Research in Gerontological Nursing

Empirical Research 

Postoperative Delirium Following Joint Replacement in Patients With Dementia in New South Wales, Australia: A State-Wide Retrospective Cohort Study

Xi Li, MPH; Wei Du, PhD; Anne Parkinson, PhD; Nicholas Glasgow, MD

Abstract

The objective of the current study was to investigate the variation in postoperative delirium in patients with dementia undergoing joint replacement in New South Wales (NSW) Australia public hospitals, identify factors related to its occurrence, and explore the volume–outcome relationship. The NSW Admitted Patient Data (July 2001 to June 2014) were used in this study and included patients with dementia undergoing joint replacement who were 65 or older with minor to severe comorbidities. Mixed-effect logistic models were applied to investigate hospital-level variation and factors associated with postoperative delirium. The between-hospital variability of postoperative delirium was 0.19% prior to 2008–2009 and 8.32% after 2008–2009. Hospital volume was not inversely associated with postoperative delirium rate. During 2001–2014, the incidence of postoperative delirium increased by 13% per annum (95% confidence interval [CI] 10% to 16%), while it increased by 15% per annum (95% CI 8% to 22%) after 2008–2009. An integrated approach addressing complex needs of patients with dementia may reduce the observed unwarranted variation and improve surgical outcomes. [Research in Gerontological Nursing, 13(5), 243–253.]

Abstract

The objective of the current study was to investigate the variation in postoperative delirium in patients with dementia undergoing joint replacement in New South Wales (NSW) Australia public hospitals, identify factors related to its occurrence, and explore the volume–outcome relationship. The NSW Admitted Patient Data (July 2001 to June 2014) were used in this study and included patients with dementia undergoing joint replacement who were 65 or older with minor to severe comorbidities. Mixed-effect logistic models were applied to investigate hospital-level variation and factors associated with postoperative delirium. The between-hospital variability of postoperative delirium was 0.19% prior to 2008–2009 and 8.32% after 2008–2009. Hospital volume was not inversely associated with postoperative delirium rate. During 2001–2014, the incidence of postoperative delirium increased by 13% per annum (95% confidence interval [CI] 10% to 16%), while it increased by 15% per annum (95% CI 8% to 22%) after 2008–2009. An integrated approach addressing complex needs of patients with dementia may reduce the observed unwarranted variation and improve surgical outcomes. [Research in Gerontological Nursing, 13(5), 243–253.]

Dementia commonly affects older people (Ritchie & Lovestone, 2002), the prevalence of which doubles every 5 years after age 65 (Fiest et al., 2016). It has been estimated that dementia affects >115 million people worldwide (Prince et al., 2013). Older people (age >65) with dementia often experience comorbid conditions, such as musculo-skeletal disorders, and 61% of patients with dementia are reported to have more than three comorbidities (Fox et al., 2014). Given the aging population, the prevalence of chronic health conditions will increase (Yancik et al., 2007) and may accelerate the development of dementia (Huang et al., 2015). Furthermore, the existence of dementia complicates clinical care for comorbidities (Griffith et al., 2016).

Worldwide, including in Australia, joint replacement surgery has dramatically increased in recent decades, especially in older people with dementia (bAustralian Institute of Health and Welfare [AIHW], 2013b; Bin Abd Razak & Yung, 2015). Although recognized as one of the most effective surgical interventions in pain relief and function restoration, this procedure is also associated with postoperative complications in older adults with comorbid conditions (Chung et al., 2015; Fox et al., 2014; SooHoo et al., 2006). As multiple morbidities increase with age, there will be a growing number of patients undergoing joint replacement who have at least one comorbid condition (Podmore et al., 2018), such as dementia, which is often reported in older people undergoing joint replacement (Teipel et al., 2018).

Dementia is reported to be one of the predisposing factors for delirium following joint replacement (Huang et al., 2019; Scott et al., 2015). The literature shows that patients with dementia are more likely to develop delirium (Australian Commission on Safety and Quality in Health Care [ACSQHC], 2018). Although there has been a reduction in postoperative complications following joint replacement over recent years due in part to advances in surgical techniques and perioperative care for inpatients (Bin Abd Razak & Yung, 2015; Partridge et al., 2018), evidence of postoperative delirium in patients with dementia remains scarce.

Variation in postoperative delirium rates following joint replacement among all patients has been reported in the current literature, with an incidence rate ranging from 28% in patients undergoing hip replacement to 41% in patients undergoing knee replacement (Jain et al., 2011). Such variability may be due to differences in study population, with patient-level factors such as comorbidities and older age possibly increasing the incidence of postoperative delirium (Huang et al., 2019; Scott et al., 2015). Given prior findings that higher procedure volume was related to less postoperative complications following joint replacement (Adelani et al., 2018; de Vries et al., 2011; Furuya-Kanamori et al., 2017), the current authors hypothesized that hospital-level difference in procedure volume also plays a role in variation in postoperative delirium rates. If there was an inverse relationship between procedure volume and postoperative delirium, a policy trade-off would occur between improvement in access to surgeries and enhancement in quality of surgical care.

Quantifying variation in the occurrence of postoperative delirium across hospitals and identifying factors contributing to its occurrence may be informative for effective allocation of resources, prevention of postoperative delirium, as well as improvements in the provision of safe and quality surgical care in patients with dementia. The current study thus aims to examine the extent to which post-operative delirium varies across hospitals in patients with dementia, identify factors in relation to its occurrence, and investigate the volume–outcome association.

Method

Data Sources

Analytic data derived from the New South Wales (NSW) Admitted Patient Data Collection (APDC) database were used. The APDC, administered by the NSW Department of Health, is a complete census of all inpatient services provided by public, private, and repatriation hospitals in NSW (Australian Bureau of Statistics [ABS], 2008). Based on the data use agreement with NSW Department of Health, for each de-identified separation record from public hospitals during the period of July 2001 to June 2014, patients' demographic information was extracted on age, sex, insurance status, marital status, postcode of residence, as well as clinical information on urgency of admission, hours in intensive care, medical diagnoses, and surgical procedures. Medical diagnoses for hospital admission were all coded using the Australian Modification of International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10-AM) (AIHW, 2018a). Ethical approval was sought from the relevant Science & Medical Delegated Ethics Review Committee (#2016/030).

