Trends of Australian emergency department care for people with mental health diagnoses : implications for service provision and policy development

Background: Mental illnesses are the leading cause of disability in the developed world. The treatment and care of people living with mental illnesses historically occurred in institutions. Deinstitutionalisation and reductions in psychiatric beds have since occurred in many developed countries including Canada, the US and Australia. In Australia, closure of psychiatric hospitals became national policy in 1992 through the First National Mental Health Plan. A potential consequence of the shift of care towards the community, is increased presentations to general hospital emergency departments (EDs). In Australia, no long-term trend analysis has been done to provide a comprehensive understanding of the burden of presentations with a MH diagnosis (MH\\(_{dx}\\)) on EDs. Method and Results of Study 1: A systematic review aims to determine if there has been an increase in the percentage of MH-related ED (MHrED) presentations between 1985 and 2015 by country. All major databases were searched for English-language peer-reviewed literature assessing ED presentations for patients with a MH-related diagnosis including psychotic disorders (ICD-9 codes 295.00-295.95 or ICD 10 codes F20-F29). There were 81 journal articles from 17 countries identified, which were primarily from the US (n=43), then Australia (n=15). National data was only reported for the US. More than nine-tenths of the studies were published in the last half of the study period, and more than half after 2003. Most studies encompassed all age-groups (n=47); just under a third reported on children (n=23). Due to study heterogeneity and limitations of data presentation, a narrative synthesis was undertaken. Overall, nationalstudies from the US for all age-groups showed an upward trend in the proportion of MHrED presentations since 1992, averaging 5.4% for 1992- 2001 and 6.5% for 1997-2003. By year, proportions increased from 3.0% in 2001 to 3.5% in 2006, and from 6.6% in 2008 to 7.3% in 2010. Australian studies reported data for individual hospitals and by region, with upward trends found for all age-groups, particularly in NSW (2.9% in 1999 to 3.7% in 2006) and SA (0.3% in 1993 to 4.3% in 2002). Methods of Study 2 to Study 5: Study 2 to Study 5 examined the trends of all ED and MH\\(_{dx}\\) presentations to Australian public hospital EDs, nationally, by jurisdiction and by characteristic of patient/presentation between 2004-05 to 2016-17, for Australia and each jurisdiction, subject to data availability. The characteristics assessed included sex of patient (Study 2), age group (Study 3), MH diagnostic group (Study 4), acuity and admission status (Study 5). These studies used aggregated data from the National Non-Admitted Patient Emergency Department Care Database (NNAPEDCD), a part of the National Minimum Data Set (NMDS), as published by the Australian Institute of Health and Welfare (AIHW). Data were derived from two series: Australian Hospital Statistics (AHS) and its special series for mental health, Mental Health Services in Australia (MHSA). The MH\\(_{dx}\\) presentation was defined as those with a principal diagnosis falling into the International Classification of Diseases, 10th Revision, Australian Modification (ICD-10-AM) codes F00-F99. Outcomes of these studies were the number and rate of presentations per 10,000 population (population rate‚ÄövÑvp), and proportion of all presentations that were MH\\(_{dx}\\) and growth, primarily assessed as x-fold change. Univariable generalised linear regression models were used to assess the trends. Breakpoints in the trends were tested using statistical goodness of fit tests, optimised by likelihood ratio tests. Multivariable generalised linear regression models were employed to: (1) compare the proportion of MH\\(_{dx}\\) presentations between jurisdictions and the national average, and between sexes (in Study 2); (2) compare the proportion of MH\\(_{dx}\\) presentations across age groups (in Study 3); and (3) compare the population rates of high acuity presentations and admitted presentations between jurisdictions and the national average (in Study 5). Statistical significance was identified with p-value ‚Äöv¢¬ß0.05 Results from Study 2: MH\\(_{dx}\\) presentations increased from 3.3% to 3.7% of all ED presentations with a diagnosis. Most growth occurred between 2010-11 and 2015- 16. NT had the highest level of MH\\(_{dx}\\) presentations per 10,000 population, more than double the Australian average. The proportion of MH\\(_{dx}\\) presentations was highest in SA in most years, and the average annual proportion of MH\\(_{dx}\\) presentations was statistically significantly higher than the national average in SA, QLD, and WA. The proportion of MH\\(_{dx}\\) presentations increased in each jurisdiction, with significant increases for VIC, QLD, WA, ACT, and the NT. Males experienced greater numbers and rates of all ED and MH\\(_{dx}\\) presentations, while the