Healthcare Quarterly

Healthcare Quarterly 23(1) April 2020 : 47-52.doi:10.12927/hcq.2020.26139
Quality Improvement

Analysis of Extreme Length of Stay Hospitalizations for Children and Youth in a Quaternary Care Hospital

Elisabeth Yorke, Lennox Huang, Julia Orkin, Tyler Chalk, Farrah Ladha and Alene Toulany

Abstract

Length of stay (LOS) is an important issue for many healthcare organizations. In-patients with extreme LOS account for a disproportionately large percentage of hospital costs. Our analysis of over 15,000 pediatric hospital discharges at The Hospital for Sick Children (Toronto, Canada) between 2015 and 2016 revealed that the vast majority of patients with extreme LOS were discharged directly home, with only a minority receiving home-based services. Patients with the greatest LOS were accounted for by primarily four subspecialty services. Although this report outlines an analysis of pediatric in-patients, our findings and implications are relevant for all jurisdictions and populations as many acute care hospitals often "hold" patients with complex, chronic illness as in-patients for extended periods because alternate appropriate services may not exist or be available. Our case study highlights three key areas to improve quality of care for patients with extreme LOS: alternate levels of care, system resources and transitions to home.

Introduction

With the rising cost of healthcare in Canada, macrosystems are often looking for ways to reduce expenditures. Since 1997, hospitals have accounted for the greatest share of health spending (28.3%), followed by drugs (16.4%) and physician services (15.4%) (CIHI 2017). Recent literature shows that a very small proportion of the population – those with medical complexity – is responsible for a large percentage of healthcare costs (Wodchis et al. 2016).

In pediatrics, children with medical complexity are defined as those with complex underlying chronic health conditions that are typically associated with significant functional status limitations (Cohen et al. 2012). In the study by Cohen et al. (2012), children with medical complexity accounted for almost one-third of all child health spending, mirroring adult population statistics. In fact, those patients with the longest lengths of stay, otherwise known as extreme LOS, were accountable for the majority of these expenses. Considering these data, pediatric in-patients with extreme lengths of stay account for a disproportionately large percentage of the healthcare resources and costs for hospitals because of complex healthcare needs, multiple interventions and possible hospital-acquired adverse events.

There are a multitude of reasons why reducing the LOS in hospital for patients may be beneficial. From the patients' perspective, extended in-hospital LOS causes family and school disruption and workforce displacement for adult caregivers, places patients at risk of nosocomial infections and medical error and has financial implications. With regard to resource utilization, extended stays are often demanding on nursing and allied health disciplines, and progressive and chronic diseases can lead to increasingly longer admissions, adding strain to an already resource-limited, publicly funded system. Identifying early predictors for extreme LOS may provide opportunities for intervention to minimize burden to the patient, family and health system.

Work is already being done to build capacity for alternate levels of care (ALC), to reduce in-hospital LOS and to move patients closer to home. However, this process is slow, and the traditional pathway of home to the hospital and back remains the most common direction for intermittently ill children with medical complexity. When examining this pathway, it becomes apparent that bottlenecks in patient flow have increasing effects downstream. For example, backlog in the emergency department leads to an increase in hospital admissions, and maximum capacity in the hospital often leads to forced discharges. This underscores the importance of ALC locations and shows the upstream implications of bottlenecks in the patient flow. Taking these factors into account, it is apparent that improving the quality of service delivery is critical to improve efficiency in our complex health system. By concentrating on those patients with extreme LOS, incremental changes in this relatively small group of patients may have broader effects to the cost and quality of life of patients at large.