Study Design and Population

This is a retrospective cohort study. ICD-10-AM diagnosis codes were used to ascertain patients with dementia, including those with Alzheimer's disease and other brain degenerative diseases (Table A, available in the online version of this article) (AIHW, 2018a). In terms of the Charlson Comorbidity Index, each comorbidity category based on ICD-10-AM codes has a correlative weight (from 1 to 6) and the sum of all weights leads to a single score for each patient (Sundararajan et al., 2004). As dementia is a component of the Charlson score, the score in the current study was calculated using other Charlson entities excluding dementia. Severity of comorbidities was thus defined as minor (total score = 0), moderate (total score = 1 or 2), or severe (total score >2). Patients with dementia age ≥65 were included in the current study setting following the approach taken in other studies (Hölttä et al., 2014; Inouye et al., 2014). A sensitivity analysis was also conducted, taking patients younger than 65 (n = 73) into account, which showed no material change (Figure A, available in the online version of this article). Therefore, the study population was restricted to patients with dementia who were 65 or older, having minor to severe comorbidities (total score ≥ 0), and undergoing a primary procedure for joint replacement, including knee, hip, and other joint replacement sites (Table A) (AIHW, 2018b). Although public and private hospitals perform joint replacement surgeries, the exact hospital identifier for individual private hospitals was not available; therefore, the study population was restricted to patients who were treated in NSW public hospitals for analyses in relation to hospital-level variation in the occurrence of postoperative delirium.

ICD-10-AM codes for primary procedures to joint replacement, dementia, and postoperative delirium. (Available in the public domain: Australian Institute of Health and Welfare. (2019). Principal diagnosis data cubes. Retrieved from https://www.aihw.gov.au/reports/hospitals/principal-diagnosis-data-cubes/contents/data-cubes)ICD-10-AM codes for primary procedures to joint replacement, dementia, and postoperative delirium. (Available in the public domain: Australian Institute of Health and Welfare. (2019). Principal diagnosis data cubes. Retrieved from https://www.aihw.gov.au/reports/hospitals/principal-diagnosis-data-cubes/contents/data-cubes)

Table A.

ICD-10-AM codes for primary procedures to joint replacement, dementia, and postoperative delirium. (Available in the public domain: Australian Institute of Health and Welfare. (2019). Principal diagnosis data cubes. Retrieved from https://www.aihw.gov.au/reports/hospitals/principal-diagnosis-data-cubes/contents/data-cubes)

Comparison between different selections of admission characteristics.a, b, ca Intraclass correlation (ICC) was estimated using the mixed-effect logistic regression models.b Hospitals were arranged from lowest to highest volume.c Hospitals were arranged from lowest to highest point estimate.Comparison between different selections of admission characteristics.a, b, ca Intraclass correlation (ICC) was estimated using the mixed-effect logistic regression models.b Hospitals were arranged from lowest to highest volume.c Hospitals were arranged from lowest to highest point estimate.

Figure A.

Comparison between different selections of admission characteristics.a, b, c

a Intraclass correlation (ICC) was estimated using the mixed-effect logistic regression models.

b Hospitals were arranged from lowest to highest volume.

c Hospitals were arranged from lowest to highest point estimate.

Outcomes

The ICD-10-AM code-based rules developed by the ACSQHC (2019) were further used to ascertain the occurrence of postoperative delirium, which was defined as binary (i.e., yes or no) (Table A).

Covariates

Patient demographic information included age group (categorized as 65 to 84 or 85+), sex (male or female), Accessibility and Remoteness Index of Australia (ARIA+)–based rurality of residence (urban or rural), intensive care (yes or no), private insurance (yes or no), and living status (partnered or single). Socioeconomic status was categorized into 1st quintile (the most disadvantaged) and the rest using the postcode-based Index of Relative Socio-Economic Disadvantage (IRSD) (ABS, 2013). Severity of comorbidity was treated as minor (total score = 0), moderate (total score = 1 or 2), or severe (total score >2) using the Charlson Comorbidity Index (Sundararajan et al., 2004), which was calculated based on ICD-10-AM codes. Co-morbid mental disorders (i.e., co-occurring mental disorders based on diagnoses other than delirium) were treated as yes or no. Urgency status of admission was categorized as yes or no. Given that different joint replacement procedures may result in differences in anesthesia and time to complete the procedures, joint replacement procedures were classified into three categories (i.e., knee replacement, hip replacement, and others) based on ICD-10-AM codes (Table A). Principal referral hospitals (with >20,000 separations per annum) were categorized as yes or no (Du et al., 2018). Hospital volume was categorized as low (≤33%), medium (33% to 67%), or high (>67%) of the distribution of procedure volumes.

Statistical Analysis

Statistical analyses were performed using Stata SE 15. Given that the Classification of Hospital-Acquired Diagnoses (CHADx) has been implemented nationally in Australia since July 2008 (Utz et al., 2012), it is likely the coding practice of hospital-acquired delirium would have been changed from that time (i.e., changes in coding practice from 2008 onwards may lead to a variation in the incidence of postoperative delirium across hospitals). Therefore, data were stratified into two sets (i.e., 2001–2002 to 2007–2008 and 2008–2009 to 2013–2014), and parallel analysis was conducted for these two study sets. The numbers and proportions of postoperative delirium in patients with dementia were calculated. Considering potential patient clustering within hospitals, mixed-effect logistic regression models with hospitals included as random intercepts were used to compare postoperative delirium occurrence between hospitals. Individual-level factors (e.g., age group, sex, rurality of residence) and hospital-level factors (e.g., principal referral hospital) were included in this model. The financial year of admitted patient records was treated as continuous. A model-derived intraclass correlation coefficient (ICC) was used to measure to what extent the occurrence of delirium differed across hospitals. Caterpillar plots were used to visualize variation in the occurrence of postoperative delirium across hospitals. Delirium rate was indirectly adjusted using the ratio of observed to expected incidence of delirium, and funnel plots with 95% and 99.8% control limits were created for the adjusted rate of postoperative delirium. If hospital procedure volume was <10, such hospitals were not included in the funnel plots. A series of repeated modelings were conducted as a sensitivity analysis excluding those with other coexisting major comorbid conditions, including sensory impairments, cerebral palsy and other paralytic syndrome, or cancer. Patients younger than 65 were also added in as a sensitivity analysis. p < 0.05 was set as statistically significant.