proportion of MH\\(_{dx}\\) presentations was 8% higher for females. Results from Study 3: children (0-14 years), followed by older persons (‚Äöv¢‚Ä¢65 years) had the highest ED utilisation; while youth (15-24 years) and younger adults (25-34 years) predominated for MH\\(_{dx}\\) presentations. As a proportion of all presentations, MH\\(_{dx}\\) presentations were lowest in children, and highest in people 35- 44 years (13.2-times higher than for children). The rate of increase in MH\\(_{dx}\\) presentations was higher than for all presentations in all age-groups, reaching almost 4-times higher for children. Results from Study 4: Increased population rates of MH\\(_{dx}\\) presentations were observed for most diagnostic groups, except for mental retardation (F70-F79) and mood disorders (F30-F39). The greatest absolute increase was for psychoactive substance use-related disorders, including alcohol use (F10-F19), and relative increase, unspecified mental disorder (F99). There was differentiation across jurisdictions regarding population rates of, and growth in MH\\(_{dx}\\) presentations for each MH-diagnostic group. In 2016-17, population rates, at least twice the national average, were observed for psychoactive substance-use (F10-F19), schizophrenia and psychotic disorders (F20-F29) and childhood and adolescence onset disorders (F90-F98) in the NT, F90-F98 in SA, personality disorders (F60-F69) in TAS and unspecified mental disorder (F99) in NSW. The most marked growth was observed for F99 in NSW. Results from Study 5: Population rates of high acuity presentations and presentations admitted increased 2.2-fold (to 417.7/10,000 people) and 1.6-fold (to 1,0256/10,000 people) respectively for all presentations and 2.1-fold (to 16.3/10,000 people) and 2.0-fold (to 43.5/10,000 people) for MH\\(_{dx}\\) presentations. Trends differed between all and MH\\(_{dx}\\) presentations for both high acuity presentations and presentations admitted. The proportion of high acuity presentations within all ED presentations increased from 9.1% to 13.5% and from 11.2% to 14.4% for MH\\(_{dx}\\) presentations; high acuity MH\\(_{dx}\\) presentations as proportion of all high acuity ED presentations was consistent. The proportion of presentations admitted within all ED presentations increased from 28.9% to 32.5%, and from 32.2% to 38.6% for MH\\(_{dx}\\) presentations, with the proportion of all presentations admitted that were for a MH\\(_{dx}\\) increasing from 3.4% to 4.2%. Annual increases in all outcomes were observed in most jurisdictions. Discussion: Internationally, there was an increasing number of studies on MHrED presentations in the peer-reviewed literature between 1985 and 2015, which may reflect an increase in clinical concern regarding MHrED presentations. An observed increase in the percentage of MHrED presentations was found at national level in the US and at individual hospital and jurisdictional levels in Australia, highlighting increasing demands on general hospital EDs. In Australia, the proportion of MH\\(_{dx}\\) presentations, narrowly defined, has increased in all Australian jurisdictions between 2004-05 to 2016-17, but particularly since 2010-11. To better identify the impact of MH on ED presentations, the AIHW should consider expanding the breadth of MH diagnoses they report. Differences between jurisdictions indicate jurisdictional specific issues. However, significant or upwards trends of MH\\(_{dx}\\) presentations across all jurisdictions indicates generic issues necessitating concern and policy development at a national level. For age-specific issues, children and older persons were found as the two highest groups of ED users, while children, youth, and younger adults had the greatest increase in the population rate of MH\\(_{dx}\\) presentations. Across MH diagnostic groups, there were increases in the population rates of MH\\(_{dx}\\) presentations for most, but particularly psychoactive substance use-related disorders. Increasing rates of MH\\(_{dx}\\) presentations and MH\\(_{dx}\\) presentations admitted underscores increasing (but underestimated) demand for MH treatment in Australian public hospitals both in EDs and inpatient care. To better identify the impact of MH on ED presentations, the AIHW should consider expanding the breadth of MH diagnoses they report. When presenting national data, cross-tabulations across patient characteristics and outcomes of presentations would provide insight into the roles of multiple factors affecting ED use for MH conditions. Clinical coding in EDs also needs to be improved to be more specific, including being able to assign multiple codes, which will enable better monitoring of trends. Conclusion: Within the peer-reviewed literature, MH-related presentations as a proportion of all ED presentations have been found to increase in the US, Canada, and Australia. In Australia, analysis of data from the national repository shows increases across all jurisdiction. From a national perspective, the increase ...

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.