If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author. C. van Lieshout, MSc, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, PO Box 85500, 3508 GA Utrecht, The Netherlands.
Delirium is a common condition of a global disturbance of cognition, triggered by underlying diseases. The objective of this study is to review the current evidence in the literature on direct healthcare costs and health-related quality of life (HRQOL) associated with delirium.
Methods
A systematic search was conducted in PubMed and Embase for relevant studies published between January 1, 2010 and November 4, 2021. Studies for inclusion reported estimates on healthcare costs or HRQOL, adjusted for relevant confounding factors.
Results
Fourteen studies on healthcare costs and eleven studies on HRQOL were included. Delirium resulted in (adjusted) increased costs ranging from $1,532 to $22,269 depending on included cost categories, the country and the type of hospital department. Increased length of stay for delirious patients ranged from 2.5 days to 10.4 days and had the largest contribution to overall, direct incremental costs. Heterogeneity was observed in HRQOL outcomes.
Conclusion
The analysis indicates that the presence of a delirium episode may lead to increased costs of hospitalisation. Changes in HRQOL due to delirium are not well demonstrated and more research is needed to determine the effect of delirium on HRQOL.
Zusammenfassung
Einleitung
Das Delir ist ein häufiges Beschwerdebild, das durch eine globale Störung der Kognition charakterisiert ist und durch zugrunde liegende Erkrankungen ausgelöst wird. Ziel dieser Studie ist es, die aktuelle Evidenz zu den direkten Gesundheitskosten und zur gesundheitsbezogenen Lebensqualität („Health-Related Quality of Life“, HRQoL) im Zusammenhang mit Delirzuständen zu sichten.
Methode
Die Datenbanken PubMed und Embase wurden systematisch nach einschlägigen, zwischen dem 1. January 2010 und dem 4. November 2021 publizierten Studien durchsucht. Eingeschlossen wurden Studien, die Schätzungen zu den Gesundheitskosten oder zur HRQoL (mit Adjustierung für relevante Störfaktoren) enthielten.
Ergebnisse
Eingeschlossen wurden 14 Studien zu Gesundheitskosten und 11 Studien zu HRQOL. In Abhängigkeit von berücksichtigten Kostenkategorien, Land und Krankenhausabteilung verursachten Delirzustände (adjustierte) Kostensteigerungen zwischen 1532 und 22 269 Dollar. Patienten mit Delir hatten um 2,5 bis 10,4 Tage verlängerte Krankhausverweildauern und den größten Anteil an den direkt zurechenbaren Gesamtkosten. Heterogenität ließ sich bezüglich der HRQoL-Outcomes beobachten.
Schlussfolgerung
Die Analyse zeigt, dass das Vorhandensein einer Delirepisode eine Steigerung der Krankenhauskosten verursachen kann. Veränderungen der gesundheitsbezogenen Lebensqualität infolge eines Delirs sind in den untersuchten Studien nicht hinreichend dargestellt; daher bedarf es weiterer Forschung, um die Auswirkungen eines Delirs auf die gesundheitsbezogene Lebensqualität zu bestimmen.
]. Three subtypes of delirium can be distinguished: hyperactive delirium, hypoactive delirium, and mixed delirium. Hyperactive delirium is characterized as highly alert and uncooperative, whereas hypoactive delirium is more common and characterized by a lot of sleep, and inattentive disorganized thought. Mixed delirium is a form where there is a fluctuation between hyperactive delirium and hypoactive delirium. This is the most frequently observed type [
]. The onset of delirium can be triggered by infections, fever, alcohol withdrawal, major surgery, terminal phases of cancer or AIDS, and medications. Delirium can occur at any age, but is most commonly seen in the elderly and young children on ICU. Estimates of the incidence of delirium as a complication in adult ICU patients are between 19 and 82 percent [
]. Different standards, such as the International Classification of Diseases, tenth edition (ICD-10), the Diagnostic Statistical Manual of Mental Disorders, fifth edition (DSM-5), the Confusion Aassessment Method (CAM) and The Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) were developed for Delirium detection [
Prevalence on hospital wards ranges from 7% to 27% depending on patient population and ward type (general surgical ward vs. geriatric ward and ICU), these figures might be an underestimation since delirium recognition is poor, only 12 to 35% of delirium patients are being recognised [
]. With use of validated assessment tools, recognition of delirium can be improved. This could result in a better management of delirium, which primarily consists of early recognition, and management of the underlying condition [
The burden of delirium is manifold, for patients, loved-ones and staff. For patients it may have an effect on mortality, morbidity, healthcare costs and health related quality of life (HRQOL) of patients. Gaining insight into the differences in costs and HRQOL is essential to further understand the impact of delirium and the potential for improvement of care. The aim of this review is therefore to summarize the available evidence on the impact of delirium on healthcare costs and HRQOL in hospitalized patients.