Results

Population Characteristics

During 2001–2014, there were 6,158 patients with dementia who received joint replacement in NSW public hospitals. Patients living in urban areas (n = 5,760), those who did not receive intensive care (n = 5,897), those without comorbid mental disorders (n = 5,719), or those undergoing hip replacement (n = 5,950) were overrepresented (Table 1). Approximately 14.7% and 26.8% of patients having comorbid mental disorders developed postoperative delirium from 2001–2002 to 2007–2008 and 2008–2009 to 2013–2014, respectively (Table 1). The majority of joint replacement was performed in principal referral hospitals (n = 5,098) and high-volume hospitals (n = 3,939) (Table 1).

Participant DemographicsParticipant Demographics

Table 1:

Participant Demographics

Factors in Relation to Postoperative Delirium

During the period of 2001–2002 to 2013–2014, the incidence of postoperative delirium increased by 13% per annum (95% confidence interval [CI] 10% to 16%), while it increased by 15% per annum (95% CI 8% to 22%) after 2008–2009 (Figure 1 and Table 2). Patients having intensive care (odds ratio [OR] = 1.65, 95% CI [1.16, 2.36]), those having moderate (OR = 1.32, 95% CI [1.08, 1.62]) or severe (OR = 1.46, 95% CI [1.17, 1.83]) comorbidities, or those having comorbid mental disorders (OR = 1.67, 95% CI [1.25, 2.24]) experienced an increased risk of postoperative delirium from 2001–2002 to 2013–2014 (Table 2). Different types of joint replacement procedures were not significantly associated with postoperative delirium in patients with dementia.

Procedure volume and adjusted postoperative delirium rate following joint replacement in New South Wales public hospitals.

Figure 1.

Procedure volume and adjusted postoperative delirium rate following joint replacement in New South Wales public hospitals.

Factors Associated With Postoperative Delirium in Patients With DementiaFactors Associated With Postoperative Delirium in Patients With Dementia

Table 2:

Factors Associated With Postoperative Delirium in Patients With Dementia

During the period of 2001–2002 to 2007–2008, patients having intensive care (OR = 1.78, 95% CI [1.04, 3.05]), those with moderate (OR = 1.35, 95% CI [1.01, 1.82]) or severe (OR = 1.72, 95% CI [1.26, 2.34]) comorbidities, or those with comorbid mental disorders (OR = 1.77, 95% CI [1.23, 2.54]) had an elevated risk of postoperative delirium compared with their counterparts (Table 2). During the period of 2008–2009 to 2013–2014, female patients (OR = 0.71, 95% CI [0.55, 0.91]) or those undergoing joint replacement in the acute care setting (OR = 0.64, 95% CI [0.43, 0.96]) were less likely to develop postoperative delirium (Table 2).

Hospital-Level Variation

Among a total of 48 NSW public hospitals involved in the current study, there were six (prior to 2008–2009) and 13 (after 2008–2009) hospitals excluded from Figure 2 due to their volume being <10. The mixed-effect models suggested that between-hospital variation (ICC) explained 0.19% (prior to 2008–2009) and 8.32% (after 2008–2009) of the total variance for postoperative delirium, respectively.

Hospital-level variation in adjusted postoperative delirium rate in patients with dementia undergoing joint replacement.

Figure 2.

Hospital-level variation in adjusted postoperative delirium rate in patients with dementia undergoing joint replacement.

During the period of 2001–2002 to 2007–2008, there was one high-volume hospital with adjusted postoperative delirium rate above the 99.8% upper control limits; whereas another high-volume hospital outperformed the other hospitals with low rate of postoperative delirium below the 99.8% lower control limits during the period of 2008–2009 to 2013–2014 (Figure 2). There was one medium-volume hospital and one high-volume hospital with high rate of postoperative delirium above the 99.8% upper control limits during the period 2008–2009 to 2013–2014 (Figure 2). There was no statistically significant association between postoperative delirium and hospital procedure volume (Table 2). When restricted to those without other major conditions, there was less variability prior to 2008–2009 and more variability after 2008–2009 (Figure A, available in the online version of this article). When taking younger patients into account, there was more variability after 2008–2009 (Figure B, available in the online version of this article.).

Comparison between different selections of dementia patients according to age.a, b, ca Intraclass correlation (ICC) was estimated using the mixed-effect logistic regression models.b Hospitals were arranged from lowest to highest volume.c Hospitals were arranged from lowest to highest point estimate.Comparison between different selections of dementia patients according to age.a, b, ca Intraclass correlation (ICC) was estimated using the mixed-effect logistic regression models.b Hospitals were arranged from lowest to highest volume.c Hospitals were arranged from lowest to highest point estimate.

Figure B.

Comparison between different selections of dementia patients according to age.a, b, c

a Intraclass correlation (ICC) was estimated using the mixed-effect logistic regression models.

b Hospitals were arranged from lowest to highest volume.

c Hospitals were arranged from lowest to highest point estimate.

Discussion

Although postoperative delirium is considered a preventable complication (Zegers et al., 2011), there is currently no single countermeasure to stop it from happening. The current study supports the collective efforts to reduce variation in postoperative delirium following joint replacement in patients with dementia. In response to the aging population worldwide, surgical safety and quality guidelines to improve surgical care among inpatients have been developed. A good example is the World Health Organization's “Safe Surgery Saves Lives” initiative that has produced a well-defined core set of safety standards for member states to adopt (World Alliance for Patient Safety, 2008). Current study findings of hospital-level variation signal a need for continuing efforts to reinforce appropriate intervention programs for postoperative delirium prevention and control across NSW. Implementation of more patient-centered strategies in surgical care, as widely recommended in the literature (ACSQHC, 2011), and hospital-level safety and quality initiatives (Grattan Institute, 2018) integrated with early cognitive interventions, validated locally agreed delirium care pathways, and coordinated mental health support during surgical care, may have the utility to reduce harms and improve outcomes in surgical patients with dementia.

