileptic agents, antiarryhtmic agents, diuretics, metaraminol
FBE = full blood exam; UEC = electrolytes, urea, creatinine; Coags = coagulation profile;
LFT = liver function tests; CMP = calcium magnesium phosphate; ECG = electrocardiogram;
CT = computed tomography scan; IV = intravenous.

Table 3: The doctors and nurses involved in the ARRC along with their time (in minutes) spent for each ARRC is described. The cost of investigations performed during each ARRC is included. Based on this, the maximum opportunity wages were calculated using Supplementary Table 1B.

c

*Based on maximum workforce wages. We have also calculated the cost based on minimum and average workforce wages. Refer Supplementary Tables 3A and 3B.

ARRC = Avoidable Rapid Response Call; Med = medical; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; ANUM = associate nurse unit manager; PSM = patient service manager; Cons = consultant.

Discussion

We found that approximately 10% of RRC at our hospital were avoidable, with many of these patients being subjected to non-beneficial investigations and interventions. This poses significant challenges for both the afferent and efferent limbs2 of the RRS. This was associated with a small but important cost (that includes both financial and opportunity costs). This is the first paper to explore the incidence and financial costs of ARRC in the Australian inpatient setting.

ARRC in the current study were associated with a maximum cost of approximately AUD $6,128.56 in equipment, investigations, management and staff time. Although it is a relatively small figure, it reflects only the expenditure related to a single inappropriate aspect of patients’ care over a six-month period, and at a single centre. It is also likely that this figure represents only a fraction of the proportion of healthcare expenditure that is unnecessarily spent on inappropriate interventions for patients who require, or would prefer, only symptom management and comfort care.

Most ARRCs happened within three days from the time that patients were documented to be not for further RRCs. The median time for the ARRCs was 22 minutes, with some RRCs observed to be as long as 71 minutes. The reasons for this are unclear; however, discussions with primary teams and patients’ family take time. A previous study reported that limitation of medical therapy-based RRC often took considerably longer than other RRCs.11

A majority of patients in our study who had a ARRC were admitted under medical teams. The reasons why ARRC may be more common for medical than surgical patients are unclear.

Performing basic interventions such as administering supplemental oxygen or intravenous fluids, suctioning patients’ upper airways, securing intravenous access and ordering basic investigations (ECG, arterial blood gases, chest x-ray and routine bloods)20 have become an expectation for RRCs. These interventions come at a cost to the patient/family’s quality of life and experience, staffing resources and finances. Rather than being reactive, the RRTs should consider whether such interventions and investigations are appropriate or necessary in dying patients.

A recent multicentre study reported that 29.5% of patients with a LOMT had a RRC on the same day that they died.11 Similarly we found that most ARRC were in the final days of patient’s lives. This suggests that an important trigger for ARRC may be due to the urgent need for symptom control at the end of life.21 Other potential triggers could include a lack of clear documentation or other communication between patients, family and carer providers regarding goals of care and limitations of treatment.

Enhanced services such as effective management of physical, psychological and spiritual distress; timely and sensitive communication about appropriate goals of patient care; alignment of treatment with patient preferences; attention to families’ needs and concerns; planning for care transitions; and support for clinicians at the end of life could help reduce the number of ARRC and potentially improve patient and carer’s experiences.21–26

A recent review observed that LOMT discussions occurred more commonly than resuscitation interventions such as endotracheal intubation. This review also highlighted the fact that none of the studies reported the frequency of palliative care interventions implemented by the MET.25,27 There is good evidence that formal involvement of palliative care at the end-of-life improves patient and carer outcomes,23,24 however, such services appear to be underutilised.4,22 Likewise, patients who had ARRC activations were less likely to have palliative care consultations to those who did not.22 The reasons for this are unclear and warrant further investigation.

We believe that a multidisciplinary team approach should be adopted in managing these patients, with timely referral to palliative care teams. The concept of a palliative emergency team (PET) warrants further investigation. This may present a more cost-effective and appropriate alternative to RRCs for patients on end-of-life pathways, once the educational and behavior shaping components are optimised.

Even though clinicians should not be discouraged from activating RRCs, ARRC should be avoided to maximise comfort and dignity in the last days of life.

Demystifying definitions and terminology such as ‘palliative care’, ‘not for resuscitation’, ‘advance care planning’ and ‘advance health directive’ is prudent to aid communication and standardising practice.28

Although most clinicians working in acute facilities will participate in providing end-of-life care to patients, it appears that the management of this care varies due to knowledge gaps, skill and experience among the clinicians caring for the patient.29,30 Active education, using simulation and role playing will be beneficial for all healthcare professionals.