Methods
We performed a scoping review to summarize the available evidence on costs and impact on HRQOL of delirium. We adhered to the PRISMA statement for scoping reviews.
Inclusion criteria
Articles were eligible for this review if: 1) patients with delirium and without delirium were compared 2) the article provided information on HRQOL and/or hospitalisation costs. Studies were excluded when: 1) confounders were not investigated and/or adjustment for confounding was not conducted, 2) disease severity at baseline was not investigated, 3) hospital charges were not included, 4) studies were published in a language other than English 5) results were only presented in conference abstracts, and 6) studies on patients < 18 years.
Search
The MEDLINE and Embase databases were searched to find relevant primary articles from January 1, 2010 up to June 19, 2019. This search was updated on November 4, 2021 to extend the time window to November 4, 2021, in order to include the latest papers. Older papers were not included since costs are subject to change over time and the most relevant papers were of interest. For every database, a search strategy was defined with the support of an information specialist of the Utrecht University Library. Terms related to delirium, health care costs and HRQOL were searched in titles and abstracts of literature as well as in MeSH and Emtree terms. The complete search strategy is provided in the Appendix A.
Study selection
After removal of duplicates, the title and abstract of the obtained studies were screened for the eligibility criteria by a single researcher. The full texts of articles were assessed when needed, for clarification. If results were published multiple times, data from the most complete article was used.
Data collection
Data were extracted from each included article on: study design and period, study setting and country, number of included participants with percentage experiencing delirium, populations characteristics; including age and gender, delirium assessment tool, covariates and time of measurement. Data extraction was conducted by one researcher, and reviewed by another.
For cost studies, the cost category and the differences in costs in the presence of delirium were reported as well as the general extracted information listed. Studies that reported on costs specified by type of costs were summarized in a separate table. No critical appraisal of the calculations that derived the costs included in the studies was performed. If the year of the cost analysis was reported in the original papers, cost estimates were converted to 2020 US Dollars with use of consumer price indeces and currency exchange rates. The studies that were identified that reported healthcare costs were also screened for data on length of stay for delirious and non-delirious patients.
For the studies on HRQOL, the measurement tool and differences in outcome were extracted in addition to the study characteristics listed above. Studies which reported on the differences for each of the eight domains of the SF-36 were summarized.
Risk of bias
Possible confounders and adjustment for confounding were evaluated during the study selection part of this study; studies were excluded if no attempt was made to adjust for confounding.
Results
Study selection
The first systematic search identified 931 studies, after the removal of duplicates. After screening titles and abstracts, 58 studies remained for full text screening, of these, 18 studies were included. The updated resulted in a further 839 papers, included duplicates with the first search, of which seven were included. Across the initial and updated search, fourteen of the included studies reported costs [
Incremental cost and length of stay associated with postprocedure delirium in transcatheter and surgical aortic valve replacement patients in the United States.
Catheterization and Cardiovascular Interventions.2019; 93: 1132-1136
Hospital Delirium and Psychological Distress at 1 Year and Health-Related Quality of Life After Moderate-to-Severe Traumatic Injury Without Intracranial Hemorrhage.
Archives of Physical Medicine and Rehabilitation.2014; 95: 2382-2389
The effects of elective aortic repair, colorectal cancer surgery and subsequent postoperative delirium on long-term quality of life, cognitive functioning and depressive symptoms in older patients.