In this multicenter retrospective cohort study, the incidence of postoperative delirium in patients with dementia undergoing joint replacement was approximately 8.9% up to 2007–8008 and 16.3% after 2008–2009. Previous findings show that the incidence of delirium ranges from 7% to 75% in patients undergoing total joint replacement, due in part to differences in study population and case ascertainment (Bin Abd Razak & Yung, 2015). In 2008, hospital-acquired delirium was defined using ICD-10-AM codes by the ACSQHC (Utz et al., 2012), which allows hospitals to identify and monitor adverse events that occur. An increasing trend of postoperative delirium was observed from 13.03% in 2008–2009 to 21.94% in 2013–2014 for patients with dementia undergoing joint replacement. Similarly, a relatively smaller increase was observed in postoperative delirium (13.7% to 19.8%) from 2008–2009 to 2013–2014 for all surgical patients in NSW public hospitals, and therefore, it is likely that the observed increase reflects an improvement in case diagnosis and ascertainment. Given that postoperative delirium could possibly remain under-diagnosed in orthopedic patients (Katznelson et al., 2010), the observed hospital variation in postoperative delirium is unlikely to be explained in principle by the improvement in diagnosis alone. Nonetheless, appropriate care should be given to interpreting the observed increase.

No hospital was observed to consistently perform better or worse during the study periods with sizeable variability in postoperative delirium across all NSW public hospitals. Between-hospital variation (ICC) accounted for 0.19% (prior to 2008–2009) and 8.32% (after 2008–2009) of the total variance for postoperative delirium, respectively. Many reasons may potentially explain such observed variation, including hospital-level differences in case-mix (Bruce et al., 2007), adoption of cognitive assessment (Ouimet et al., 2007), and appropriate staff training and education (NSW Agency for Clinical Innovation [ACI], 2014), which is consistent with previous findings (Furuya-Kanamori et al., 2017). Given it is also possible that the observed hospital-level differences found in the current setting could be related to differences in procedure protocols, future studies are needed to identify the comparative effectiveness in postoperative delirium reduction for different surgical protocols.

Due to the presence of dementia complicating surgical care for comorbidities (Griffith et al., 2016), it is imperative to implement hospital-based delirium management guidelines for surgical patients with dementia and provide education and reinforcement during its implementation. Nonetheless, it will be challenging, because of the difficulty in case identification as not all people with dementia who experience delirium are aggressive. The extra workload required for dementia and comorbidity care is well-recognized, and in addition to case management for those who have challenging behaviors, health care providers also lack knowledge about delirium assessment/screening tools (NSW ACI, 2014). However, integrated psychosocial support is necessary, and should encompass perioperative assessment of cognitive impairment, preoperative physical and cognitive exercises (Luan Erfe et al., 2018), and enhanced mental health training for staff, all of which may have the potential to prevent postoperative delirium from arising in surgical patients with dementia during their hospital care.

The observed variation in postoperative delirium was not found to be related to the hospital procedure volume. Previous studies reported that higher procedure volume was related to better outcomes such as lower mortality rate or less infections (de Vries et al., 2011; Katz et al., 2001; Kreder et al., 1997). Despite differences in the study settings, this observed lack of volume–outcome relationship may be due to the regionalization of surgical care becoming more common for surgical patients in Australia (Du et al., 2018); and therefore, surgeries are more likely to be performed at institutions where resources are centralized. Under such circumstances, the observed variation may be mainly explained by unobserved patient-level determinants rather than hospital-specific ones, such as volume per surgeon and surgeons' proficiency in operation (de Vries et al., 2011). Nevertheless, the impact of hospital procedure volume on postsurgical delirium is still unclear due in part to lack of information on these unobserved factors in the current study. Further studies are therefore needed to explore hospital-level perioperative processes and care pathways for a better understanding of volume–outcome association in surgical patients with dementia.

Consistent with previous studies (Huang et al., 2019; Kassahun, 2018; Scott et al., 2015; SooHoo et al., 2006), post-operative delirium was commonplace in older people with dementia, especially in those with severe comorbidities or receiving intensive care. Such commonly observed delirium in older people is largely preventable (ACSQHC, 2016). It is important to develop patient-centered initiatives as an integral part of postoperative care strategies to improve surgical outcomes for patients with dementia undergoing joint replacement, in which case, patients, their carers/family members, and health care professionals are expected to collaboratively design an individualized care plan that allows surgical goals to be well tailored to the patient's needs (NSW ACI, 2014). In addition, pre-screening and risk stratification using the short form of the Informant Questionnaire on Cognitive Decline in the Elderly has been considered a useful preventive strategy in decreasing the risk for delirium following total joint arthroplasty (Bin Abd Razak & Yung, 2015). For surgical patients with dementia, prevention initiatives to reduce postoperative complications may include the development and provision of a psychosocial care pathway, recognition and risk assessment of postoperative delirium, efficient perioperative management, and early interventions to improve cognitive and functional outcomes (Loughlin & Brown, 2015). The current study reinforces the need for continuing investment in mental health support in this surgical patient group.

Limitations

There are several limitations in this study. First, perioperative scenarios were not able to be fully captured using the hospital administrative APDC data, for example, lack of detailed information on modifiable factors such as type of anesthesia, peri-operative medications, and postoperative pain control, which are all important risk factors for postoperative delirium following joint replacement (Weinstein et al., 2018), thereby making it difficult to identify any effects resulting from un-measured confounders or by chance alone. Second, there is a growing need for joint replacement surgeries in the aging population (Bin Abd Razak & Yung, 2015; Norton, 1998), which was not fully reflected in the current study as procedure volumes changed over time. Fluctuation in procedure volume probably relates to the exclusion of private hospitals in this study; consequently, results should be interpreted with caution. Third, there was possible misclassification bias due to use of ICD-10-AM codes for case selection (e.g., possible under-diagnosis of postoperative delirium [Scott et al., 2015] and dementia [AIHW, 2013a]). Fourth, most of the variables were treated as binary in the study to avoid over-parametrization when modeling categorical variables. Although sensitivity analysis did not show any material changes when using full information of selected variables, special care is warranted for result interpretation. Fifth, the existence of any hospital-specific policies or pathways to mitigate the risk of postoperative delirium were unable to be determined, which may explain in part the observed variation across hospitals. However, findings present the challenges health care providers are facing. Given that delirium during hospital care places heavy burden on service provision and resource allocation (Scott et al., 2015), the authors advocate for the ongoing need for mental health support to optimize outcomes in patients with dementia undergoing joint replacement.