A MEDLINE search was conducted looking for LOS data, with search parameters for extreme LOS (90–99th percentile), children's hospitals and in the past 10 years. This search returned 63 abstracts, which were screened for clinical relevance. Of these, 18 studies were identified as relevant. In one American study looking at infants staying six months or longer in a newborn unit, increased LOS was associated with extreme prematurity, respiratory distress and necrotizing enterocolitis. Interestingly, when qualitative data were collected from the in-patient teams looking after these patients, the healthcare providers in these teams viewed much of the prolonged care provided as ongoing and invasive management (noted to be unlikely to help the infant), also identifying this care as the main driver of LOS (Catlin 2008). Other noted influences on LOS were patient compliance with medications while in the hospital (Dekker et al. 2016), infection with resistant organisms (sometimes secondary to exposure during previous admission; Fan et al. 2014), age at admission (younger children compared to older children with bacterial pneumonia; Gajewska et al. 2016) and hospital-acquired adverse drug reactions (Khan 2013). In one American study looking at the demographics of hospital admissions for children with high-risk conditions causing death, the children who were the most ill were the ones with the longest LOS and most frequent readmissions (Miller et al. 2012). Taken together, these results highlight the importance of examining this complex patient population and identifying areas for intervention in the pathway from home to the hospital in providing, or preventing, hospital admission.

Case Study: Analysis of Extreme Hospital LOS Trends for Children and Youth at The Hospital for Sick Children

We aimed to identify and characterize pediatric patients with extreme LOS at a large quaternary care hospital in Toronto, Canada, to identify possible strategies for quality care improvement and reduced hospital stay.

A retrospective analysis of 15,980 pediatric hospital discharges from April 2015 to March 2016 was performed at The Hospital for Sick Children. LOS data for the 99th percentile (162 patients) and the 95th percentile (816 patients) were analyzed separately for identification of population characteristics and trends. Variables included duration of admission, discharge destination, most responsible admitting medical service (pediatrics, cardiology, etc.) and surgical service (to further delineate interventions completed during admission).

The vast majority of patients with extreme LOS (68% of 99th percentile; 78% of 95th percentile) were discharged directly to home, the minority of whom received publicly funded support services (private homecare services were not captured in our data) after discharge (16% of 99th percentile; 13% of 95th percentile; Figures 1 and 2). Four services (general pediatrics, neonatology, cardiology and hematology/oncology) contributed to more than 60% of the extreme outlier cases. However, one service (neonatology) had a significantly higher rate of discharge to ALC (49%) than the other services (Figures 3 and 4).


Click to Enlarge
 


Click to Enlarge
 


Click to Enlarge
 


Click to Enlarge
 

Within these four services, the patients with the longest LOS had multiple comorbidities. Patients admitted under the cardiology service had a diagnosis of congenital heart defects (e.g., tetralogy of Fallot and hypoplastic left heart syndrome), along with histories of multiple interventions and surgeries. Within neonatology, infants with the longest LOS included those with multiple congenital abnormalities primarily involving the respiratory, cardiac and renal systems, followed by complications of prematurity, such as very low birth weight and necrotizing enterocolitis. Neonatology patients with prolonged LOS had, on average, over 10 different medical comorbidities. For patients admitted under the hematology/oncology service, acute myeloid leukemia and acute lymphoblastic leukemia were seen most often, with the least number of comorbidities identified of any of the top four clinical services. Within pediatrics, a variety of conditions leading to prolonged hospital admission were observed. Many cases were a direct result of the intrahospital transfer process: for example, patients in the neonatal intensive care unit (NICU) or cardiology units were often transferred to the general pediatrics ward for feeding or respiratory support when their acute medical complications were resolved (tetralogy of Fallot and esophageal atresia being most common). These patients at the 99th percentile of LOS also had, on average, over 10 medical comorbidities.

With regard to the cost of stay during admission, a summary of mean direct cost of admission for patients at the 99th percentile of LOS compared to mean cost of admission for patients with median LOS is provided in Table 1. The mean direct cumulative cost for patients at the 99th percentile of LOS was $337,610 per admission, whereas that for patients with median LOS was $3,959 per admission.


Table 1. Cost comparison of extreme LOS (>99th percentile) with median hospital LOS in a tertiary care pediatric hospital (2015–2016)
  Patients with >99th percentile LOS Patients with median LOS (days)
Total number of inpatient discharges 157 2,424
Mean (SD) number of comorbidities 12 (8) 2
Mean (SD) LOS (days) 135.3 (65.1) 2.0
Mean (SD) direct cost/day/patient (CAD) $2,495 $1,979
Mean direct cost for hospitalization (CAD) $337,610 (225,636) $3,959
CAD = Canadian dollars; LOS = length of stay; SD = standard deviation.

