Abstract
Purpose
People with severe mental illness (SMI) experience high levels of unemployment. We aimed to better understand the associations between clinical, social, and demographic inequality indicators and unemployment.
Methods
Data were extracted from de-identified health records of people with SMI in contact with secondary mental health services in south London, UK. A Natural Language Processing text-mining application was applied to extract information on unemployment in the health records. Multivariable logistic regression was used to assess associations with unemployment, in people with SMI.
Results
Records from 19,768 service users were used for analysis, 84.9% (n = 16,778) had experienced unemployment. In fully adjusted models, Black Caribbean and Black African service users were more likely to experience unemployment compared with White British service users (Black Caribbean: aOR 1.62, 95% CI 1.45–1.80; Black African: 1.32, 1.15–1.51). Although men were more likely to have experienced unemployment relative to women in unadjusted models (OR 1.36, 95% CI 1.26–1.47), differences were no longer apparent in the fully adjusted models (aOR 1.05, 95% CI 0.97–1.15). The presence of a non-affective (compared to affective) diagnosis (1.24, 1.13–1.35), comorbid substance use (2.02, 1.76–2.33), previous inpatient admissions (4.18, 3.71–4.70), longer inpatient stays (78 + days: 7.78, 6.34–9.54), and compulsory admissions (3.45, 3.04–3.92) were associated with unemployment, in fully adjusted models.
Conclusion
People with SMI experience high levels of unemployment, and we found that unemployment was associated with several clinical and social factors. Interventions to address low employment may need to also address these broader inequalities.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Background
People who are diagnosed with schizophrenia and related disorders or bipolar disorders often experience significant disability and impairment. Unemployment rates remain high for those with these severe mental illnesses (SMI), estimated to reach between 80 and 90% for people with schizophrenia [1,2,3,4,5]. This has major implications for people with SMI on an individual level, as barriers to participating in the job market contribute to the social exclusion of people with SMI and can adversely impact quality of life and clinical outcomes [6]. In qualitative research, service users have highlighted the important role that employment can play in their lives and recovery [7, 8]. Unemployment is a sizeable contributor to the societal and economic costs of these mental illnesses, with lost productivity due to unemployment representing an estimated 29% of the total costs [5, 9].
Positive work histories and cognitive functioning are some of the significant predictors of employment outcomes for people with SMI [1, 10]. However, there is a relative dearth of information on associations between sociodemographic and clinical characteristics with unemployment outcomes for people with SMI. There is a need to assess these associations in SMI populations and to understand how these link to broader inequities impacting these groups, which could be used to inform intervention development and implementation.
The digitization of health records has allowed for large scale, naturalistic studies of the health of populations. However, studies using routine data to understand and assess inequalities in mental disorders have to date been hampered by a lack of individual-level measures of socioeconomic position, including employment and other indicators, as these are poorly or incompletely captured in standard structured fields in electronic health records (EHRs). In mental health EHRs, detail on the service user’s occupation may only be recorded in unstructured free-text fields in clinical note form [11], which limits its use to inform analyses, particularly in large-scale health records. However, text-mining methods can enable the extraction of this information. The present study utilizes this methodology to assess occupational status at scale in the health records of service users with SMI.
The aim of this study was to investigate sociodemographic, clinical, and service use variables associated with unemployment for service users with SMI. In addition, through using large-scale electronic health records from a diverse catchment area [12], we further sought to assess ethnic inequalities impacting this indicator. Based on the previous literature, we hypothesized that service users with SMI who had any recorded experience of unemployment would be more likely to be older [13,14,15,16,17,18,19,20,21,22,23], male [1], from a minority ethnic group [24], have a diagnosis of non-affective SMI [1], and have more intensive contact with mental health services as indicated by inpatient admissions [25, 26].
Methods
Participants and setting
The South London and Maudsley (SLaM) National Health Service (NHS) Trust serves a large, ethnically diverse catchment area across four inner-city London boroughs (Southwark, Croydon, Lambeth, and Lewisham), with a population of approximately 1.3 million people [27]. Compared with London as a whole, UK Census data indicate that there are higher proportions of people from Black African and Black Caribbean ethnic groups, and lower proportions from South Asian groups in the SLaM catchment areas [28]. Since 2008, the EHRs of service users have been de-identified and are accessible to approved researchers through the Clinical Record Interactive Search (CRIS) platform [27], which was accessed for this study.