By reducing the number of ARRC, there may be some financial support available to fund these under-utilised strategies (eg, additional resources in palliative care).4,22

Strengths and limitations

To our knowledge, this is the first study to directly evaluate the incidence and the costs associated with ARRC. A strength of this study was, given the small sample size, we were able to micro-cost individual times and associated financial costs of each individual ARRC. However, the study also has a few limitations. The small sample size at a single site and retrospective design limits broader generalisation of our results. Given the retrospective design of our study, we are unable to exactly identify the reasons why ARRC occurred, despite the documentation to confirm the patients should only be treated symptomatically and a higher incidence of such calls in patients admitted for medical conditions. Some patients had more than one ARRC and the reasons for this remain unclear due to the retrospective nature of our study. We relied upon documentation to identify ARRC. This study was unable to quantify other costs associated with ARRC such as emotional and physical discomfort of the patient and their relatives’, and the negative impact on the care of other patients. It was also not possible to retrospectively investigate the specific wishes of patients and their families using available medical records.

Conclusion

Avoidable RRC (ARRC) occur in 10% of RRCs for hospital inpatients aged over 80 years, and are associated with potentially increased time and financial costs. Further research is required to understand the triggers to these ARRC and to study effectiveness of new strategies designed to reduce the burden of ARRC on patients, carers and hospital staff.

Supplementary Figure 1: Example of a Peninsula Health Resuscitation Plan defining avoidable rapid response calls (ARRC).  

c

We defined ARRC as RRC activations despite clear documentation confirming not for further RRCs. Patients were included only if a resuscitation plan document was completed with the Patient is for treatment aimed at “Symptom Management” box ticked and additional hand-written documentation of “not for further MET calls”.

Supplementary Table 1A and 1B. Items and item costs used to calculate financial cost of rapid response calls (RRCs). Table 1A: Most common blood and radiological investigations. Table 1B: Costing of workforce who were involved in the RRC.

*/**AMA List of Medical Services and Fees, Medical Benefit Scheme.

*Dorevitch pathology (Peninsula Health pathology provider).

**MIA radiology.

IV = intravenous; CT = computed tomography scan; ECG = electrocardiogram.

Supplementary Table 1B: Costing of workforce who attend and look after the RRC. Wages are represented as entry level (minimum) and maximum.

CCLN = Critical Care Liaison Nurse.

Doctors wages based on Award as of December 2015.

http://admin.anfvic.asn.au/multiversions/2776/FileName/RN_RM_PubSect2014.pdf (assessed 18 January 2017).

RRC = Rapid Response Call; ICU = intensive care unit; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; PSM = patient service manager; Cons = consultant.

Supplementary Table 3B: The doctors and nurses involved in the RRC along with their time (in minutes) spent for each RRC is described. The cost of investigations performed during each RRC is included. Based on this, the average workforce wages calculated using Supplementary Table 1B.

c

RRC = Rapid Response Call; ICU = intensive care unit; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; PSM = patient service manager; Cons = consultant.

Summary

Abstract

Rapid response calls (RRCs) are designed to appropriately manage clinical deterioration. However, not all RRCs are appropriate due to medical futility or the patient s wishes. Incidence and costs associated with avoidable-RRC (ARRC) remain underexplored.

Aim

The aim of this study was to describe the incidence and costs of ARRC activations in older patients.

Method

We retrospectively reviewed RRCs in patients aged 580 years over six months. We defined ARRC as RRC activations despite clear documentation confirming not for further RRCs. Data on investigations, equipment and management of each ARRC were analysed. We then micro-costed each ARRC using standard references.

Results

Ten percent (25/255) of RRCs were ARRC (mean age 85.6 years) with most patients (88%) admitted under medical teams. Median duration of ARRC was 22 minutes (IQR 7-38 minutes). Palliative care services were underutilised (40%). Most patients (94.4%) died soon after ARRC. The costs for investigations, equipment and management was AUD $2,267.01, opportunity costs were AUD $3,861.55, with a grand total cost of AUD $6,128.56.

Conclusion

ARRC, noted in 10% of RRCs, are associated with increased time and financial costs. Further research is required to better understand ARRC triggers to reduce the burden of ARRC on patients, carers and hospital staff.

Author Information

- Ashwin Subramaniam, Intensivist, Intensive Care, Frankston Hospital, Australia; Cameron Green, Intensive Care, Frankston Hospital, Australia; Mohammad Omair, Department of Medicine, Frankston Hospital, Australia; Lionel Soh, Department of

Acknowledgements

We gratefully acknowledge the assistance of Dr Christopher Moran (Geriatrician) for his valuable suggestions in preparing this manuscript. We would also like to thank the Critical Care Liaison Nurse service (in particular, Jo Molloy and Naomi Pratt) of Peninsula Health for the use of their patient databases, and for their commitment to maintaining the rapid response system.

Correspondence

Dr Ashwin Subramaniam, Intensivist, Intensive Care, Frankston Hospital, Hastings Road, Frankston 3199, Australia.

Correspondence Email

catchdrash@yahoo.com

Competing Interests

Nil.