 

In summary, the majority of patients with the greatest LOS could be accounted for by only four services, highlighting key areas of focus for improvement of quality of care delivery. In addition, there was a significant cost associated with patients at the 99th percentile of LOS, as well as an increase in medical complexity, compared to the patients with median time of admission (two days) during the period studied.

Discussion

Our case study highlights three key areas for improvement when addressing extreme LOS: ALC, system resources and transitioning out of hospital.

Alternate levels of care

Current LOS literature in the adult population shows that delayed placement into ALC is an important factor contributing to the prolonged LOS (Black and Pearson 2002). In our case study, many of the patients with the longest LOS were discharged directly home. It is possible that their LOS could have been shortened if an alternate destination was available (such as a step-down facility or improved homecare resources). However, in many jurisdictions, spaces in these alternate destinations are limited, and home resources for the level of care needed are not always accessible in every jurisdiction. Therefore, acute care facilities often provide continued care for ill, complex patients until they are fully ready for discharge. This not only contributes to added costs on the healthcare system but also limits the availability of critical acute care beds to those who need them more urgently. When these admissions are further characterized in the adult literature, patients with stroke, psychiatric conditions, morbid obesity and abusive behaviours accounted for the largest proportion of ALC days (Costa et al. 2012). A Canadian study reported that the wait time for nursing home beds was one of the most important factors contributing to LOS and ALC days (Costa et al. 2012).

In our hospital, pathways exist to improve the flow of patients between acute care units and general in-patient units. For example, many of the patients with extreme LOS on the cardiac critical care service are transferred to the general cardiology unit and then to pediatrics for general pediatric care, including feeding and weight gain before repeat surgery or management of intercurrent illness. A similar pathway also exists from the NICU to pediatrics, although there is a higher proportion of NICU transfers to community hospitals, reflecting the capacity of community hospitals to manage these patients (and in many cases, these would be repatriation after acute illness, such as necrotizing enterocolitis). The flow of a substantial group of these complex, long-stay patients to the pediatrics department highlights a care pathway within the hospital to achieve less acute levels of care. A bottleneck exists within the more general services that can provide "holistic care" as there is often no other place closer to home for these patients where the expertise is available to manage the level of complexity seen in this population.

System resources

The Canadian healthcare system was structured to manage acute care concerns, and to this day, the majority of healthcare expenditures are in acute care settings (Rosella et al. 2014). However, with our aging population, the increased chronicity of illness and the multimorbid nature of health concerns for patients, there is an increased need for ALC for patients, outside of the in-patient setting. When ALC beds are not readily available (or available at all), patients can continue to have prolonged admissions as they may no longer require quaternary-level care but are not yet ready to return home. This can result in upstream effects, such as emergency department crowding due to reduced patient flow, which in turn is associated with increased healthcare costs and delays in service (Krochmal and Riley 1994). As patients with medical complexity often require frequent admission to a hospital (Berry et al. 2011), it is apparent that this pathway of care may not be sustainable with the current healthcare system resources, and alternatives are greatly needed.

This is true in the general pediatric setting as well. Not only is the current medical system structured and funded for acute care or short stays, but also the nature of the care is often "one size fits all," with limited capacity to adjust for the needs of longer-term, more complex admissions. For example, frequent rotations of staff physicians and trainees often disadvantage patients who are admitted for long periods as they are not provided with coordinated, holistic care because of poor continuity and frequent handovers. In our hospital, there have been efforts to manage this gap in service by implementing a standardized handover checklist and employing clinical nurse specialists who can provide the continuity and knowledge of previous treatments and future plans for the longer-stay patients. The prolonged LOS population with significant complexity identified in our study highlights the need for a "niche" team, skilled in providing the specialized, complex care for this patient population and the continuity needed to plan and facilitate discharge. In addition to the need for specialized resources in the in-patient setting, in our case example, one of the greatest barriers to providing effective home care, besides the resource availability of this publicly funded service, is maintaining the level of training among care providers to support the complexity of this patient population (e.g., managing respiratory support for a patient with congenital heart disease or tracheostomy support for a child with a craniofacial abnormality). It is apparent that our current system of "one size fits all" care is a barrier to facilitating timely and supported discharge of the extreme LOS patients.