] (Figure 1). The most frequent reasons for exclusion during the full text screening were related to papers not concerning costs or HRQoL of Delirium (wrong outcome) and no or inappropriate adjustment for confounding. In the included papers on costs the prevalence of delirium ranged from 1.61% to 78.5% in these studies, combining figures from regular wards and ICUs. The majority of analysis were from the United States of America (9/14 in Costs, 5/11 in HRQOL). The assessment tools for delirium that were used in cost studies identified were: Confusion Assessment Method (CAM) (4, 29%), Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) (4, 29%), ICD-9 codes (4, 29%) and the Delirium observation screening scale (DOSS) (2, 14%). For the HRQOL papers the used assessment tools were: CAM-ICU (6, 55%), Intensive Care Delirium Screening Checklist (ICDSC) (2, 18%), Delirium symptom Interview (DSI) (1, 9%), Delirium Rating Scale (DRS) (1, 9%), and Delirium observation screening scale (DOSS) (1, 9%). All of these assessment tools have been validated [
In Table 1, study characteristics are given for the 14 included studies that reported costs. In the study of Potter et al. (2018), costs were stratified by intervention; resulting in two costs estimates from this study [
Incremental cost and length of stay associated with postprocedure delirium in transcatheter and surgical aortic valve replacement patients in the United States.
Catheterization and Cardiovascular Interventions.2019; 93: 1132-1136
]. Four studies, Potter et al, Ha et al and the two Kwak et al papers identified cases only by previously registered ICD-9 code. Nine studies retrieved data from existing databases, the remaining 5 articles looked at data prospectively. The majority of included papers (11 out of 14) based costs on direct healthcare costs and not on reimbursement, the other three did not specify the origin of the figures used.
Table 1Study characteristics and incremental costs in the presence of delirium. If reported, age and gender are shown by D = delirium; ND = no delirium.
Incremental cost and length of stay associated with postprocedure delirium in transcatheter and surgical aortic valve replacement patients in the United States.
Catheterization and Cardiovascular Interventions.2019; 93: 1132-1136
Insurance status, age, Charlson comorbidity score, pre-existing cognitive dysfunction, and APACHE II at enrolment, daily modified SOFA, daily severe sepsis status, daily mechanical ventilation status, day of ICU stay.
30 days ICU stay, mean incremental cumulative ICU costs in survivors
NA
Total costs: $22,269 (95% CI $13,897-$29,333) attributed to intensity of utilization; attributed to mortality $5,707 ($2,567–$9,824)
Infrastructure, medications, materials, instruments, fees of the surgeon, anesthesiologist, consulting physician, and so on. Converted from Won to US dollars, applying mean exchange rate in June 2015.
Studies were conducted in different countries, and costs were reported in in each national currency; all currencies were converted to US Dollars. Table 2 shows these amounts in the original currency with and in 2020 US Dollars. Exchange rates were accurate to two decimal places. An inflation rate was not taken into account. Costs were higher amongst patients with delirium in all studies. Unadjusted incremental healthcare costs (range $3,191 to $27,491) were higher than adjusted incremental costs: adjusted estimates ranged from $1,532 to $22,269.
Table 2Converted amounts to US dollars. Exchange rates were accurate to two decimal places.
Three studies stratified data by the type of costs. Ha et al. (2018) reported healthcare costs by category only in a figure, therefore values could not be derived from that study [
]. For the other studies, costs are given by cost category in Table 3. In the remaining two studies, the highest proportion of costs was made up of routine care, generally per diem hospital charges. The study of Vasilevskis, et al. (2018) gave estimates corrected for confounding; all costs were higher for patients with delirium compared with those without delirium [
All but two papers also reported the hospital length of stay for delirious and non-delirious patients (12-14, 16-23, 25). Potter et al. (2018) reported costs split by the procedure that led to hospitalization (SAVR and TAVR) and Schubert et al. (2018) reported detailed information on hospital length of stay for different subgroups (Table 1) [
Incremental cost and length of stay associated with postprocedure delirium in transcatheter and surgical aortic valve replacement patients in the United States.
Catheterization and Cardiovascular Interventions.2019; 93: 1132-1136
Eleven articles that assessed HRQOL were included in this review. Study characteristics are presented in Table 4, with delirium prevalence ranging from 8% to 56% on regular wards and ICUs. All of the studies collected data within prospective cohort studies.
Table 4Study characteristics and quality of life (QOL) for delirium. If reported, age and gender are shown by D = delirium; ND = no delirium.