SMI was defined as an F2* or F30/F31 disorder (schizophrenia and related disorders, and bipolar disorder) according to the International Classification of Diseases (ICD-10), where this diagnosis was not preceded by an F0* organic disorder (for example, dementia). Service user record IDs were extracted from CRIS where there had been a recording of a primary or secondary diagnosis of SMI, either in the structured field or the clinical notes in the record. The latter diagnoses were identified in the notes using an NLP application, developed using the Generalized Architecture for Text Engineering platform [29] to identify mental disorders according to ICD-10 diagnosis [30]. Service users were included if they were recognized as an adult in the UK (over the age of 16) at the time of occupation data extraction (29th January 2020), as this is the legal full-time working age in the UK. Sample sizes were small for inferential analyses in some of the ethnic groups (Fig. 1). Aggregating service users from these ethnic groups would have led to a heterogeneous group, which would be challenging to draw meaningful interpretations from. Therefore, similar to previous research, we excluded service users who did not belong to White British, Irish, Black African, Black Caribbean, Indian, Pakistani, or Bangladeshi ethnic groups [30].
Measures
The service user’s date of birth was used to generate age at the time of data extraction. This was grouped into 10-year bands, except for the youngest group (age 16–29) to maintain sample power for this group with a start age of 16. Relationship status was grouped to include service users who were in a relationship (married/civil partnership/cohabiting), or not in a relationship (single/divorced/partnership dissolved/separated/widowed). Ethnicity was categorized to correspond with the Office for National Statistics categories for ethnicity [30]. We included service users who had ethnicity recorded as either White British, Irish, Black Caribbean, Black African, and South Asian (including Indian, Pakistani, Bangladeshi—these groups were aggregated due to smaller sample sizes). Area-level deprivation was assessed through the Index of Multiple Deprivation (IMD) [31], which is an aggregate of domains of area-level deprivation including income, employment, education, health, crime, barriers to housing and services, and living environment deprivation. IMD scores for the service user’s address closest to diagnosis date were mapped onto national quintiles of deprivation according to the Office for National Statistics 2015 guidelines related to Lower Super Output Area (LSOA), which have a mean of 1500 people residing in each area [32].
Several clinical and service use variables were extracted from the health record. Service users’ first SMI diagnosis (either from the structured field or clinical notes) was extracted and grouped into affective disorders (ICD-10 codes F30 and F31) and non-affective disorders (all F2* codes). The date of birth was used to calculate age at diagnosis date, which was grouped as working age SMI onset (< 64) or retirement age SMI onset (≥ 65). Service users who also had an ICD-10 diagnosis of F10–F19 were identified as having a history of alcohol or substance use disorder. We also extracted psychiatric inpatient admissions data which included: total number of inpatient admissions, total number of inpatient bed days, and total number of inpatient bed days under compulsory detention (the Mental Health Act). The median for cumulative bed days was calculated and used as a cut-off for the bed days groups.
Employment status
To identify occupation mentions in the clinical notes in the EHR, we applied a validated text-mining application to extract occupation mentions from the clinical notes via CRIS [33]. The application uses a combination of machine-learning and rule-based approaches—the development of the application is described in further detail elsewhere [33]. The application has good levels of precision (0.88) and recall (0.90) on documents which describe the service user’s ‘personal history’ [33]. If the structured, drop-down field for employment was completed, this was also extracted. At least one mention of unemployment in either the clinical notes or structured field was used as an indicator of unemployment. Where a service user’s record had at least one mention of unemployment, they were included in the ‘unemployed group’. Where service users had multiple occupation types recorded in their records alongside unemployment (for example, ‘accountant’ and ‘unemployed’), they were also included in this ‘unemployed’ group. Conversely, if the record did not mention unemployment, the service user was included in the ‘never unemployed’ group. If occupation was not recorded in the health record (i.e., data were missing), these service user records were not included in the sample (Fig. 1). We validated this by manually checking the clinical notes for a random sub-sample of the cohort and comparing this to the extracted text-mined and structured field occupations. In this sample of 100 service users, this methodology gave a positive predictive value of 79% for the construct of interest (experience of unemployment) in manually screened full case notes.