  1. Lee A, Bishop G, Hillman KM, et al. The Medical Emergency Team. Anaesth Intensive Care. 1995; 23(2):183–6.
  2. Jones DA, DeVita MA, Bellomo R. Rapid-response teams. N Engl J Med. 2011; 365(2):139–46.
  3. Hillman KM, Chen J, Jones D. Rapid response systems. Med J Aust. 2014; 201(9):519–21.
  4. Hillman K. Dying safely. Int J Qual Health Care. 2010; 22(5):339–40.
  5. Hillman K, Chen J, Cretikos M, et al. Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. Lancet. 2005; 365(9477):2091–7.
  6. Buist MD, Moore GE, Bernard SA, et al. Effects of a medical emergency team on reduction of incidence of and mortality from unexpected cardiac arrests in hospital: preliminary study. BMJ. 2002; 324(7334):387–90.
  7. Ruiz M, Bottle A, Long S, et al. Multi-Morbidity in Hospitalised Older Patients: Who Are the Complex Elderly? PLoS One. 2015; 10(12):e0145372.
  8. Karlamangla A, Tinetti M, Guralnik J, et al. Comorbidity in older adults: nosology of impairment, diseases, and conditions. J Gerontol A Biol Sci Med Sci. 2007; 62(3):296–300.
  9. Guralnik JM. Assessing the impact of comorbidity in the older population. Ann Epidemiol. 1996; 6(5):376–80.
  10. Parr JD, Zhang B, Nilsson ME, et al. The influence of age on the likelihood of receiving end-of-life care consistent with patient treatment preferences. J Palliat Med. 2010; 13(6):719–26.
  11. Jones DA, Bagshaw SM, Barrett J, et al. The role of the medical emergency team in end-of-life care: a multicenter, prospective, observational study. Crit Care Med. 2012; 40(1):98–103.
  12. Subramaniam A, Botha J, Tiruvoipati R. The limitations in implementing and operating a rapid response system. Intern Med J. 2016; 46(10):1139–45.
  13. Smith RL, Hayashi VN, Lee YI, et al. The medical emergency team call: a sentinel event that triggers goals of care discussion. Crit Care Med, 2014; 42(2):322–327.
  14. Medicare Benefits Schedule Book. In: Ageing AGDoHa, editor.: Canberra; 2015.
  15. AMA. List of Medical Services and Fees: Australian Meical Association Limited; 2015.
  16. Nurses And Midwives (Victorian Public Health Sector) (Single Interest Employers) Enterprise Agreement 2012–2016, (2014).
  17. AMA. Rates of Pay: Doctors in Training 2015 [18/01/2017]. Available from: http://amavic.com.au/page/Doctors_in_Training/What_is_the_AMA_doing_for_you/Rates_of_Pay_Doctors_in_Training/
  18. Federation AN. Quick Reference Pay Rates for Registered Nurses and Midwives 2012–2016 Agreement - Public Sector 2014.
  19. AMA Victoria - Victorian Public Health Sector Medical Specialists Enterprise Agreement 2013, (2013).
  20. Ranji SR, Auerbach AD, Hurd CJ, et al. Effects of rapid response systems on clinical outcomes: systematic review and meta-analysis. J Hosp Med. 2007; 2(6):422–32.
  21. Qaseem A, Snow V, Shekelle P, et al. Evidence-based interventions to improve the palliative care of pain, dyspnea, and depression at the end of life: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2008; 148(2):141–6.
  22. Downar J, Barua R, Rodin D, et al. Changes in end of life care 5 years after the introduction of a rapid response team: a multicentre retrospective study. Resuscitation. 2013; 84(10):1339–44.
  23. Lorenz KA, Lynn J, Dy SM, et al. Evidence for improving palliative care at the end of life: a systematic review. Ann Intern Med. 2008; 148(2):147–59.
  24. Lynn J, Schall MW, Milne C, et al. Quality improvements in end of life care: insights from two collaboratives. Jt Comm J Qual Improv. 2000; 26(5):254–67.
  25. Tan LH, Delaney A. Medical emergency teams and end-of-life care: a systematic review. Crit Care Resusc. 2014; 16(1):62–8.
  26. Aslakson RA, Curtis JR, Nelson JE. The changing role of palliative care in the ICU. Crit Care Med. 2014; 42(11):2418–28.
  27. Vazquez R, Gheorghe C, Grigoriyan A, et al. Enhanced end-of-life care associated with deploying a rapid response team: a pilot study. J Hosp Med. 2009; 4(7):449–52.
  28. Care ACoSaQiH. safety and quality of end-of-life care in acute hospitals: a background paper: ACSQHC, 2013. Sydney, Australia2013. p. 52.
  29. Parish K, Glaetzer K, Grbich C, et al. Dying for attention: palliative care in the acute setting. Aust J Adv Nurs. 2006; 24(2):21–5.
  30. Le BH, Watt JN. Care of the dying in Australia’s busiest hospital: benefits of palliative care consultation and methods to enhance access. J Palliat Med. 2010; 13(7):855–60.
  31. Federation AN. Quick Reference Pay Rates Registered Nurses and Midwives 2012–2016 Agreement -PUBLIC SECTOR 2014 [18/01/2017]. Available from: http://admin.anfvic.asn.au/multiversions/2776/FileName/RN_RM_PubSect2014.pdf

Contact diana@nzma.org.nz
for the PDF of this article

View Article PDF

Rapid response systems (RRS) were developed to improve patient safety1 and consist of an afferent arm for identification of deteriorating hospital patients and triggering of a clinical review, and an efferent arm that responds to patient deterioration by rapidly deploying a team led by a physician or a nurse with appropriate critical care expertise.2 RRS have been widely adopted throughout Australia and New Zealand to identify deteriorating patients with the aim to prevent serious adverse events such as cardiac arrest.3 Operating a RRS has financial and resource implications, although they are hard to evaluate in the context of wider hospital costs. We believe that RRC activations may reduce the overall healthcare expenditure by expediting optimal delivery of care to the deteriorating patient.