Transitioning out of hospital

Transition between levels of care is a continued challenge for the healthcare system and families alike. With multiple siloes of care, including primary care, acute care, rehabilitation services and home care, patients and their healthcare records are often fragmented, with limited pathways for communication. In Ontario, the varied electronic medical record systems further complicate the communication between levels of care. Patients often experience a sudden step-down in care when they are transitioned from the hospital to home, from having access to 24-hour nursing and physician care, allied care services and provided medications to managing most of their care independently (Soong et al. 2014). Within our own hospital, private homecare services are recorded in a separate data collection system, limiting connection with in-patient services. Regardless, many of the families in our population would not be able to afford private care, and many more publicly funded options are needed, as well as better communication and shared electronic healthcare records between levels of care, to support the transition to home safely and efficiently.

Conclusions

The patients with the longest LOS in our cohort had conditions similar to those seen in previous studies of high-cost users, particularly in the NICU and the oncology unit (Wodchis et al. 2016). A new area of focus is children with medical complexity, who are most often treated by pediatric hospitalists and who may be more comparable to older patients with multiple morbidities (on average, over 10 associated medical co-morbidities in the 99th percentile population), as seen in the adult literature (Reid et al. 2003).

There are ALC pathways within the pediatric system that are paving the way as models of care for patients with medical complexity. For example, our hospital has an existing pathway to transition patients with medical complexity to a pediatric rehabilitation facility for ongoing and less intensive recovery care, although the capacity of this facility to accept these patients and scope of the patients who meet referral criteria is currently limited, making this pathway not an option for the majority of medically complex pediatric patients transitioning out of hospital. Finally, as previously discussed, partnerships are being developed to support the training of care workers in community settings to provide the level of specialized care needed for more complex patients, understanding that greater capacity is needed to support this growing population outside of the acute care hospital setting.

Our case study highlights the importance of further investigating ALC for patients who stay the longest in hospital and broadening our support for home services, while reducing barriers to transition between levels of care. There may be opportunities to provide patients and their families with alternatives to prolonged hospital admission if these patients are not quite ready for home but no longer require full quaternary-care provision of service. In addition, there are limited data from a pediatric perspective on the characteristics of children at the extreme outliers of LOS, and more research is required to better understand this important group of patients.

About the Author(s)

Elisabeth Yorke, MD, MSc, MSc, FRCPC, is an adolescent medicine fellow in the Department of Paediatrics, University of Toronto, Ontario.

Lennox Huang, FRCPC, is the chief medical officer and vice president of medical & academic affairs at The Hospital for Sick Children in Toronto, Ontario.

Julia Orkin, MD, MSc, FRCPC, is the medical director of the Complex Care Program and an assistant professor in the Department of Paediatrics at the University of Toronto and The Hospital for Sick Children in Toronto, Ontario.

Tyler Chalk is the chief strategy officer at the Southlake Regional Health Centre in Newmarket, Ontario.

Farrah Ladha is the director of business development at Well.ca in Toronto, Ontario.

Alene Toulany, MD, FRCPC, is an assistant professor in the Department of Paediatrics, University of Toronto, and a staff physician in the Division of Adolescent Medicine, The Hospital for Sick Children in Toronto, Ontario.

References

Berry, J.G., R. Agrawal, D.Z. Kuo, E. Cohen, W. Risko, M. Hall et al. 2011. Characteristics of Hospitalizations for Patients Who Use a Structured Clinical Care Program for Children with Medical Complexity. The Journal of Pediatrics 159(2): 284–90. doi:10.1016/j.jpeds.2011.02.002.

Black, D. and M. Pearson. 2002. Average Length of Stay, Delayed Discharge, and Hospital Congestion. BMJ 325(7365): 610–11. doi:10.1136/bmj.325.7365.610.