Statistical analyses
Stata-15 software [33] was used for all analyses. The characteristics of the sample were first cross tabulated with employment group. Univariable and multivariable logistic regression were used to assess associations between social, clinical, and service use characteristics and unemployment. A priori confounders included age and sex. We next included age, sex, area-level deprivation, relationship status, ethnicity, diagnosis type (affective/non-affective), age at diagnosis, and history of substance use disorder, as additional confounders in fully adjusted models. Likelihood ratio tests were conducted for all models. As an additional set of sensitivity analyses, we removed people above working age (65 +) from the sample and repeated the analyses.
Results
Figure 1 displays the number of service user records identified with SMI (n = 37,092), and subsequent exclusion based on completeness of data and inclusion criteria. In total, 19,768 service users were included in the final analytical sample for complete case analysis (53.3% of original sample), of whom 16,778 (84.9%) had unemployment recorded.
Table 1 shows the descriptive data for service users in the complete sample. Most service users (85.3%) were recorded as not being in a relationship. While 51% of service users were White British, 28% of service users were Black Caribbean and 14% were Black African. A high proportion of service users were living in the most deprived areas at time of diagnosis (highest IMD quintile n = 8,978, 45%).
Table 2 shows the proportion of service users in the unemployed group for each sociodemographic group, and results from both the crude and adjusted multivariable logistic regression analyses. After adjusting for all covariates (age, sex, relationship status, ethnicity, diagnosis type, age at diagnosis, and substance use disorder), area-level deprivation was strongly associated with unemployment: service users who lived in the most deprived national quintile had twofold relative odds of experiencing unemployment compared with service users living in the least deprived areas. Service users had an increased odds of experiencing unemployment if they were middle aged at the time of data extraction, and not in a relationship. Although men were more likely to have experienced unemployment relative to women in unadjusted models (OR 1.36, 95% CI 1.26–1.47), differences were no longer apparent in the fully adjusted models (aOR 1.05, 95% CI 0.97–1.15).
Across unadjusted and fully adjusted models for ethnicity, Black Caribbean service users had the highest odds of experiencing unemployment compared with White British service users (aOR 1.62, 95% CI 1.45–1.80). Black African service users were also more likely to have experienced unemployment compared with White British service users (aOR 1.30, 95% CI 1.14–1.49). There were no differences in the odds of unemployment between White British and Irish service users (aOR 1.18, 95% CI 0.93–1.50) or South Asian service users (aOR 1.17, 95% CI 0.95–1.43).
There were also associations between clinical and service use characteristics and unemployment (Table 3). In unadjusted models, service users had an increased odds of 1.43 (95% CI 1.32–1.55) of unemployment if they had a non-affective SMI diagnosis compared with an affective disorder. This association persisted after adjustment for all other covariates. The inpatient admission variables had the strongest associations with unemployment out of all variables: the odds of experiencing unemployment were elevated fourfold in service users that had experienced inpatient admissions in fully adjusted models. Service users who had inpatient admissions of a longer duration and compulsory admissions also had high odds of unemployment.
In the sensitivity analysis restricting the sample to working age adults only, no substantial differences were noted (Supplementary File 1).
Discussion
In this large-scale sample of 19,768 service users with SMI, using novel text-mining methods, we found that an extremely high proportion (85%) of service users had experienced unemployment. Our findings are consistent with previous studies [2, 34]. Unemployment rates in the general population for the catchment area were estimated to be 4.3% in London by the Office for National Statistics at the time of the study [35], providing further context. Although these unemployment statistics for the local area capture unemployment at one particular timepoint, and in our study we looked at any recorded experiences of unemployment in the health record, our findings highlight the potentially deep inequalities which impact this group.