RRS are increasingly detecting patients who are actively dying and have therefore become a surrogate way of providing end-of-life care (EOLC) to such patients.4 It is also evident that not all rapid response call (RRC) activations are clinically appropriate and may not necessarily reflect a person’s wishes.5,6 Older people are particularly at risk of such avoidable RRCs as they represent a substantial proportion of the inpatient population7 and have multiple comorbidities, placing them at a higher risk of clinical deterioration.8,9 These patients may not wish to have life-prolonging measures performed but have not had the opportunity to have that adequately communicated to the treating team.10

There are significant knowledge gaps with regards to erroneously activating RRCs in patients with clearly documented treatment limitations specifying that they are not for further RRCs. Such erroneous calls are avoidable. This has the potential to cause not only distress to the patients and their relatives but could also lead to increased costs to healthcare settings. Previous studies have specifically excluded such patients who were “not for a RRC” because their needs and management were “complex”.11 In addition to contravening patients’ wishes, avoidable RRCs (ARRC) are also associated with time and financial costs. This includes obvious expenses such as equipment usage and staff wages but also the opportunity cost of removing clinical staff from attending to duties they would otherwise be performing. Despite this, no studies have examined the financial cost associated with ARRC.12

We aimed to retrospectively identify the incidence of ARRC in older people and explore the time and financial costs associated with them.

Methodology

We retrospectively reviewed the medical records of all RRC activations in people 80 years of age and over admitted to Frankston Hospital (Victoria, Australia) from 1 March 2015 to 31 August 2015 who had at least one avoidable RRC (ARRC). Frankston hospital is the major hospital within Peninsula Health serving the metropolitan and regional areas of Mornington Peninsula. Frankston hospital has 454 beds, a 15-bedded ICU and had 61,711 admissions and discharges in the last 12 months.

Patient selection

We have included patients over 80 years, as these patients are the most vulnerable and are likely to have limitations of medical treatment (LOMT).13 Hence, we chose them as the target study group.

Classification of appropriateness

A resuscitation care plan is discussed with all appropriate patients and recorded on a dedicated document (see Supplementary Figure 1). This document indicates the outcome of the discussion between doctors, patient and occasionally carers regarding the nature of appropriate interventions in the event of clinical deterioration or cardiac arrest. Any plan to limit resuscitation is discussed with senior medical staff directly responsible for the patient’s management during that hospital admission.

We defined ARRC as a RRC activation occurring despite a patient having prior documentation limiting their medical interventions to have no further RRC activations. The decision that a patient was not for further RRCs was made as a direct consequence of earlier RRCs, when resuscitation plans were revised to “treatment aimed at symptom management” and “not for further RRC”. The decision that a patient should not have any further RRCs was made by the patient or their relatives, treating team, and/or by the RRC team in conjunction with the treating team. Once it was documented that a patient should not have any further RRCs, any further RRC during the same admission was defined as avoidable.

Rapid response system (RRS) in Frankston Hospital

The RRS in Frankston Hospital was established in 1999. The rapid response team (RRT) in this hospital includes a critical care liaison nurse (CCLN) (between 8:30 and 22:00) or ICU nurse (between 22:00 and 08:30), an ICU registrar and a medical registrar. If a “Code Blue” is called, an additional ICU nurse and a coronary-care trained nurse also attend. An anaesthetist is also available if required for airway management.

Approximately 8–10 local staff members are usually present during a RRC including: 2–3 bedside nurses; ward in-charge nurses; home-team doctors (which may include a hospital medical officer or registrar, an intern and a consultant); and a patient services manager (PSM).

End-of-life care in Peninsula Health

“Medical Futility” in the study hospital is established by the admitting senior clinicians at the time of admission to hospital in patients who are in a persistent vegetative state or who are terminally ill. This process usually takes a shared decision approach. A ‘Care of the Dying Patient’ (CDP) pathway may be implemented by the treating team consultant or multidisciplinary team decision for such patients in whom death is deemed to be inevitable. The CDP pathways are algorithms designed by the palliative care team, and are aimed at managing symptoms and ensuring patients are made comfortable. The CDP pathway is reassessed every three days if required.

Data collection and outcomes

The date, time and clinical indication for RRC activation was retrieved from CCLN database and confirmed against the patients’ electronic medical records. The database maintained by the dedicated CCLN team captures every RRC and subsequent reviews of involved patients. We extracted basic demographic data, RRC clinical trigger indication, staff in attendance, duration of RRC and equipment used from the electronic medical record.