Conclusion

There was variation across hospitals with regard to the incidence of postoperative delirium in patients with dementia undergoing joint replacement, suggesting continuing need for mental health support during postoperative care in this vulnerable population. Intervention strategies such as early screening and patient-centered initiatives could potentially prevent the development of postoperative delirium in patients with dementia and address their complex needs. Future studies may focus more on hospital-level perioperative processes to optimize postoperative outcomes in this surgical patient group.

References

  • Adelani, M. A., Keller, M. R., Barrack, R. L. & Olsen, M. A. (2018). The impact of hospital volume on racial differences in complications, readmissions, and emergency department visits following total joint arthroplasty. The Journal of Arthroplasty, 33(2), 309–315.e20. doi:10.1016/j.arth.2017.09.034 [CrossRef] PMID:29066108
  • Australian Bureau of Statistics. (2008). NSW Health Department, Admitted Patient Data Collection 2008. Retrieved from https://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/1368.1Explanatory%20Notes1452007
  • Australian Bureau of Statistics. (2013). The index of relative socio-economic disadvantage (IRSD). Retrieved from http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/2033.0.55.001main+features100052011
  • Australian Commission on Safety and Quality in Health Care. (2011). Patient-centred care: Improving quality and safety through partnerships with patients and consumers. Retrieved from https://www.safetyandquality.gov.au/sites/default/files/migrated/PCC_Paper_August.pdf
  • Australian Commission on Safety and Quality in Health Care. (2016). Delirium clinical care standard. Retrieved from https://www.safetyandquality.gov.au/wp-content/uploads/2016/07/Delirium-Clinical-Care-Standard-Web-PDF.pdf
  • Australian Commission on Safety and Quality in Health Care. (2018). Selected best practices and suggestions for improvement for clinicians and health system managers: Hospital-acquired complication 11: Delirium. Retrieved from https://www.safetyandquality.gov.au/wp-content/uploads/2018/03/Delirium-detailed-fact-sheet.pdf
  • Australian Commission on Safety and Quality in Health Care. (2019). Hospital-acquired complications (HACs) list - specifications -version 3.0. Retrieved from https://www.safetyandquality.gov.au/publications-and-resources/resource-library/hospital-acquired-complications-hacs-list-specifications-version-30
  • Australian Institute of Health and Welfare. (2013a). Dementia care in hospitals: Costs and strategies. Retrieved from https://www.aihw.gov.au/getmedia/9c7deaeb-1b8c-4e40-8763-cc01560642cc/14347_20130502.pdf.aspx
  • Australian Institute of Health and Welfare. (2013b). Rise in hospitalizations for osteoarthritis leads to rise in joint replacements. Retrieved from https://www.aihw.gov.au/news-media/media-releases/2013/2013-may/rise-in-hospitalisations-forosteoarthritis-leads
  • Australian Institute of Health and Welfare. (2018a). Principal diagnosis data cubes. Retrieved from https://www.aihw.gov.au/reports/hospitals/principal-diagnosis-data-cubes/contents/data-cubes
  • Australian Institute of Health and Welfare. (2018b). Procedures data cubes. Retrieved from https://www.aihw.gov.au/reports/hospitals/procedures-data-cubes/contents/data-cubes
  • Bin Abd Razak, H. R. & Yung, W. Y. A. (2015). Postoperative delirium in patients undergoing total joint arthroplasty: A systematic review. The Journal of Arthroplasty, 30(8), 1414–1417 doi:10.1016/j.arth.2015.03.012 [CrossRef] PMID:25818653
  • Bruce, A. J., Ritchie, C. W., Blizard, R., Lai, R. & Raven, P. (2007). The incidence of delirium associated with orthopedic surgery: A meta-analytic review. International Psychogeriatrics, 19(2), 197–214 doi:10.1017/S104161020600425X [CrossRef] PMID:16973101
  • Chung, K. S., Lee, J. K., Park, J. S. & Choi, C. H. (2015). Risk factors of delirium in patients undergoing total knee arthroplasty. Archives of Gerontology and Geriatrics, 60(3), 443–447 doi:10.1016/j.archger.2015.01.021 [CrossRef] PMID:25704295
  • de Vries, L. M., Sturkenboom, M. C., Verhaar, J. A., Kingma, J. H. & Stricker, B. H. (2011). Complications after hip arthroplasty and the association with hospital procedure volume. Acta Orthopaedica, 82(5), 545–552 doi:10.3109/17453674.2011.618907 [CrossRef] PMID:21895498
  • Du, W., Glasgow, N., Smith, P., Clements, A. & Sedrakyan, A. (2018). Major inpatient surgeries and in-hospital mortality in New South Wales public hospitals in Australia: A state-wide retrospective cohort study. International Journal of Surgery, 50, 126–132 doi:10.1016/j.ijsu.2017.12.023 [CrossRef] PMID:29288807
  • Fiest, K. M., Jetté, N., Roberts, J. I., Maxwell, C. J., Smith, E. E., Black, S. E., Blaikie, L., Cohen, A., Day, L., Holroyd-Leduc, J., Kirk, A., Pearson, D., Pringsheim, T., Venegas-Torres, A. & Hogan, D. B. (2016). The prevalence and incidence of dementia: A systematic review and meta-analysis. The Canadian Journal of Neurological Sciences, 43(Suppl. 1), S3–S50 doi:10.1017/cjn.2016.18 [CrossRef] PMID:27307127
  • Fox, C., Smith, T., Maidment, I., Hebding, J., Madzima, T., Cheater, F., Cross, J., Poland, F., White, J. & Young, J. (2014). The importance of detecting and managing comorbidities in people with dementia. Age and Ageing, 43(6), 741–743 doi:10.1093/ageing/afu101 [CrossRef] PMID:25038831
  • Furuya-Kanamori, L., Doi, S. A. R., Smith, P. N., Bagheri, N., Clements, A. C. A. & Sedrakyan, A. (2017). Hospital effect on infections after four major surgical procedures: Outlier and volume-outcome analysis using all-inclusive state data. The Journal of Hospital Infection, 97(2), 115–121 doi:10.1016/j.jhin.2017.05.021 [CrossRef] PMID:28576454
  • Grattan Institute. (2018). All complications should count: Using our data to make hospitals safer. Retrieved from https://grattan.edu.au/wp-content/uploads/2018/02/897-All-complications-should-count.pdf
  • Griffith, L. E., Gruneir, A., Fisher, K., Panjwani, D., Gandhi, S., Sheng, L., Gafni, A., Patterson, C., Markle-Reid, M. & Ploeg, J. (2016). Patterns of health service use in community living older adults with dementia and comorbid conditions: A population-based retrospective cohort study in Ontario, Canada. BMC Geriatrics, 16(1), 177 doi:10.1186/s12877-016-0351-x [CrossRef] PMID:27784289
  • Hölttä, E. H., Laurila, J. V., Laakkonen, M. L., Strandberg, T. E., Tilvis, R. S. & Pitkala, K. H. (2014). Precipitating factors of delirium: Stress response to multiple triggers among patients with and without dementia. Experimental Gerontology, 59, 42–46 doi:10.1016/j.exger.2014.04.014 [CrossRef] PMID:24809631
  • Huang, J., Sprung, J. & Weingarten, T. N. (2019). Delirium following total joint replacement surgery. Bosnian Journal of Basic Medical Sciences, 19(1), 81–85 doi:10.17305/bjbms.2018.3653 [CrossRef] PMID:29984677