We also found key clinical differences between service users who had been unemployed compared with those who had no recorded unemployment. As hypothesized, service users who had experienced inpatient admissions, longer inpatient stays and compulsory admissions were more likely to have experienced unemployment. This can be considered in the context of previous work which has found that service users with SMI who have more severe symptoms and lower functioning scores are more likely to experience unemployment [36]. Our study provides a new perspective on this by investigating associations between inpatient stays and unemployment. An inpatient stay may suggest that the service user had more intensive contact with services and may have been more clinically unwell, which could have made it difficult to then initiate and maintain employment [36]. We also found that patients with comorbid substance use disorders had over twice the odds of experiencing unemployment. Substance use disorders were ascertained using ICD-10 codes; however, rates were lower than expected [37]. The presence of substance use disorders could therefore be an underestimate, due to under-recording in the structured field of the health record for some service users. In addition, service users with an earlier age of SMI onset and a non-affective SMI diagnosis were also more likely to experience unemployment in the sample—these could also be indicators of illness severity and functioning.
We found evidence to support our hypotheses for associations between sociodemographic characteristics and unemployment including age and relationship status, but not for sex. We found that service users who were aged between 50 and 59 were more likely to have experienced unemployment. These findings corroborate previous research: investigators have previously found that being of an older or middle age is associated with an increased likelihood of unemployment in people with SMI [13,14,15,16,17,18,19,20,21,22,23]. This may be a consequence of having more chances of being unemployed compared with younger service users. We found that service users who were not in a relationship were more likely to experience unemployment—this observation has been supported by work in other countries [38,39,40,41,42]. Although men had a higher likelihood of unemployment in the age adjusted model, this association was no longer evident in fully adjusted models, taking into account area deprivation, relationship status, ethnicity, and other clinical factors. Previous research on the relationship between sex and unemployment in SMI samples have also found no evidence of an association [17, 19,20,21, 25, 40, 42,43,44,45,46,47,48].
A key strength of the present study is that, using methodologies developed for large-scale textual analysis, we were able to assess a large sample of almost 20,000 service users. This electronic health record dataset provides ‘real-world’ data and insights [28] into occupation and unemployment for service users accessing mental health services in south London. As the study catchment area includes a high proportion of service users from Black African and Black Caribbean ethnic groups, we were able to undertake comparisons between minority ethnic subgroups, which has in general been limited and not usually possible to this extent. We hypothesized that there would be higher odds of unemployment for ethnic minority groups—this hypothesis was supported for some ethnic minority groups, but not others. Relative to White British service users, we found that Black Caribbean and Black African service users were more likely to have experienced unemployment, after adjusting for other variables, including area deprivation. Irish service users were also more likely to have experienced unemployment in age and sex adjusted models, although this was less apparent in fully adjusted models. We observed no differences between South Asian service users and White British service users in this sample. However, as Indian, Pakistani, and Bangladeshi ethnic groups were grouped as ‘South Asian’ due to smaller sample sizes, differences between these groups may have been masked, and it is a limitation that we could not disaggregate this group further. Overall, these findings suggest ethnic inequalities in employment outcomes for people with SMI are particularly apparent for Black Caribbean and Black African service users, but not Irish or South Asian service users.
Ethnic inequalities in unemployment are also evident in the general population in the UK: individuals from Black ethnic groups are currently twice as likely to be unemployed compared to White British individuals [49], and are more likely to be in precarious, temporary employment, which may exacerbate the risk of unemployment when unwell [50]. In addition, Black Caribbean, and Black African people in the general population have been shown to be more likely to receive lower earnings compared to White British people; these trends have also been noted in people of Pakistani and Bangladeshi origin [50]. A range of factors, including prior experiences of unemployment, younger age, lower education levels, and the impact of cumulative discrimination and racism over the life-course, have been implicated [50,51,52]. Our findings reflect these wider structural inequities, but with the additional focus on SMI impacting racially minoritized groups. Our findings are also consistent with other UK cohort studies: the investigators of the AESOP longitudinal study found that Black Caribbean people with first-episode psychosis were more likely to be unemployed compared with the White British reference group [24, 34].