We defined an ‘in-hours’ RRC as that occurring between the 8:30 and 17:00 and RRCs occurring outside of this time-period as ‘out-of-hours’. The time to RRC activation was calculated as the time from amended resuscitation form documentation requesting no further RRC activations to the next RRC activation.

Costings

The costings were classified into financial costs (usage of investigations, equipment and management) and opportunity costs (defined as staff duration of attendance). Other less-tangible costs (such as emotional costs) can be estimated; however, this is difficult and at times controversial. Therefore, we have limited our analysis to include only estimated financial and opportunity costs for all patients who had an ARRC. The reference year of 2015 was used to calculate the costs associated with each resource use item. All costs are presented in Australian dollars (AUD) and evaluated from the perspective of a health service provider. The medical records relating to each inappropriate RRC were reviewed and costs of an episode were summed for each individual patient based on actual resource use (Supplementary Table 1A and 1B).

Investigation, management and equipment costs

The occurrence of each investigation was recorded from the medical record and assigned a corresponding standard unit price based on the 2015 Medicare Benefit Schedule of the Australian government14 and the AMA list of Medical Services.15 See Supplementary Table 1A for a list of the common equipment and investigations costs.

Workforce costs

The hourly wages for staff were calculated based on that published in 2015 sources16–19 (See Supplementary Table 1B). Although a proportion of the nursing staff are employed on a casual basis (who are paid 25% more than a regular employee for their level of experience), we assumed all nurses to be contracted by the hospital. The doctors and nurses have varied clinical experience. We included their minimum, maximum and average wages in costing for both nurses and doctors.

Analysis

We used N (%), mean (standard deviation) and median (interquartile range) as appropriate to describe the data. We calculated the cost of each individual ARRC based on the usage of investigations, equipment and management (defined as financial costs) and staff duration of attendance (defined as opportunity costs). All statistical analysis was conducted using SPSS (IBM, v.20) statistical package.

Ethical approval

This project was reviewed and approved as audit activity by Research Governance of Peninsula Health (ref. QA/15/PH/23). Informed consent was not required as this study was a retrospective chart review that required no patient contact.

Results

A total of 1,974 patients ≥80 years of age were admitted to Frankston hospital during the study period. A total of 255 RRC occurred in 186 patients (Figure 1). Of these, a total of 25 (9.8%) RRCs that involved 18 individual patients were ARRC (mean age = 85.6 years, SD = 3.94). Table 1 describes the characteristics of patients who had a ARRC. Six patients had more than one ARRC. Most patients (88%) were admitted under a medical team with “functional decline” being the most common (25%) presenting complaint on admission.

Figure 1: Flowchart to describe that about 10% of the RRC activations were avoidable.

c

Table 1: Characteristics of avoidable rapid response call (ARRC) activations.

*Other refers to RRCs due to staff concern, including: uncontrolled or severe pain (including chest pain).
ARRC were not trigged due to other RRC triggers as listed in Supplementary Table 2.
N = number; IQR = interquartile range; UTI = urinary tract infection; COPD = chronic obstructive pulmonary disease; APO = acute pulmonary oedema; NSTEMI – non-ST elevation myocardial infarction; RRC; RR = Respiratory rate; SaO2 = blood oxygen saturation; SBP = Systolic Blood Pressure; HR = Heart Rate.

A majority (88%) of ARRCs occurred during weekdays, with similar proportions occurring in (52%) and out (48%) of business hours. The triggers for ARRC activations were: SaO2 <90% (20%); SBP <90mmHg (20%); respiratory distress (12%); tachycardia (HR >130 bpm) (12%); tachypnoea (respiratory rate >36) (8%); SBP >180 mmHg (8%); bradycardia (HR <40 bpm) (4%) and seizure (4%) (Table 1). The median duration of each RRC was 22 min (IQR 13–40 minutes).

The median time to ARRCs was within three days (IQR—one day to six days) from the time that patients were documented to be not for further RRCs. Most patients (94.4%) who had a ARRC died in hospital within three days or were transferred directly to palliative care. Twelve out of the 18 patients (66.7%) were already on CDP with syringe-driver infusions of morphine and midazolam at the time of their ARRC; however, palliative care team utilisation was seen in only in 40% of patients.

Costings

The calculated financial costs (usage of investigations, equipment and management) and estimated the opportunity costs (defined as staff duration of attendance) for all patients who had an ARRC. Table 2 describes the investigation, management and staff costs associated with ARRC. The total costs of investigations and management was AUD $2,267.01, and maximum workforce cost was AUD $3,861.55. The total financial cost of all 25 ARRC was AUD $6,128.56 (Table 3 and Supplementary Tables 3A and 3B).

Table 2: Investigation, management and workforce costs during avoidable RRC (ARRC) activations. The most common investigations, all necessary piece of equipment and staff member at the ARRC identified and costed.