Canadian Institute for Health Information (CIHI). 2017. National Health Expenditure Trends, 1975 to 2017. Retrieved March 14, 2020. <https://secure.cihi.ca/free_products/nhex-trends-narrative-report-2019-en-web.pdf>.

Catlin, A. 2008. Extremely Long Hospitalizations of Newborns in the United States: Data, Descriptions, Dilemmas. Advances in Neonatal Care 8(2): 125–32. doi:10.1097/01.ANC.0000317261.99072.e7.

Cohen E., J.G. Berry, X. Camacho, G. Anderson, W. Wodchis and A. Guttmann. (2012). Patterns and Costs of Health Care Use of Children with Medical Complexity. Pediatrics 130(6): e1463–70. doi:10.1542/peds.2012-0175.

Costa, A.P., J.W. Poss, T. Peirce and J.P. Hirdes. 2012. Acute Care Inpatients with Long-Term Delayed-Discharge: Evidence from a Canadian Health Region. BMC Health Services Research 12: 172. doi:10.1186/1472-6963-12-172.

Dekker, K.R., B.L. Myers and M.A. Barras. 2016. Compliance with Enoxaparin Dosing and Monitoring Guidelines and the Impact on Patient Length of Stay: A Pilot Study. Therapeutic Drug Monitoring 38(1): 59–63. doi:10.1097/FTD.0000000000000233.

Fan, N.C., H.H. Chen, C.L. Chen, L.S. Ou, T.Y. Lin, M.H. Tsai et al. 2014. Rise of Community-Onset Urinary Tract Infection Caused by Extended-Spectrum Beta-Lactamase-Producing Escherichia coli in Children. Journal of Microbiology, Immunology and Infection 47(5): 399–405. doi:10.1016/j.jmii.2013.05.006.

Gajewska, M., K. Lewtak, J. Scheres, P. Albrecht and P. Gory?ski. 2016. Trends in Hospitalization of Children with Bacterial Pneumonia in Poland. Central European Journal of Public Health 24(3): 188–92. doi:10.21101/cejph.a4164.

Khan, L.M. 2013. Comparative Epidemiology of Hospital-Acquired Adverse Drug Reactions in Adults and Children and Their Impact on Cost and Hospital Stay–A Systematic Review. European Journal of Clinical Pharmacology 69(12): 1985–96. doi:10.1007/s00228-013-1563-z.

Krochmal, P. and T.A. Riley. 1994. Increased Health Care Costs Associated with ED Overcrowding. The American Journal of Emergency Medicine 12(3): 265–66. doi:10.1016/0735-6757(94)90135-X.

Miller, R.L., A. Gebremariam and F.O. Odetola. 2012. Pediatric High-Impact Conditions in the United States: Retrospective Analysis of Hospitalizations and Associated Resource Use. BMC Pediatrics 12: 61. doi:10.1186/1471-2431-12-61.

Reid, R., R. Evans, M. Barer, S. Sheps, K. Kerluke, K. McGrail et al. 2003. Conspicuous Consumption: Characterizing High Users of Physician Services in One Canadian Province. Journal of Health Services Research and Policy 8(4): 215–24. doi:10.1258/135581903322403281.

Rosella, L.C., T. Fitzpatrick, W.P. Wodchis, A. Calzavara, H. Manson and V. Goel. 2014. High-Cost Health Care Users in Ontario, Canada: Demographic, Socio-economic, and Health Status Characteristics. BMC Health Services Research 14: 532. doi:10.1186/s12913-014-0532-2.

Soong, C., B. Kurabi, D. Wells, L. Caines, M.W. Morgan, R. Ramsden et al. 2014. Do Post Discharge Phone Calls Improve Care Transitions? A Cluster-Randomized Trial. PLoS One 9(11): e112230. doi:10.1371/journal.pone.0112230.

Wodchis, W.P., P.C. Austin and D.A. Henry. 2016. A 3-Year Study of High-Cost Users of Health Care. CMAJ 188(3): 182–88. doi:10.1503/cmaj.150064.

Comments

Be the first to comment on this!

Related Articles

Note: Please enter a display name. Your email address will not be publically displayed