Age, gender, indigenous and Australian peoples, alcohol use, comorbidities, urgent surgery, left ventricular ejection fraction, time spend on the aortic cross-clamp, and baseline QOL
SF-36 (Prior surgery, day 4, six months. (Not clear which measurement is used))
Age, gender, ASA physical status, BMI, duration of anesthesia, type of anesthesia, locoregional anesthesia, emergency surgery, temperature at SICU admission, troponin I at SICU admission, comorbidities, high-risk surgery, total RCRI, crystalloids, colloids, erythrocytes, fresh frozen plasma, platelets, and APACHE II scores
SF-36 (before and at 6 months)
Physical functioning (beta 17.402), vitality (beta 8.221) and social function (beta 16.802)
Hospital Delirium and Psychological Distress at 1 Year and Health-Related Quality of Life After Moderate-to-Severe Traumatic Injury Without Intracranial Hemorrhage.
Archives of Physical Medicine and Rehabilitation.2014; 95: 2382-2389
The effects of elective aortic repair, colorectal cancer surgery and subsequent postoperative delirium on long-term quality of life, cognitive functioning and depressive symptoms in older patients.
Age, gender, Race, BMI, FEV, LoS, Bypass time, 6MWD)
SF-12 (3,6,12,18, 24,30 and 36 months)
SF12-PCS -4.0; (95%CI: − 7.4 to − 0.7) SF12-MCS 2.2; (95%CI: − 0.7 to 5.7)
APACHE = Acute Physiology and Chronic Health Evaluation; ASA = American Society of Anesthesiologists; BMI = body mass index; CABG = Coronary Artery Bypass Grafting; CAM = Confusion Assessment Method; CumSOFA = cumulative total daily Sequential Organ Failure Assessment; DRS = Delirium Rating Scale; DSI = Delirium Symptom Interview; ICU = intensive care unit; IQR = Interquartile Range; ISS = Injury Severity Score; n = number of subjects; NIHSS = National Institutes of Health Stroke Scale; RCRI = Revised Cardiac Risk Index; SD = Standard Deviation; SF-12 = 12-Item Short Form Health Survey; SF-36 = 36-Item Short Form Health Survey; SICU = surgical intensive care unit.
In this review HRQOL was measured with different tools: the EuroQol five-dimension instrument (EQ-5D) (used once), Short Form-36 (SF-36) (used four times), SF-12 (used once), Neuro-QOL (used twice), and WHOQOL-BREF (used once). Seven of the 11 studies found statistically significant differences in HRQOL between patients with delirium and without delirium (p < 0.05). All differences were towards a lower HRQOL when delirium was detected.
Four of the included studies used the SF-36 for HRQOL assessment. These 4 studies evaluated the 8 domains of the SF-36 separately. In Table 5, differences in QOL by domain of the SF-36 are shown for patients with delirium episode versus patients without delirium. When differences were statistically significant, they indicated that delirium led to a decrease in HRQOL in all cases. However, there were some discrepancies in HRQOL when examining all estimates, regardless of statistical significance. Except for the domain on vitality, all significant estimates indicate a decrease in HRQOL when delirium was detected.
Table 5Differences in SF-36 domains between patients with delirium and patients without delirium.
Hospital Delirium and Psychological Distress at 1 Year and Health-Related Quality of Life After Moderate-to-Severe Traumatic Injury Without Intracranial Hemorrhage.
Archives of Physical Medicine and Rehabilitation.2014; 95: 2382-2389
In this review we found a potential meaningful difference in hospitalisation costs between patient having a delirium episode and those who had not. In contrast to the studies on healthcare costs, HRQOL changes due to delirium were not consistent between studies, although results may suggest a decrease in HRQOL in patients with delirium.
The diagnosis of delirium is frequently missed, only 12 to 35% of delirium cases are recognized during hospital stay [
]. In particular for studies that reviewed patient record databases; screening of delirium might not have been performed (or noted) and screening might only be considered when delirium was suspected. Not all cases of delirium would therefore be recognized and be present in the patient record databases. This might result in overestimation of the costs estimates based on patient record databases.
Hospitalisation costs were stratified by cost type in two studies, in order to determine the component that contributed most to the increase in total hospitalisation costs [
]. Clinical admissions resulted in the highest costs of the evaluated cost components. One study (Vasilevskis, et al. 2018) provided adjusted estimates [
]. Not all studies included the same costs for their estimates. Some studies only included costs that contributed to an increased length of stay, whereas other studies looked at hospital charges between patients with delirium and without delirium. Hospital reimbursement charges might give a better reflection of other additional underlying costs such as diagnostics and therapeutics. Increased length of stay is only a part of overall total hospitalisation costs.