*Medications included bronchodilators, steroids, analgesia (excluding opiates), antiepileptic agents, antiarryhtmic agents, diuretics, metaraminol
FBE = full blood exam; UEC = electrolytes, urea, creatinine; Coags = coagulation profile;
LFT = liver function tests; CMP = calcium magnesium phosphate; ECG = electrocardiogram;
CT = computed tomography scan; IV = intravenous.

Table 3: The doctors and nurses involved in the ARRC along with their time (in minutes) spent for each ARRC is described. The cost of investigations performed during each ARRC is included. Based on this, the maximum opportunity wages were calculated using Supplementary Table 1B.

c

*Based on maximum workforce wages. We have also calculated the cost based on minimum and average workforce wages. Refer Supplementary Tables 3A and 3B.

ARRC = Avoidable Rapid Response Call; Med = medical; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; ANUM = associate nurse unit manager; PSM = patient service manager; Cons = consultant.

Discussion

We found that approximately 10% of RRC at our hospital were avoidable, with many of these patients being subjected to non-beneficial investigations and interventions. This poses significant challenges for both the afferent and efferent limbs2 of the RRS. This was associated with a small but important cost (that includes both financial and opportunity costs). This is the first paper to explore the incidence and financial costs of ARRC in the Australian inpatient setting.

ARRC in the current study were associated with a maximum cost of approximately AUD $6,128.56 in equipment, investigations, management and staff time. Although it is a relatively small figure, it reflects only the expenditure related to a single inappropriate aspect of patients’ care over a six-month period, and at a single centre. It is also likely that this figure represents only a fraction of the proportion of healthcare expenditure that is unnecessarily spent on inappropriate interventions for patients who require, or would prefer, only symptom management and comfort care.

Most ARRCs happened within three days from the time that patients were documented to be not for further RRCs. The median time for the ARRCs was 22 minutes, with some RRCs observed to be as long as 71 minutes. The reasons for this are unclear; however, discussions with primary teams and patients’ family take time. A previous study reported that limitation of medical therapy-based RRC often took considerably longer than other RRCs.11

A majority of patients in our study who had a ARRC were admitted under medical teams. The reasons why ARRC may be more common for medical than surgical patients are unclear.

Performing basic interventions such as administering supplemental oxygen or intravenous fluids, suctioning patients’ upper airways, securing intravenous access and ordering basic investigations (ECG, arterial blood gases, chest x-ray and routine bloods)20 have become an expectation for RRCs. These interventions come at a cost to the patient/family’s quality of life and experience, staffing resources and finances. Rather than being reactive, the RRTs should consider whether such interventions and investigations are appropriate or necessary in dying patients.

A recent multicentre study reported that 29.5% of patients with a LOMT had a RRC on the same day that they died.11 Similarly we found that most ARRC were in the final days of patient’s lives. This suggests that an important trigger for ARRC may be due to the urgent need for symptom control at the end of life.21 Other potential triggers could include a lack of clear documentation or other communication between patients, family and carer providers regarding goals of care and limitations of treatment.

Enhanced services such as effective management of physical, psychological and spiritual distress; timely and sensitive communication about appropriate goals of patient care; alignment of treatment with patient preferences; attention to families’ needs and concerns; planning for care transitions; and support for clinicians at the end of life could help reduce the number of ARRC and potentially improve patient and carer’s experiences.21–26

A recent review observed that LOMT discussions occurred more commonly than resuscitation interventions such as endotracheal intubation. This review also highlighted the fact that none of the studies reported the frequency of palliative care interventions implemented by the MET.25,27 There is good evidence that formal involvement of palliative care at the end-of-life improves patient and carer outcomes,23,24 however, such services appear to be underutilised.4,22 Likewise, patients who had ARRC activations were less likely to have palliative care consultations to those who did not.22 The reasons for this are unclear and warrant further investigation.

We believe that a multidisciplinary team approach should be adopted in managing these patients, with timely referral to palliative care teams. The concept of a palliative emergency team (PET) warrants further investigation. This may present a more cost-effective and appropriate alternative to RRCs for patients on end-of-life pathways, once the educational and behavior shaping components are optimised.

Even though clinicians should not be discouraged from activating RRCs, ARRC should be avoided to maximise comfort and dignity in the last days of life.

Demystifying definitions and terminology such as ‘palliative care’, ‘not for resuscitation’, ‘advance care planning’ and ‘advance health directive’ is prudent to aid communication and standardising practice.28

Although most clinicians working in acute facilities will participate in providing end-of-life care to patients, it appears that the management of this care varies due to knowledge gaps, skill and experience among the clinicians caring for the patient.29,30 Active education, using simulation and role playing will be beneficial for all healthcare professionals.

By reducing the number of ARRC, there may be some financial support available to fund these under-utilised strategies (eg, additional resources in palliative care).4,22

Strengths and limitations

To our knowledge, this is the first study to directly evaluate the incidence and the costs associated with ARRC. A strength of this study was, given the small sample size, we were able to micro-cost individual times and associated financial costs of each individual ARRC. However, the study also has a few limitations. The small sample size at a single site and retrospective design limits broader generalisation of our results. Given the retrospective design of our study, we are unable to exactly identify the reasons why ARRC occurred, despite the documentation to confirm the patients should only be treated symptomatically and a higher incidence of such calls in patients admitted for medical conditions. Some patients had more than one ARRC and the reasons for this remain unclear due to the retrospective nature of our study. We relied upon documentation to identify ARRC. This study was unable to quantify other costs associated with ARRC such as emotional and physical discomfort of the patient and their relatives’, and the negative impact on the care of other patients. It was also not possible to retrospectively investigate the specific wishes of patients and their families using available medical records.

