536933 research-article2014
ANP0010.1177/0004867414536933Australian & New Zealand Journal of PsychiatryHowlett et al.
Research
Ambulatory mental health data demonstrates the high needs of people with an intellectual disability: Results from the New South Wales intellectual disability and mental health data linkage project
Australian & New Zealand Journal of Psychiatry 2015, Vol. 49(2) 137–144 DOI: 10.1177/0004867414536933 © The Royal Australian and New Zealand College of Psychiatrists 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav anp.sagepub.com
Sophie Howlett1, Tony Florio2, Han Xu1 and Julian Trollor1
Abstract Objective: To investigate the ambulatory mental health service profile of persons with intellectual disability (ID) in a representative sample of New South Wales (NSW). Methods: A statistical linkage key was used to link the Disability Services Minimum Data Set (DS-MDS) and a community mental health services dataset (MH-COM) for a representative area of NSW for the period 2005–2010. Linkage was undertaken in four NSW local health districts (LHDs), covering 36.7% of the NSW population. The mental health profiles and service use characteristics of people with an ID were compared to mental health service users without an ID. Results: Of the 89,262 people in the MH-COM over this 6-year period, 1,459 people (1.6 %) were identified as having an ID. Compared to those without an ID, people with an ID were more likely to have psychotic disorders, developmental disorders and personality disorders, and more than twice as likely to have an ‘unknown’ diagnosis. Compared to those without an ID, people with an ID were less likely to have depressive disorders, adjustment disorders and other disorders. Service use profiles revealed that people with an ID had 1.6 times more face-to-face contacts, and a total faceto-face contact time which was 2.5 times longer than people without an ID. Conclusions: Ambulatory mental health data from NSW indicates that people with an ID receive care for a distinct range of mental disorders, and experience uncertainty regarding their diagnosis. People with an ID have higher service needs which require recognition and the allocation of specific resources. Such data provide a suitable baseline for future evaluation of the impact of ambulatory mental health reforms for people with an ID. Keywords Ambulatory mental health services, data linkage, disability services minimum dataset, intellectual disability, mental health services
Introduction People with an intellectual disability (ID) are a minority group in Australia, accounting for approximately 400,000, or 1.8% of the population (Australian Institute of Health and Welfare (AIHW), 2003) and 2.2% of the NSW population (AIHW, 2008). The majority of these individuals (57%) have a psychiatric disability or mental disorder (Australian Bureau of Statistics, 2009). Compared to the general population, people with an ID experience elevated rates of common mental disorders which are approximately two to three times those experienced by the general population (Cooper et al., 2007; Einfeld et al., 2006, 2011;
Emerson, 2007; Smiley et al., 2007). Predisposition to mental ill-health is apparent across the lifespan of people 1Department
of Developmental Disability Neuropsychiatry, School of Psychiatry, The University of New South Wales, Sydney, Australia 2South Eastern Sydney Local Health District, Sydney, Australia Corresponding author: Julian Trollor, Department of Developmental Disability Neuropsychiatry, The University of New South Wales, 34 Botany Street, Randwick, NSW 2052, Australia. Email: [email protected]
Australian & New Zealand Journal of Psychiatry, 49(2)
Downloaded from anp.sagepub.com at CORNELL UNIV WEILL MED COLG on April 25, 2015