  • Huang, S. W., Wang, W. T., Chou, L. C., Liao, C. D., Liou, T. H. & Lin, H. W. (2015). Osteoarthritis increases the risk of dementia: A nationwide cohort study in Taiwan. Scientific Reports, 5(1), 10145 doi:10.1038/srep10145 [CrossRef] PMID:25984812
  • Inouye, S. K., Westendorp, R. G. J. & Saczynski, J. S. (2014). Delirium in elderly people. Lancet, 383(9920), 911–922 doi:10.1016/S0140-6736(13)60688-1 [CrossRef] PMID:23992774
  • Jain, F. A., Brooks, J. O. III. , Larsen, K. A., Kelly, S. E., Bode, R. H., Sweeney, G. A. & Stern, T. A. (2011). Individual risk profiles for postoperative delirium after joint replacement surgery. Psychosomatics, 52(5), 410–416 doi:10.1016/j.psym.2011.03.011 [CrossRef] PMID:21907058
  • Kassahun, W. T. (2018). The effects of pre-existing dementia on surgical outcomes in emergent and nonemergent general surgical procedures: Assessing differences in surgical risk with dementia. BMC Geriatrics, 18(1), 153–161 doi:10.1186/s12877-018-0844-x [CrossRef] PMID:29970028
  • Katz, J. N., Losina, E., Barrett, J., Phillips, C. B., Mahomed, N. N., Lew, R. A., Guadagnoli, E., Harris, W. H., Poss, R. & Baron, J. A. (2001). Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States Medicare population. Journal of Bone and Joint Surgery American Volume, 83(11), 1622–1629 doi:10.2106/00004623-200111000-00002 [CrossRef] PMID:11701783
  • Katznelson, R., Djaiani, G., Tait, G., Wasowicz, M., Sutherland, A. M., Styra, R., Lee, C. & Beattie, W. S. (2010). Hospital administrative database underestimates delirium rate after cardiac surgery. Canadian Journal of Anaesthesia, 57(10), 898–902 doi:10.1007/s12630-010-9355-8 [CrossRef] PMID:20645040
  • Kreder, H. J., Deyo, R. A., Koepsell, T., Swiontkowski, M. F. & Kreuter, W. (1997). Relationship between the volume of total hip replacements performed by providers and the rates of postoperative complications in the state of Washington. Journal of Bone and Joint Surgery American Volume, 79(4), 485–494 doi:10.2106/00004623-199704000-00003 [CrossRef] PMID:9111392
  • Loughlin, D. & Brown, M. (2015). Improving surgical outcomes for people with dementia. Nursing Standard, 29(38), 50–58 doi:10.7748/ns.29.38.50.e9925 [CrossRef] PMID:25990182
  • Luan Erfe, B. M., Boehme, J., Erfe, J. M., Brovman, E. Y., Bader, A. M. & Urman, R. D. (2018). Postoperative outcomes in primary total knee arthroplasty patients with preexisting cognitive impairment: A systematic review. Geriatric Orthopaedic Surgery & Rehabilitation, 9, 2151459318816482 doi:10.1177/2151459318816482 [CrossRef] PMID:30622833
  • Norton, E. C., Garfinkel, S. A., McQuay, L. J., Heck, D. A., Wright, J. G., Dittus, R. & Lubitz, R. M. (1998). The effect of hospital volume on the in-hospital complication rate in knee replacement patients. Health Services Research, 33(5), 1191–1210 PMID:9865217
  • NSW Agency for Clinical Innovation. (2014). ACI Aged Health Network: Key principles for care of confused hospitalized older persons. Retrieved from https://www.aci.health.nsw.gov.au/__data/assets/pdf_file/0006/249171/CHOPS-key-principles1-2-web.pdf
  • Ouimet, S., Kavanagh, B. P., Gottfried, S. B. & Skrobik, Y. (2007). Incidence, risk factors and consequences of ICU delirium. Intensive Care Medicine, 33(1), 66–73 doi:10.1007/s00134-006-0399-8 [CrossRef] PMID:17102966
  • Partridge, T., Jameson, S., Baker, P., Deehan, D., Mason, J. & Reed, M. R. (2018). Ten-year trends in medical complications following 540,623 primary total hip replacements from a national database. The Journal of Bone and Joint Surgery American Volume, 100(5), 360–367. doi:10.2106/JBJS.16.01198 [CrossRef] PMID:29509612
  • Podmore, B., Hutchings, A., van der Meulen, J., Aggarwal, A. & Konan, S. (2018). Impact of comorbid conditions on outcomes of hip and knee replacement surgery: A systematic review and meta-analysis. BMJ Open, 8(7), e021784 doi:10.1136/bmjopen-2018-021784 [CrossRef] PMID:29997141
  • Prince, M., Bryce, R., Albanese, E., Wimo, A., Ribeiro, W. & Ferri, C. P. (2013). The global prevalence of dementia: A systematic review and metaanalysis. Alzheimer's & Dementia, 9(1), 63–75.e2. doi:10.1016/j.jalz.2012.11.007 [CrossRef] PMID:23305823
  • Ritchie, K. & Lovestone, S. (2002). The dementias. Lancet, 360(9347), 1759–1766 doi:10.1016/S0140-6736(02)11667-9 [CrossRef] PMID:12480441
  • Scott, J. E., Mathias, J. L. & Kneebone, A. C. (2015). Incidence of delirium following total joint replacement in older adults: A meta-analysis. General Hospital Psychiatry, 37(3), 223–229 doi:10.1016/j.genhosppsych.2015.02.004 [CrossRef] PMID:25774049
  • SooHoo, N. F., Lieberman, J. R., Ko, C. Y. & Zingmond, D. S. (2006). Factors predicting complication rates following total knee replacement. The Journal of Bone and Joint Surgery American Volume, 88(3), 480–485 PMID:16510811
  • Sundararajan, V., Henderson, T., Perry, C., Muggivan, A., Quan, H. & Ghali, W. A. (2004). New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. Journal of Clinical Epidemiology, 57(12), 1288–1294 doi:10.1016/j.jclinepi.2004.03.012 [CrossRef] PMID:15617955
  • Teipel, S. J., Fritze, T., Ellenrieder, M., Haenisch, B., Mittelmeier, W. & Doblhammer, G. (2018). Association of joint replacement surgery with incident dementia diagnosis in German claims data. International Psychogeriatrics, 30(9), 1375–1383 doi:10.1017/S1041610217002976 [CrossRef] PMID:29559010
  • Utz, M., Johnston, T. & Halech, R. (2012). A review of the classification of hospital-acquired diagnoses (CHADx). Retrieved from https://www.health.qld.gov.au/__data/assets/pdf_file/0028/362845/techreport_12.pdf
  • Weinstein, S. M., Poultsides, L., Baaklini, L. R., Mörwald, E. E., Cozowicz, C., Saleh, J. N., Arrington, M. B., Poeran, J., Zubizarreta, N. & Memtsoudis, S. G. (2018). Postoperative delirium in total knee and hip arthroplasty patients: A study of perioperative modifiable risk factors. British Journal of Anaesthesia, 120(5), 999–1008 doi:10.1016/j.bja.2017.12.046 [CrossRef] PMID:29661417
  • World Alliance for Patient Safety. (2008). Safe surgery saves lives. Retrieved from https://www.who.int/patientsafety/safesurgery/knowledge_base/SSSL_Brochure_finalJun08.pdf
  • Yancik, R., Ershler, W., Satariano, W., Hazzard, W., Cohen, H. J. & Ferrucci, L. (2007). Report of the National Institute on Aging task force on comorbidity. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 62(3), 275–280 doi:10.1093/gerona/62.3.275 [CrossRef] PMID:17389724
  • Zegers, M., de Bruijne, M. C., de Keizer, B., Merten, H., Groenewegen, P. P., van der Wal, G. & Wagner, C. (2011). The incidence, root-causes, and outcomes of adverse events in surgical units: Implication for potential prevention strategies. Patient Safety in Surgery, 5(1), 13–23 doi:10.1186/1754-9493-5-13 [CrossRef] PMID:21599915