Conclusion

Avoidable RRC (ARRC) occur in 10% of RRCs for hospital inpatients aged over 80 years, and are associated with potentially increased time and financial costs. Further research is required to understand the triggers to these ARRC and to study effectiveness of new strategies designed to reduce the burden of ARRC on patients, carers and hospital staff.

Supplementary Figure 1: Example of a Peninsula Health Resuscitation Plan defining avoidable rapid response calls (ARRC).  

c

We defined ARRC as RRC activations despite clear documentation confirming not for further RRCs. Patients were included only if a resuscitation plan document was completed with the Patient is for treatment aimed at “Symptom Management” box ticked and additional hand-written documentation of “not for further MET calls”.

Supplementary Table 1A and 1B. Items and item costs used to calculate financial cost of rapid response calls (RRCs). Table 1A: Most common blood and radiological investigations. Table 1B: Costing of workforce who were involved in the RRC.

*/**AMA List of Medical Services and Fees, Medical Benefit Scheme.

*Dorevitch pathology (Peninsula Health pathology provider).

**MIA radiology.

IV = intravenous; CT = computed tomography scan; ECG = electrocardiogram.

Supplementary Table 1B: Costing of workforce who attend and look after the RRC. Wages are represented as entry level (minimum) and maximum.

CCLN = Critical Care Liaison Nurse.

Doctors wages based on Award as of December 2015.

http://admin.anfvic.asn.au/multiversions/2776/FileName/RN_RM_PubSect2014.pdf (assessed 18 January 2017).

RRC = Rapid Response Call; ICU = intensive care unit; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; PSM = patient service manager; Cons = consultant.

Supplementary Table 3B: The doctors and nurses involved in the RRC along with their time (in minutes) spent for each RRC is described. The cost of investigations performed during each RRC is included. Based on this, the average workforce wages calculated using Supplementary Table 1B.

c

RRC = Rapid Response Call; ICU = intensive care unit; reg = registrar; CCLN = critical care liaison nurse; RN = registered nurse; PSM = patient service manager; Cons = consultant.

Summary

Abstract

Rapid response calls (RRCs) are designed to appropriately manage clinical deterioration. However, not all RRCs are appropriate due to medical futility or the patient s wishes. Incidence and costs associated with avoidable-RRC (ARRC) remain underexplored.

Aim

The aim of this study was to describe the incidence and costs of ARRC activations in older patients.

Method

We retrospectively reviewed RRCs in patients aged 580 years over six months. We defined ARRC as RRC activations despite clear documentation confirming not for further RRCs. Data on investigations, equipment and management of each ARRC were analysed. We then micro-costed each ARRC using standard references.

Results

Ten percent (25/255) of RRCs were ARRC (mean age 85.6 years) with most patients (88%) admitted under medical teams. Median duration of ARRC was 22 minutes (IQR 7-38 minutes). Palliative care services were underutilised (40%). Most patients (94.4%) died soon after ARRC. The costs for investigations, equipment and management was AUD $2,267.01, opportunity costs were AUD $3,861.55, with a grand total cost of AUD $6,128.56.

Conclusion

ARRC, noted in 10% of RRCs, are associated with increased time and financial costs. Further research is required to better understand ARRC triggers to reduce the burden of ARRC on patients, carers and hospital staff.

Author Information

- Ashwin Subramaniam, Intensivist, Intensive Care, Frankston Hospital, Australia; Cameron Green, Intensive Care, Frankston Hospital, Australia; Mohammad Omair, Department of Medicine, Frankston Hospital, Australia; Lionel Soh, Department of

Acknowledgements

We gratefully acknowledge the assistance of Dr Christopher Moran (Geriatrician) for his valuable suggestions in preparing this manuscript. We would also like to thank the Critical Care Liaison Nurse service (in particular, Jo Molloy and Naomi Pratt) of Peninsula Health for the use of their patient databases, and for their commitment to maintaining the rapid response system.

Correspondence

Dr Ashwin Subramaniam, Intensivist, Intensive Care, Frankston Hospital, Hastings Road, Frankston 3199, Australia.

Correspondence Email

catchdrash@yahoo.com

Competing Interests

Nil.