A limitation of this study is that, by looking at mentions of ‘unemployment’, this may miss some service users who were unemployed where this was not mentioned in the health record. As unemployment was our primary outcome, we excluded patients who were missing occupation data from our sample. This recording and selection bias may impact some groups more than others. For example, older service users of retirement age may only be described as ‘retired’, despite experiencing unemployment prior to retirement age; therefore, unemployment may be disproportionately missed in this group. In a previous study, we found that people with more contacts with services, who may have a more severe course of illness, were more likely to have employment status recorded [33]. This may have partly influenced associations between unemployment and service use variables in the present study. However, our comparison group included service users who had no mentions of unemployment but did have other occupations, which could similarly be better recorded where service users had more contacts. Some other patient groups may be less likely to have occupation recorded and may therefore not be represented in the complete sample, for example those with communication difficulties, although this is likely to be a small number in the sample. Compared to a previous study with all secondary mental health service users, this group of patients with SMI had comparatively lower levels of missing employment data (20.6% in the present study, compared to 43.3% with all service users) [33].
We were unable to evaluate recall rates for the text-mining algorithms, as it was unfeasible to read through the service user’s whole care record when occupation is rarely mentioned in the context of the wider healthcare notes. However, our estimates of occupation recording after deploying the NLP application, which indicated that 79.4% of records had an occupation recorded (Fig. 1), approximated closely to the proportions previously reported in a national audit of SMI patients’ case notes [53]. Occupations should be well recorded for service users with SMI diagnoses, as employment history forms part of a holistic psychiatric assessment, with occupation support for people living with schizophrenia or bipolar disorders evidenced as a quality standard for delivering care [54]. As clinical notes tend to repeat information multiple times across the record [55], this increased opportunities for unemployment to be identified by the application. Furthermore, despite a risk that unemployment may have been under-recorded, our estimates of unemployment were still highly consistent with findings from other studies [1, 2, 56].
The present study was cross-sectional in design by extracting data held in the EHR in January 2020. It would have been desirable to look at patterns of unemployment longitudinally in service users with SMI—this was a key strength in the AESOP longitudinal study of outcomes for people with first-episode psychosis [56]. Limitations of the data and current methodology meant that we were unable to look at unemployment temporally in this study: we were unable to identify when a service user became unemployed, how long they were unemployed, or whether they were recently or currently unemployed. This further detail on unemployment would be of clinical and research interest. There were also other characteristics which could not be examined here—including service user’s educational attainment, and duration of untreated psychosis, which could have explained some of the variation in the analysis [13, 47, 57]. We were, however, able to measure other indicators of illness severity and functioning (inpatient admissions, length of inpatient stay, and compulsory admissions) and found that these indicators for more severe illness had strong and substantial associations with unemployment.
This study was conducted at a single secondary care mental health provider in an inner-city area in the UK, albeit with near-complete coverage of the catchment area of approximately 1.3 million people. The catchment area reflects an urban, ethnically diverse but highly deprived area in the UK [58]. The study catchment may be similar to other metropolitan/urban areas in the UK; however, the findings may be less generalizable to more rural catchment areas.
This study demonstrates that text-mining methodology can be useful to access and analyze the social determinants of mental health conditions in EHR data [59]. These approaches may be further developed in future to enhance understandings of inequalities in routinely collected health records data [60].
Conclusions
The findings of this study make important contributions to the current literature by providing a large-scale assessment of the prevalence of unemployment experienced by secondary mental health service users with SMI. Our findings highlight the importance of a range of clinical and social indicators as impacting on the likelihood of service users experiencing unemployment. Our findings also suggest a potential ‘inequality within an inequality’ for service users from Black minority ethnic groups living with severe mental health conditions. Addressing high unemployment rates through support and interventions (for example, Individual Placement and Support [61]), which potentially take into account wider structural inequities, remains of paramount importance and could play an important role in future.
Data availability
Data are owned by a 3rd party SLaM BRC CRIS tool which provides access to anonymised data derived from SLaM electronic medical records. These data can only be accessed by permitted individuals from within a secure firewall (i.e. the data cannot be sent elsewhere) in the same manner as the authors.