138 with an ID, including in children, younger people, adults (Einfeld et al., 2006) and older adults. Schizophrenia appears to be over-represented by two to four times and has an earlier onset in people with an ID (Borthwick-Duffy, 1994; Morgan et al., 2008). Over-representation has also been noted for common mental disorders such as affective disorders, anxiety disorders and the dementias (Cooper, 1997; Cooper et al., 2007; Strydom et al., 2007, 2009). This vulnerability to mental illness highlights the importance of timely access to psychiatry services for this group (Borthwick-Duffy, 1994; Emerson, 2003). Vulnerability to mental disorders in people with an ID is underpinned by a variety of biological, psychological and social factors. For example, specific genetic conditions associated with ID can increase the risk of psychopathology (Cooper et al., 2007), as can developmental brain abnormalities and pharmacological treatments and their side effects (Tonge and Einfeld, 2000). People with an ID are also at increased risk of a range of physical health conditions (O’Hara et al., 2008; Scheepers et al., 2005; Wallace and Beange, 2008), a factor which increases the risk for mental ill-health. Despite the over-representation of mental disorders in people with an ID, access to mental health services for this population is limited and falls far short of that for the general population. Australian research following a cohort of children and adolescents with ID for 14 years demonstrated that just 10% of those with a mental disorder received an intervention (Einfeld et al., 2006). This compares unfavourably with access to mental health supports for the general population, which has been estimated to be about 34.9% over a 1-year period (Slade et al., 2009). Access to mental health services for this group is influenced by a number of factors. These include: the difficulty a person with an ID may have in seeking mental health care; lack of knowledge, training and confidence of mental health professionals (Edwards et al., 2007; Jess et al., 2008; Phillips et al., 2004; Torr et al., 2008); poor understanding by carers and disability and mental health workers of the manifestations of mental disorders in people with an ID (Costello et al., 2007; Moss et al., 1997; Patel et al., 1993); and a lack of data on the interaction between, and distinct roles and responsibilities of disability and mental health services (Evans et al., 2012). Data linkage is increasingly used as a tool to investigate health and mental health service use and outcomes (Holman et al., 1999; Morgan et al., 2000; Williams et al., 2005). It is a relatively cheap and accurate method of determining the epidemiology of mental disorders, profile of service use, and outcomes of specific sub-groups of interest. Presently, Western Australia (WA) is the only Australian jurisdiction to have linked disability and mental health datasets. Established in 1953, the WA Disability Register (now known as the Intellectual Disability Exploring Answers (IDEA) database), (Petterson et al., 2005) has been used for
ANZJP Articles a number of purposes, including to provide accurate prevalence estimates of ID in children (Leonard et al., 2003) and to investigate the causes of ID (O’Leary et al., 2012) and correlates of ID (Leonard et al., 2005). Linkage between IDEA and the WA Mental Health Information System, the general psychiatric case register, has indicated that approximately 32% of people with an ID had a psychiatric comorbidity and 2% in the WA mental health system had an ID (Morgan et al., 2008). This study indicated that, compared to people without an ID, people with an ID had a higher prevalence and earlier onset of schizophrenia (Morgan et al., 2008). We linked disability services data from the New South Wales Disability Services Minimum Data Set (DS-MDS) with ambulatory mental health services data from a community mental health services dataset (MH-COM) to examine the mental health profile and service use of people with an ID receiving an ambulatory mental health service. We hypothesised that people with an ID would be under-represented in ambulatory mental health services relative to available data on the prevalence of mental disorders in people with an ID. Further, we predicted that people with an ID being managed within ambulatory mental health services would have service use profiles which reflected the complexity of their condition and its management.