  1. Lee A, Bishop G, Hillman KM, et al. The Medical Emergency Team. Anaesth Intensive Care. 1995; 23(2):183–6.
  2. Jones DA, DeVita MA, Bellomo R. Rapid-response teams. N Engl J Med. 2011; 365(2):139–46.
  3. Hillman KM, Chen J, Jones D. Rapid response systems. Med J Aust. 2014; 201(9):519–21.
  4. Hillman K. Dying safely. Int J Qual Health Care. 2010; 22(5):339–40.
  5. Hillman K, Chen J, Cretikos M, et al. Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. Lancet. 2005; 365(9477):2091–7.
  6. Buist MD, Moore GE, Bernard SA, et al. Effects of a medical emergency team on reduction of incidence of and mortality from unexpected cardiac arrests in hospital: preliminary study. BMJ. 2002; 324(7334):387–90.
  7. Ruiz M, Bottle A, Long S, et al. Multi-Morbidity in Hospitalised Older Patients: Who Are the Complex Elderly? PLoS One. 2015; 10(12):e0145372.
  8. Karlamangla A, Tinetti M, Guralnik J, et al. Comorbidity in older adults: nosology of impairment, diseases, and conditions. J Gerontol A Biol Sci Med Sci. 2007; 62(3):296–300.
  9. Guralnik JM. Assessing the impact of comorbidity in the older population. Ann Epidemiol. 1996; 6(5):376–80.
  10. Parr JD, Zhang B, Nilsson ME, et al. The influence of age on the likelihood of receiving end-of-life care consistent with patient treatment preferences. J Palliat Med. 2010; 13(6):719–26.
  11. Jones DA, Bagshaw SM, Barrett J, et al. The role of the medical emergency team in end-of-life care: a multicenter, prospective, observational study. Crit Care Med. 2012; 40(1):98–103.
  12. Subramaniam A, Botha J, Tiruvoipati R. The limitations in implementing and operating a rapid response system. Intern Med J. 2016; 46(10):1139–45.
  13. Smith RL, Hayashi VN, Lee YI, et al. The medical emergency team call: a sentinel event that triggers goals of care discussion. Crit Care Med, 2014; 42(2):322–327.
  14. Medicare Benefits Schedule Book. In: Ageing AGDoHa, editor.: Canberra; 2015.
  15. AMA. List of Medical Services and Fees: Australian Meical Association Limited; 2015.
  16. Nurses And Midwives (Victorian Public Health Sector) (Single Interest Employers) Enterprise Agreement 2012–2016, (2014).
  17. AMA. Rates of Pay: Doctors in Training 2015 [18/01/2017]. Available from: http://amavic.com.au/page/Doctors_in_Training/What_is_the_AMA_doing_for_you/Rates_of_Pay_Doctors_in_Training/
  18. Federation AN. Quick Reference Pay Rates for Registered Nurses and Midwives 2012–2016 Agreement - Public Sector 2014.
  19. AMA Victoria - Victorian Public Health Sector Medical Specialists Enterprise Agreement 2013, (2013).
  20. Ranji SR, Auerbach AD, Hurd CJ, et al. Effects of rapid response systems on clinical outcomes: systematic review and meta-analysis. J Hosp Med. 2007; 2(6):422–32.
  21. Qaseem A, Snow V, Shekelle P, et al. Evidence-based interventions to improve the palliative care of pain, dyspnea, and depression at the end of life: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2008; 148(2):141–6.
  22. Downar J, Barua R, Rodin D, et al. Changes in end of life care 5 years after the introduction of a rapid response team: a multicentre retrospective study. Resuscitation. 2013; 84(10):1339–44.
  23. Lorenz KA, Lynn J, Dy SM, et al. Evidence for improving palliative care at the end of life: a systematic review. Ann Intern Med. 2008; 148(2):147–59.
  24. Lynn J, Schall MW, Milne C, et al. Quality improvements in end of life care: insights from two collaboratives. Jt Comm J Qual Improv. 2000; 26(5):254–67.
  25. Tan LH, Delaney A. Medical emergency teams and end-of-life care: a systematic review. Crit Care Resusc. 2014; 16(1):62–8.
  26. Aslakson RA, Curtis JR, Nelson JE. The changing role of palliative care in the ICU. Crit Care Med. 2014; 42(11):2418–28.
  27. Vazquez R, Gheorghe C, Grigoriyan A, et al. Enhanced end-of-life care associated with deploying a rapid response team: a pilot study. J Hosp Med. 2009; 4(7):449–52.
  28. Care ACoSaQiH. safety and quality of end-of-life care in acute hospitals: a background paper: ACSQHC, 2013. Sydney, Australia2013. p. 52.
  29. Parish K, Glaetzer K, Grbich C, et al. Dying for attention: palliative care in the acute setting. Aust J Adv Nurs. 2006; 24(2):21–5.
  30. Le BH, Watt JN. Care of the dying in Australia’s busiest hospital: benefits of palliative care consultation and methods to enhance access. J Palliat Med. 2010; 13(7):855–60.
  31. Federation AN. Quick Reference Pay Rates Registered Nurses and Midwives 2012–2016 Agreement -PUBLIC SECTOR 2014 [18/01/2017]. Available from: http://admin.anfvic.asn.au/multiversions/2776/FileName/RN_RM_PubSect2014.pdf

Contact diana@nzma.org.nz
for the PDF of this article

Subscriber Content

The full contents of this pages only available to subscribers.

LOGINSUBSCRIBE