The question whether the additional costs of delirium are reimbursed is very much related to the health system and country in question [
]. There are some indications that the additional workload for patients with delirium is not recognized by default with common scores and therefore the costs are not reimbursed [
]. Even though the costs reported in each of the studies is different, evidence for increased health care costs for patients with delirium is consistent among studies, indicating that health care costs are very likely to be increased when delirium was detected. Hospital length of stay was longer for patients with delirium, with the exception of one cardio thoracic surgery study by Brown et al [
]. The question remains what the drivers of added costs are. The additional admission days could have several different reasons such as additional time required for the treatment of the delirious episode, additional time required for treatment of the underlying illness due to delirium or the occurrence of complications due to delirium.
Different measurement tools were used to obtain HRQOL estimates, with the SF-36 as the most frequently used tool in this review. Estimates for each of the eight dimensions of the SF-36 could be retrieved from the included studies that used the instrument. Statistically significant differences were found in the domains of physical functioning, social functioning, vitality and general health perception; indicating worse HRQOL for patients with delirium. For vitality, all estimates were worse for patients with delirium without being statistically significantly different in each study. However, for all other domains (physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning, mental health), the results showed a mixed pattern, pointing towards both a negative and a positive association [
Hospital Delirium and Psychological Distress at 1 Year and Health-Related Quality of Life After Moderate-to-Severe Traumatic Injury Without Intracranial Hemorrhage.
Archives of Physical Medicine and Rehabilitation.2014; 95: 2382-2389
]. Given the reduction in consciousness, focus and attention, a decrease in HRQOL could be expected.
Strengths and limitations
A strength of this analysis is that it provides a single overview of the impact delirium can have on healthcare costs and HRQOL. A single overview of costs and HRQOL outcomes can help in better understanding of the effects delirium on patients. The combination of these outcomes may also prove to be useful for future health economic evaluations.
A limitation could be the absence of evidence synthesis. The synthesis was not performed due to multiple challenges: 1) cost charges might differ for the same procedures between countries and settings, 2) different settings were included in this review; 3) delirium might have a different impact in each setting. A different limitation could be the limited number of search terms and the limited time window used for delirium papers, which may have reduced the number of records found.
In this analysis no formal evaluation of the risk of bias was performed. Some of the included studies did not adjust for characteristics such as disease severity. These studies may have a high risk of bias and might have been removed if a formal evaluation of risk of bias was performed. However, by excluding these studies, the analysis of the remaining studies would result in insufficient data to report possible outcomes. In future research the adequate analysis and mitigation of confounders should be carefully considered as this may lead to a better insight on how total costs differ in different countries, settings and patient groups.
Conclusions
Patients experiencing delirium during hospital may incur increased hospitalisation costs mainly due to increases in length of stay. An overview of HRQOL outcomes in delirious patients showed ambiguity in outcomes with a suggestion towards a decrease in HRQOL in delirious patients. Well designed, prospective studies into healthcare costs and HRQOL in delirious and non-delirous patients may allow for stronger conclusions.
Funding
CvL, ES, MK, GWJF received a consultancy fee through Prolira B.V. for this research.
CH has not received any material or immaterial compensation and fees for creating this publication.
Conflict of interest
CH: Fees for lectures: Dräger, TapMed, Getinge, Arjo, Orion Pharma, HillRom, Stryker, Atmos, DG-Med and various non-profit institutions, academies and clinics in the health care sector; non-material conflicts of interest: functions in professional societies: DGIIN (spokesperson of the nursing section), DIVI (member), Deutsche Gesellschaft für Fachkrankenpflege und Funktionsdienste e. V. (DGF; member), founding member of the German Delir Network e. V.
Incremental cost and length of stay associated with postprocedure delirium in transcatheter and surgical aortic valve replacement patients in the United States.
Catheterization and Cardiovascular Interventions.2019; 93: 1132-1136
Hospital Delirium and Psychological Distress at 1 Year and Health-Related Quality of Life After Moderate-to-Severe Traumatic Injury Without Intracranial Hemorrhage.
Archives of Physical Medicine and Rehabilitation.2014; 95: 2382-2389
The effects of elective aortic repair, colorectal cancer surgery and subsequent postoperative delirium on long-term quality of life, cognitive functioning and depressive symptoms in older patients.
00026-5&id=gr1.jpg){kind=link}