Methods Datasets and linkage The DS-MDS is a de-identified dataset which is routinely collected by each Australian State and Territory under the National Disability Agreement. It collates information about the services provided to persons with a disability, and the persons receiving these services. The NSW DS-MDS related to State-funded disability service provision was obtained from the NSW Department of Family & Community Services – Ageing Disability and Home Care (ADHC). The MH-COM refers to ambulatory mental health data which is collected at the local health district (LHD) level within the NSW Mental Health Outcomes and Treatment (MHOAT) data collection system and fed into the local Health Information Exchange (HIE) system. It contains information about community mental health contacts and the patients making those contacts. The MH-COM was provided to the research team from extractions at the local HIE level. The MH-COM represented data from the ethnically and economically diverse catchment areas of South Western Sydney, Sydney, Illawarra Shoalhaven and South Eastern Sydney LHDs. Pre-processing was performed prior to linkage to remove those records in either dataset with missing components in the date of birth (DOB) field and those records with dubious DOB as determined by a calculated age of either 100 using a record’s DOB and a census date of 01/01/2008 (the mid-point of the study period). This resulted in the removal of 59 records from the DS-MDS and 126 records from the MH-COM as there was no way of verifying these dates. The DS-MDS was linked to the MH-COM for the period 2005–2010. The sample area of NSW covered by the study had a total population of 2.64 million, or approximately 36.7% of the State of NSW at the time of the 2011 census. Deterministic linkage was performed using the SLK581 Statistical Linkage Key devised by the AIHW (Karmel, 2005). The SLK581 is constructed using: the 2nd, 3rd and 5th letters of the surname, the 2nd and 3rd letters of the given name, the date of birth as a character string of the form ddmmyyyy, followed by the character ‘1’ for male or ‘2’ for female. Non-alphabetic letters in names are excluded (e.g. hyphens and apostrophes), and where a name contains insufficient letters, the character ‘2’ is used as a placemarker for absent key letters. The character ‘9’ is used for any other missing data so that the linkage key always has a length of 14 characters. The SLK581 has a linkage rate of 99.8% (Karmel, 2005), maintains the integrity and privacy of the data and provides the links in a de-identified format. The SLK581 was generated onsite by local staff for each of the sites in which the MH-COM data were held for the four LHDs, and the data were extracted using an in-house algorithm which collected the required target variables. Access to data was granted by both the data custodians at ADHC and at the Mental Health and Drug and Alcohol Office (MHDAO), NSW Ministry of Health, as well as Directors of Mental Health (and Drug and Alcohol) for the four LHDs. Ethics approval was granted from The
The MH-COM recorded community-based mental health services delivered to 89,262 people. The DS-MDS recorded 34,432 people for whom ID was coded as a primary or secondary disability. Individuals with ID within this dataset meet strict eligibility criteria which includes: intellectual functioning measured at two or more standard deviations below the mean for the full-scale score on a recognised test of intelligence; significant deficits in adaptive functioning in two or more areas, as defined by performance on domains or factors of recognised tests of adaptive functioning which is two or more standard deviations below the mean; and evidence that the deficits in cognitive and adaptive functioning are manifest prior to 18 years of age (ADHC, 2002). Of the 34,432 people with ID in the DS-MDS, 1,159 people were also in the MH-COM. A further 300 people in the MH-COM were identified as having ID as they had an International Classification of Diseases 10 (ICD-10) (WHO, 2010) code in the range F70–79 (Mental Retardation). Therefore, a total of 1,459 individuals with ID were identified within the MH-COM. The comparison group were people without linked records and without an ICD-10 coding in the F70–79 range in the MH-COM (87,803 people). We specified 10 mental disorder diagnostic categories of interest for this study, subsuming related individual ICD-10 diagnoses into overarching categories as described in Table 1. For each individual, each ICD-10 diagnosis recorded over the 6-year study period was assigned a binary flag, which was mapped onto the overarching diagnostic category. An ‘unknown’ diagnosis was also assigned where this was represented in the data. The prevalence of each diagnostic category was then calculated for those with an ID and those without an ID. The number and duration of direct ‘face-toface contacts’ for each MH-COM client were then aggregated for each individual with ID and without an ID over the 6-year period.
Statistical analyses After excluding records with missing data for specific variables of interest, mean age differences for those with and without an ID were determined using a two-tailed Student’s Australian & New Zealand Journal of Psychiatry, 49(2)
Downloaded from anp.sagepub.com at CORNELL UNIV WEILL MED COLG on April 25, 2015
140
ANZJP Articles
Table 2. Demographic profile of ambulatory mental health service users with and without an ID. Demographic
ID
Without an ID
Total
1,459
87,803
Age, years, mean (SE)
28.1 (0.41)
38.3 (0.07)
t(1,544) = −24.79
ANP0010.1177/0004867414536933Australian & New Zealand Journal of PsychiatryHowlett et al.
