Trends in child and adolescent discharges at a New Zealand psychiatric inpatient unit between 1998 and 2007

Research on child and adolescent inpatient mental health populations can provide useful information on changes in service utilisation, psychiatric diagnoses, and demographic variables, and can have implications for service planning with regard to young people's psychiatric inpatient care.There has been little research in New Zealand (NZ) on trends, demographic and illness variables in child and adolescent inpatient populations. A Christchurch study gathered admission data over an 18-month period in an adolescent inpatient unit in the South Island.1 In a sample of 72 subjects the most common diagnosis was mood disorder (54%) followed by anxiety or adjustment disorder (25%) and major psychosis (21%).1 Unfortunately, no demographic information (gender, ethnicity) was reported in this study.In terms of ethnic differences, the available evidence from the few NZ studies of inpatient populations suggests an over-representation of admissions for young Māori. A 25-year longitudinal study of adolescents looked at, among other things, ethnic identification and mental health problems.2 The authors described increased rates of psychiatric disorder amongst Māori youth (age 18-25 years)—depression, anxiety and substance dependence were all over-represented in the 18-21 year age group, in comparison to non-Māori.2A study on a cohort of Māori and non-Māori patients (aged between 15 and 45) admitted to inpatient services in Otago between 1990 and 1992 found Māori were over represented among first admissions.3 Māori were also found to be a more disadvantaged group with respect to financial support, education and other health problems.3 These findings are consistent with adult studies which have found that patterns of diagnosis, trends over time and use of psychiatric services in adult patients also vary between different ethnic groups in NZ, and data suggest that rates of admission to hospital are higher for Māori than non-Māori.4A retrospective file review of more than 900 adult patients from three Auckland acute inpatient psychiatric units reported that, based on the community population, Māori admissions were double the expected rate and Asian admissions were lower than expected.5 Compared to European admissions, Māori, Pacific, and Asian admissions were all more likely to have a diagnosis of a psychotic disorder.5Given the lack of NZ studies on child and adolescent psychiatric inpatient populations, the current retrospective analysis tried to address this gap by depicting trends in sociodemographic and diagnostic characteristics in a NZ child and adolescent inpatient unit over a period of 10 years between 1998 and 2007. It was also hoped that this information could then be used to inform future service planning.Method The study setting was an acute child and adolescent psychiatric inpatient unit with 23 beds, based in a public health service in the North Island. The unit includes a small eight bedded locked section, with minimal outdoor access and no capacity for family/whānau to stay. The rest of the unit is ‘open', and more spacious, with access to a garden and activity areas. Admissions cover the age range 0-18 years, but the vast majority are for adolescents aged 13-18 years. The geographic area served is large and encompasses both rural and urban regions. The population served is approximately 2 million, with an ethnically diverse mix including most of the Pacific Island population in NZ and a growing number of people of South East Asian descent. The model of practice used is bio-psycho-social with a strong focus on family/whānau participation and continuity of care with the referring service. Ethical approval was obtained from the Northern Y Regional Ethics Committee and the relevant health service. All consecutive discharges between 1st January 1998 to 31st December 2007 were identified on an electronic database.Repeat discharges were examined and any temporary discharges (patients readmitted within 14 days) were re-coded as one continuous inpatient admission and one discharge. The data obtained from the electronic database included the number of discharges, demographic characteristics (age, ethnicity, gender) and clinical data (primary diagnosis at discharge, length of stay). For the purpose of the descriptive analyses, primary diagnoses at discharge were grouped into the following diagnostic groups: Psychotic Disorders (incl. psychosis, schizophrenia), Affective Disorders (incl. depression, dysthymia), Anxiety Disorders (incl. phobia, post traumatic stress disorder, obsessive compulsive disorder, adjustment disorder), Bipolar Affective Disorder, Developmental Disorders (incl. pervasive development disorder, attention deficit hyperactivity disorder, autistic spectrum disorder), Eating Disorders (incl. anorexia nervosa, eating disorders not otherwise specified), Externalising Behaviour Disorders (incl. disruptive behaviour disorder, oppositional defiant disorder, conduct disorder), Substance Abuse Disorders, Suicide/Self-harm, Personality Traits, Other and None. Summary statistics (proportions, means/medians, and 95% confidence intervals) were calculated for the distributions of demographic and clinical data, both overall and on a yearly basis. Trends across time for changes in annual discharge numbers were investigated by regression analyses. Trends across time for changes in annual patient proportions (and therefore controlled for any entire sample change in numbers) were investigated by the chi-square based linear trend test in the Stats Direct Version 2.7.8.software (Stats Direct Ltd, UK). Results There were 1109 discharges in the 10-year review period. This constituted 899 individual people, with 150 of those individuals having more than one discharge from the unit. In terms of demographic variables, 50.6% (n=561) of the discharges involved a female patient and the mean age was 15.6 years (range 2.6-19.7 years). Over half of all discharges (53.4%, n=588) involved young people who identified as European, with the remaining sample identifying as NZ Māori (29.1%, n=321), Pacific Islander (7.4%, n=82), Asian (7.1%, n=78), or Other (3.0%, n=33). Ethnicity was not recorded for seven discharges. Māori young people were over-represented in the discharges and Pacific Island young people under-represented compared to the ethnic proportions of the relevant age group of the catchment area of the unit (Table 1, p<0.0001). Table 1. Ethnic proportions of relevant age group in catchment area (Census data) compared to unit population Ethnicity Census 1996-2006 % Unit population % (95% CI) European NZ Māori Pacific Islander Asian Other 54.8 21.5 12.0 8.7 3 53.4 (50.4 to 56.3) 29.1 (26.5 to 31.9) 7.4 (6.0 to 9.1) 7.1 (5.7 to 8.7) 3.0 (2.1 to 4.2) The annual changes in the numbers and proportions of discharges from 1998 to 2007 are shown in Tables 2 and 3. Table 2 shows the number of discharges over 10 years increased, with 68% more discharges in 2007 compared to 1998, and a linear trend of eight additional discharges per year (p=0.007). No significant trend of change in gender proportions was observed reflecting similar rates of male and female patients over the 10-year period. Changes over time were observed for ethnicity, with a significant decrease in the proportion of European patients (p=0.002) and a significant increase in the numbers and proportions of NZ Māori patients (p=0.0004 and p=0.0001). However, there was little change over time in discharges from young people of Pacific Island, Asian or Other ethnicity. In terms of clinical variables, Table 3 shows that the majority of discharges received a primary diagnosis of either Psychosis (34%, n=382), Anxiety (16%, n=181) or Affective disorders (15%, n=167). More than 50% of the total number of both Māori (321) and Pacific Island (82) patients was discharged with a diagnosis of psychosis compared to significantly smaller proportions in the other ethnic groups (p=0.0001) (data not shown). Psychosis was the most common diagnosis for Māori and Pacific Island patients in this child and adolescent psychiatric inpatient setting. Table 2. Number of discharges and demographic characteristics of the unit population between 1998 and 2007 Variables 1998-2007 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 P value n % n (%) Discharges 1109 79 82 78 75 129 127 157 115 134 133 0.007 Gender Female Male 561 548 50.6 49.4 40 (51) 39 40 (49) 42 35 (45) 43 39 (52) 36 63 (49) 66 61 (48) 66 75 (48) 82 71 (62) 44 71 (53) 63 66 (50) 67 0.001 0.33 0.05 Ethnicity European 588 53.4 43 (56) 49 (60) 40 (51) 43 (57) 82 (64) 73 (57) 86 (55) 56 (49) 61 (46) 55 (43) 0.21 0.002 NZ Māori 321 29.1 14 (18) 23 (28) 15 (19) 18 (24) 29 (23) 39 (31) 49 (31) 40 (35) 50 (37) 44 (34) 0.0004 0.0001 Pacific Islander 82 7.4 5 (6) 1 (1) 12 (15) 11 (15) 11 (9) 7 (6) 8 (5) 7 (6) 8 (6) 12 (9) 0.27 0.68 Asian 78 7.1 8 (10) 8 (10) 9 (12) 2 (3) 6 (5) 7 (6) 7 (4) 7 (6) 11 (8) 13 (10) 0.19 0.71 Other 33 3.0 7 (9) 1 (1) 2 (3) 1 (1) 0 (0) 1 (1) 7 (4) 5 (4) 4 (3) 5 (4) 0.47