Participant Demographics

DemographicStudy Period
2001–2002 to 2013–20142001–2002 to 2007–20082008–2009 to 2013–2014
Total (n)Delirium (n, %)Total (n)Delirium (n, %)Total (n)Delirium (n, %)
Age (years)
  65 to 842,938364 (12.39)1,791173 (9.66)1,147191 (16.65)
  85+3,220385 (11.96)1,768150 (8.48)1,452235 (16.18)
Sex
  Male1,650260 (15.76)913106 (11.61)737154 (20.90)
  Female4,508489 (10.85)2,646217 (8.20)1,862272 (14.61)
Rurality of residence
  Urban5,760720 (12.50)3,330313 (9.40)2,430407 (16.75)
  Rural39829 (7.29)22910 (4.37)16919 (11.24)
Private insurance
  No4,898561 (11.45)2,888250 (8.66)2,010311 (15.47)
  Yes1,260188 (14.92)67173 (10.88)589115 (19.52)
IRSD (quintile)
  Most disadvantaged1,245121 (9.72)66947 (7.03)57674 (12.85)
  Less disadvantaged4,913628 (12.78)2,890276 (9.55)2,023352 (17.40)
Intensive care
  No5,897702 (11.90)3,410303 (8.89)2,487399 (16.04)
  Yes26147 (18.01)14920 (13.42)11227 (24.11)
Severity of comorbidities
  Minor4,071450 (11.05)2,218172 (7.75)1,853278 (15.00)
  Moderate1,218167 (13.71)76678 (10.18)45289 (19.69)
  Severe869132 (15.19)57573 (12.70)29459 (20.07)
Comorbid mental disorders
  No5,719675 (11.80)3,230275 (8.51)2,489400 (16.07)
  Yes39670 (17.68)29944 (14.72)9726 (26.80)
Living status
  Partnered1,961281 (14.33)1,044108 (10.34)917173 (18.87)
  Single4,197468 (11.15)2,515215 (8.55)1,682253 (15.04)
Urgency
  No714112 (15.69)47051 (10.85)24461 (25.00)
  Yes5,444637 (11.70)3,089272 (8.81)2,355365 (15.50)
Joint replacement
  Hip5,950713 (11.98)3,433310 (9.03)2,517403 (16.01)
  Knee20035 (17.50)11812 (10.17)8223 (28.05)
  Other sites81 (12.50)81 (12.50)
Principal referral hospital
  No1,060108 (10.19)67044 (6.57)39064 (16.41)
  Yes5,098641 (12.57)2,889279 (9.66)2,209362 (16.39)
Hospital volume
  Low36729 (7.90)23611 (4.66)13118 (13.74)
  Medium1,852196 (10.58)1,03569 (6.67)817127 (15.54)
  High3,939524 (13.30)2,288243 (10.62)1,651281 (17.02)