Research
Ambulatory mental health data demonstrates the high needs of people with an intellectual disability: Results from the New South Wales intellectual disability and mental health data linkage project
Australian & New Zealand Journal of Psychiatry 2015, Vol. 49(2) 137–144 DOI: 10.1177/0004867414536933 © The Royal Australian and New Zealand College of Psychiatrists 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav anp.sagepub.com
Sophie Howlett1, Tony Florio2, Han Xu1 and Julian Trollor1
Abstract Objective: To investigate the ambulatory mental health service profile of persons with intellectual disability (ID) in a representative sample of New South Wales (NSW). Methods: A statistical linkage key was used to link the Disability Services Minimum Data Set (DS-MDS) and a community mental health services dataset (MH-COM) for a representative area of NSW for the period 2005–2010. Linkage was undertaken in four NSW local health districts (LHDs), covering 36.7% of the NSW population. The mental health profiles and service use characteristics of people with an ID were compared to mental health service users without an ID. Results: Of the 89,262 people in the MH-COM over this 6-year period, 1,459 people (1.6 %) were identified as having an ID. Compared to those without an ID, people with an ID were more likely to have psychotic disorders, developmental disorders and personality disorders, and more than twice as likely to have an ‘unknown’ diagnosis. Compared to those without an ID, people with an ID were less likely to have depressive disorders, adjustment disorders and other disorders. Service use profiles revealed that people with an ID had 1.6 times more face-to-face contacts, and a total faceto-face contact time which was 2.5 times longer than people without an ID. Conclusions: Ambulatory mental health data from NSW indicates that people with an ID receive care for a distinct range of mental disorders, and experience uncertainty regarding their diagnosis. People with an ID have higher service needs which require recognition and the allocation of specific resources. Such data provide a suitable baseline for future evaluation of the impact of ambulatory mental health reforms for people with an ID. Keywords Ambulatory mental health services, data linkage, disability services minimum dataset, intellectual disability, mental health services
Introduction People with an intellectual disability (ID) are a minority group in Australia, accounting for approximately 400,000, or 1.8% of the population (Australian Institute of Health and Welfare (AIHW), 2003) and 2.2% of the NSW population (AIHW, 2008). The majority of these individuals (57%) have a psychiatric disability or mental disorder (Australian Bureau of Statistics, 2009). Compared to the general population, people with an ID experience elevated rates of common mental disorders which are approximately two to three times those experienced by the general population (Cooper et al., 2007; Einfeld et al., 2006, 2011;
Emerson, 2007; Smiley et al., 2007). Predisposition to mental ill-health is apparent across the lifespan of people 1Department
of Developmental Disability Neuropsychiatry, School of Psychiatry, The University of New South Wales, Sydney, Australia 2South Eastern Sydney Local Health District, Sydney, Australia Corresponding author: Julian Trollor, Department of Developmental Disability Neuropsychiatry, The University of New South Wales, 34 Botany Street, Randwick, NSW 2052, Australia. Email: [email protected]
Australian & New Zealand Journal of Psychiatry, 49(2)
Downloaded from anp.sagepub.com at CORNELL UNIV WEILL MED COLG on April 25, 2015
138 with an ID, including in children, younger people, adults (Einfeld et al., 2006) and older adults. Schizophrenia appears to be over-represented by two to four times and has an earlier onset in people with an ID (Borthwick-Duffy, 1994; Morgan et al., 2008). Over-representation has also been noted for common mental disorders such as affective disorders, anxiety disorders and the dementias (Cooper, 1997; Cooper et al., 2007; Strydom et al., 2007, 2009). This vulnerability to mental illness highlights the importance of timely access to psychiatry services for this group (Borthwick-Duffy, 1994; Emerson, 2003). Vulnerability to mental disorders in people with an ID is underpinned by a variety of biological, psychological and social factors. For example, specific genetic conditions associated with ID can increase the risk of psychopathology (Cooper et al., 2007), as can developmental brain abnormalities and pharmacological treatments and their side effects (Tonge and Einfeld, 2000). People with an ID are also at increased risk of a range of physical health conditions (O’Hara et al., 2008; Scheepers et al., 2005; Wallace and Beange, 2008), a factor which increases the risk for mental ill-health. Despite the over-representation of mental disorders in people with an ID, access to mental health services for this population is limited and falls far short of that for the general population. Australian research following a cohort of children and adolescents with ID for 14 years demonstrated that just 10% of those with a mental disorder received an intervention (Einfeld et al., 2006). This compares unfavourably with access to mental health supports for the general population, which has been estimated to be about 34.9% over a 1-year period (Slade et al., 2009). Access to mental health services for this group is influenced by a number of factors. These include: the difficulty a person with an ID may have in seeking mental health care; lack of knowledge, training and confidence of mental health professionals (Edwards et al., 2007; Jess et al., 2008; Phillips et al., 2004; Torr et al., 2008); poor understanding by carers and disability and mental health workers of the manifestations of mental disorders in people with an ID (Costello et al., 2007; Moss et al., 1997; Patel et al., 1993); and a lack of data on the interaction between, and distinct roles and responsibilities of disability and mental health services (Evans et al., 2012). Data linkage is increasingly used as a tool to investigate health and mental health service use and outcomes (Holman et al., 1999; Morgan et al., 2000; Williams et al., 2005). It is a relatively cheap and accurate method of determining the epidemiology of mental disorders, profile of service use, and outcomes of specific sub-groups of interest. Presently, Western Australia (WA) is the only Australian jurisdiction to have linked disability and mental health datasets. Established in 1953, the WA Disability Register (now known as the Intellectual Disability Exploring Answers (IDEA) database), (Petterson et al., 2005) has been used for
ANZJP Articles a number of purposes, including to provide accurate prevalence estimates of ID in children (Leonard et al., 2003) and to investigate the causes of ID (O’Leary et al., 2012) and correlates of ID (Leonard et al., 2005). Linkage between IDEA and the WA Mental Health Information System, the general psychiatric case register, has indicated that approximately 32% of people with an ID had a psychiatric comorbidity and 2% in the WA mental health system had an ID (Morgan et al., 2008). This study indicated that, compared to people without an ID, people with an ID had a higher prevalence and earlier onset of schizophrenia (Morgan et al., 2008). We linked disability services data from the New South Wales Disability Services Minimum Data Set (DS-MDS) with ambulatory mental health services data from a community mental health services dataset (MH-COM) to examine the mental health profile and service use of people with an ID receiving an ambulatory mental health service. We hypothesised that people with an ID would be under-represented in ambulatory mental health services relative to available data on the prevalence of mental disorders in people with an ID. Further, we predicted that people with an ID being managed within ambulatory mental health services would have service use profiles which reflected the complexity of their condition and its management.