Factors Associated With Postoperative Delirium in Patients With Dementia

CharacteristicOdds Ratio [95% Confidence Interval]
Study Period
2001–2002 to 2013–20142001–2002 to 2007–20082008–2009 to 2013–2014
Age (years)
  65 to 84
  85+0.98 [0.83, 1.16]0.94 [0.73, 1.20]1.04 [0.82, 1.31]
Sex
  Male
  Female0.78 [0.62, 0.90]0.76 [0.58, 1.01]0.71 [0.55, 0.91]
Rurality of residence
  Urban
  Rural0.95 [0.60, 1.51]0.72 [0.34, 1.53]0.93 [0.51, 1.70]
Private insurance
  No
  Yes1.17 [0.96, 1.42]1.33 [0.99, 1.79]1.13 [0.87, 1.47]
Socioeconomic status (quintile)
  Most disadvantaged
  Less disadvantaged1.14 [0.88, 1.46]1.14 [0.79, 1.64]1.10 [0.79, 1.52]
Intensive care
  No
  Yes1.65 [1.16, 2.36]1.78 [1.04, 3.05]1.50 [0.92, 2.45]
Severity of comorbidities
  Minor
  Moderate1.32 [1.08, 1.62]1.35 [1.01, 1.82]1.27 [0.95, 1.68]
  Severe1.46 [1.17, 1.83]1.72 [1.26, 2.34]1.24 [0.88, 1.73]
Comorbid mental disorders
  No
  Yes1.67 [1.25, 2.24]1.77 [1.23, 2.54]1.58 [0.97, 2.59]
Living status
  Partnered
  Single0.86 [0.71, 1.03]0.94 [0.71, 1.25]0.82 [0.64, 1.04]
Urgency
  No
  Yes0.79 [0.59, 1.04]0.94 [0.63, 1.40]0.64 [0.43, 0.96]
Joint replacement
  Other sites
  Knee1.25 [0.14, 11.22]0.89 [0.09, 8.92]
  Hip0.99 [0.11, 8.67]0.75 [0.08, 7.09]0.70 [0.37, 1.31]
Principal referral hospital
  No
  Yes1.11 [0.64, 1.93]1.18 [0.65, 2.12]1.10 [0.54, 2.22]
Hospital volume
  Low
  Medium0.09 [0.55, 2.15]1.18 [0.53, 2.63]1.06 [0.45, 2.52]
  High1.50 [0.71, 3.16]1.62 [0.65, 4.05]1.37 [0.53, 3.56]
Financial years1.13 [1.10, 1.16]1.06 [1.00, 1.13]1.15 [1.08, 1.22]

ICD-10-AM codes for primary procedures to joint replacement, dementia, and postoperative delirium. (Available in the public domain: Australian Institute of Health and Welfare. (2019). Principal diagnosis data cubes. Retrieved from https://www.aihw.gov.au/reports/hospitals/principal-diagnosis-data-cubes/contents/data-cubes)

ICD-10-AM codes
Joint replacementKnee: 4951700 Hemiarthroplasty of knee; 4951800 Total arthroplasty of knee, unilateral; 4951900 Total arthroplasty of knee, bilateral; 4953401 Total replacement arthroplasty of patellofemoral joint of knee; 4952100 Total arthroplasty of knee with bone graft to femur, unilateral; 4952101 Total arthroplasty of knee with bone graft to femur, bilateral; 4952102 Total arthroplasty of knee with bone graft to tibia, unilateral; 4952103 Total arthroplasty of knee with bone graft to tibia, bilateral; 4952400 Total arthroplasty of knee with bone graft to femur and tibia, unilateral; 4952401 Total arthroplasty of knee with bone graft to femur and tibia, bilateral; 4952700 Revision of total arthroplasty of knee; 4953000 Revision of total arthroplasty of knee with bone graft to femur; 4953001 Revision of total arthroplasty of knee with bone graft to tibia; 4953300 Revision of total arthroplasty of knee with bone graft to femur and tibia; 4955400 Revision of total arthroplasty of knee with anatomic specific allograft. Hip: 4752200 Hemiarthroplasty of femur; 4931500 Partial arthroplasty of hip; 4931800 Total arthroplasty of hip, unilateral; 4931900 Total arthroplasty of hip, bilateral; 4932400 Revision of total arthroplasty of hip; 4932700 Revision of total arthroplasty of hip with bone graft to acetabulum; 4933000 Revision of total arthroplasty of hip with bone graft to femur; 4933300 Revision of total arthroplasty of hip with bone graft to acetabulum and femur; 4933900 Revision of total arthroplasty of hip with anatomic specific allograft to acetabulum; 4934200 Revision of total arthroplasty of hip with anatomic specific allograft to femur; 4934500 Revision of total arthroplasty of hip with anatomic specific allograft to acetabulum and femur; 4934600 Revision of partial arthroplasty of hip. Other sites: 5012700 Arthroplasty of joint, mot elsewhere; 5021503 En bloc resection of lesion of soft tissue affecting the long bones of lower limb, with intercalary reconstruction using prothesis; 5021803 En bloc resection of lesion of long bone of lower limb with replacement of adjacent joint.
DementiaF00 Dementia in Alzheimer's disease; F01 Vascular dementia; F02 Dementia in other diseases classified elsewhere; F03 Unspecified dementia; G30 Alzheimer's disease; G31 Other degenerative diseases of nervous system, not elsewhere classified; G32 Other degenerative diseases of nervous system in diseases classified elsewhere.
DeliriumF05.0 Delirium not superimposed on dementia, so described; F05.1 Delirium superimposed on dementia; F05.8 Other delirium (includes delirium of mixed origin); F05.9 Delirium, unspecified; R41.0 Disorientation, unspecified.
Authors

Ms. Li is PhD Student, Dr. Du is Lecturer, Dr. Parkinson is Senior Fellow, and Dr. Glasgow is Professor, Research School of Population Health, Australian National University, Acton ACT; and Ms. Li is also PhD Student, Centre for Research and Action in Public Health, University of Canberra, Bruce ACT, Australia.

The authors have disclosed no potential conflicts of interest, financial or otherwise. This work was supported by an Australian National Health and Medical Research Council Senior Research Fellowship (#1058878). The study sponsors had no further role in the study design, data collection, statistical analyses, interpretation of results, writing of the article, or the decision to submit it for publication.

The authors thank the NSW Department of Health for providing the Admitted Patient Data Collection data.

Address correspondence to Xi Li, MPH, PhD Student, Centre for Research and Action in Public Health, Building 22, University Drive, University of Canberra, Bruce ACT 2601, Australia; e-mail: xi.li@canberra.edu.au.

Received: June 15, 2019
Accepted: October 16, 2019
Posted Online: February 27, 2020

10.3928/19404921-20200214-01

Sign up to receive

Journal E-contents