Methods Datasets and linkage The DS-MDS is a de-identified dataset which is routinely collected by each Australian State and Territory under the National Disability Agreement. It collates information about the services provided to persons with a disability, and the persons receiving these services. The NSW DS-MDS related to State-funded disability service provision was obtained from the NSW Department of Family & Community Services – Ageing Disability and Home Care (ADHC). The MH-COM refers to ambulatory mental health data which is collected at the local health district (LHD) level within the NSW Mental Health Outcomes and Treatment (MHOAT) data collection system and fed into the local Health Information Exchange (HIE) system. It contains information about community mental health contacts and the patients making those contacts. The MH-COM was provided to the research team from extractions at the local HIE level. The MH-COM represented data from the ethnically and economically diverse catchment areas of South Western Sydney, Sydney, Illawarra Shoalhaven and South Eastern Sydney LHDs. Pre-processing was performed prior to linkage to remove those records in either dataset with missing components in the date of birth (DOB) field and those records with dubious DOB as determined by a calculated age of either 100 using a record’s DOB and a census date of 01/01/2008 (the mid-point of the study period). This resulted in the removal of 59 records from the DS-MDS and 126 records from the MH-COM as there was no way of verifying these dates. The DS-MDS was linked to the MH-COM for the period 2005–2010. The sample area of NSW covered by the study had a total population of 2.64 million, or approximately 36.7% of the State of NSW at the time of the 2011 census. Deterministic linkage was performed using the SLK581 Statistical Linkage Key devised by the AIHW (Karmel, 2005). The SLK581 is constructed using: the 2nd, 3rd and 5th letters of the surname, the 2nd and 3rd letters of the given name, the date of birth as a character string of the form ddmmyyyy, followed by the character ‘1’ for male or ‘2’ for female. Non-alphabetic letters in names are excluded (e.g. hyphens and apostrophes), and where a name contains insufficient letters, the character ‘2’ is used as a placemarker for absent key letters. The character ‘9’ is used for any other missing data so that the linkage key always has a length of 14 characters. The SLK581 has a linkage rate of 99.8% (Karmel, 2005), maintains the integrity and privacy of the data and provides the links in a de-identified format. The SLK581 was generated onsite by local staff for each of the sites in which the MH-COM data were held for the four LHDs, and the data were extracted using an in-house algorithm which collected the required target variables. Access to data was granted by both the data custodians at ADHC and at the Mental Health and Drug and Alcohol Office (MHDAO), NSW Ministry of Health, as well as Directors of Mental Health (and Drug and Alcohol) for the four LHDs. Ethics approval was granted from The
The MH-COM recorded community-based mental health services delivered to 89,262 people. The DS-MDS recorded 34,432 people for whom ID was coded as a primary or secondary disability. Individuals with ID within this dataset meet strict eligibility criteria which includes: intellectual functioning measured at two or more standard deviations below the mean for the full-scale score on a recognised test of intelligence; significant deficits in adaptive functioning in two or more areas, as defined by performance on domains or factors of recognised tests of adaptive functioning which is two or more standard deviations below the mean; and evidence that the deficits in cognitive and adaptive functioning are manifest prior to 18 years of age (ADHC, 2002). Of the 34,432 people with ID in the DS-MDS, 1,159 people were also in the MH-COM. A further 300 people in the MH-COM were identified as having ID as they had an International Classification of Diseases 10 (ICD-10) (WHO, 2010) code in the range F70–79 (Mental Retardation). Therefore, a total of 1,459 individuals with ID were identified within the MH-COM. The comparison group were people without linked records and without an ICD-10 coding in the F70–79 range in the MH-COM (87,803 people). We specified 10 mental disorder diagnostic categories of interest for this study, subsuming related individual ICD-10 diagnoses into overarching categories as described in Table 1. For each individual, each ICD-10 diagnosis recorded over the 6-year study period was assigned a binary flag, which was mapped onto the overarching diagnostic category. An ‘unknown’ diagnosis was also assigned where this was represented in the data. The prevalence of each diagnostic category was then calculated for those with an ID and those without an ID. The number and duration of direct ‘face-toface contacts’ for each MH-COM client were then aggregated for each individual with ID and without an ID over the 6-year period.
Statistical analyses After excluding records with missing data for specific variables of interest, mean age differences for those with and without an ID were determined using a two-tailed Student’s Australian & New Zealand Journal of Psychiatry, 49(2)
Downloaded from anp.sagepub.com at CORNELL UNIV WEILL MED COLG on April 25, 2015
140
ANZJP Articles
Table 2. Demographic profile of ambulatory mental health service users with and without an ID. Demographic
ID
Without an ID
Total
1,459
87,803
Age, years, mean (SE)
28.1 (0.41)
38.3 (0.07)
t(1,544) = −24.79