References
Marwaha S, Johnson S (2004) Schizophrenia and employment. Social Psychiatry Psychiatr Epidemiol 39(5):337–349
Marwaha S et al (2007) Rates and correlates of employment in people with schizophrenia in the UK, France and Germany. Br J Psychiatry 191(1):30–37
Morselli P, Elgie R, Cesana B (2004) GAMIAN-Europe/BEAM survey II: cross-national analysis of unemployment, family history, treatment satisfaction and impact of the bipolar disorder on life style. Bipolar Disord 6(6):487–497
Burke-Miller JK et al (2006) Demographic characteristics and employment among people with severe mental illness in a multisite study. Commun Mental Health J 42(2):143–159
Evensen S et al (2016) Prevalence, employment rate, and cost of schizophrenia in a high-income welfare society: a population-based study using comprehensive health and welfare registers. Schizophr Bull 42(2):476–483
Bouwmans C et al (2015) Employment and the associated impact on quality of life in people diagnosed with schizophrenia. Neuropsychiatr Dis Treat 11:2125
Dunn EC, Wewiorski NJ, Rogers ES (2008) The meaning and importance of employment to people in recovery from serious mental illness: results of a qualitative study. Psychiatr Rehabil J 32(1):59
Walsh FP, Tickle AC (2013) Working towards recovery: the role of employment in recovery from serious mental health problems: a qualitative meta-synthesis. Int J Psychosoc Rehabil 17(2):35–49
Jin H, Mosweu I (2017) The societal cost of schizophrenia: a systematic review. Pharmacoeconomics 35(1):25–42
Tsang HW et al (2010) Review on vocational predictors: a systematic review of predictors of vocational outcomes among individuals with schizophrenia: an update since 1998. Aust N Z J Psychiatry 44(6):495–504
Aldekhyyel R et al (2016) Content and quality of free-text occupation documentation in the electronic health record. In: AMIA annual symposium proceedings. American Medical Informatics Association, Bethesda, MD
Das-Munshi J et al (2016) Ethnicity and cardiovascular health inequalities in people with severe mental illnesses: protocol for the E-CHASM study. Soc Psychiatr Psychiatr Epidemiol 51(4):627–638
Tse S et al (2014) Meta-analysis of predictors of favorable employment outcomes among individuals with bipolar disorder. Bipolar Disord 16(3):217–229
Butler G et al (2010) Characteristics of people with severe mental illness who obtain employment. Psychiatrist 34(2):47–50
Depp CA et al (2012) Determinants of occupational and residential functioning in bipolar disorder. J Affect Disord 136(3):812–818
Kaneda Y, Jayathilak K, Meltzer HY (2009) Determinants of work outcome in schizophrenia and schizoaffective disorder: role of cognitive function. Psychiatry Res 169(2):178–179
Kaneda Y, Jayathilak K, Meltzer H (2010) Determinants of work outcome in neuroleptic-resistant schizophrenia and schizoaffective disorder: cognitive impairment and clozapine treatment. Psychiatry Res 178(1):57–62
Kortrijk HE et al (2019) Employment rates in flexible assertive community treatment teams in the Netherlands: an observational study. Commun Mental Health J 55(2):350–359
Drakopoulos J et al (2020) Executive functioning but not IQ or illness severity predicts occupational status in bipolar disorder. Int J Bipolar Disord 8(1):7
Salkever DS et al (2007) Measures and predictors of community-based employment and earnings of persons with schizophrenia in a multisite study. Psychiatr Serv 58(3):315–324
Carstens C, Massatti R (2018) Predictors of labor force status in a random sample of consumers with serious mental illness. J Behav Health Serv Res 45(4):678–689
Gilbert A et al (2010) Self-reported cognitive problems predict employment trajectory in patients with bipolar I disorder. J Affect Disord 124(3):324–328
Waghorn G, Chant D, Whiteford H (2003) The strength of self-reported course of illness in predicting vocational recovery for persons with schizophrenia. J Vocation Rehabil 18(1):33–41
Morgan C et al (2005) Pathways to care and ethnicity. 1: Sample characteristics and compulsory admission: report from the AeSOP study. Br J Psychiatry 186(4):281–289
Nordt C et al (2007) Predictors and course of vocational status, income, and quality of life in people with severe mental illness: a naturalistic study. Soc Sci Med 65(7):1420–1429
Babalola O et al (2014) Length of hospitalisation for people with severe mental illness. Cochrane Database Syst Rev 2014(1):CD000384
Stewart R et al (2009) The South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLAM BRC) case register: development and descriptive data. BMC Psychiatry 9(1):1–12
Perera G et al (2016) Cohort profile of the South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLAM BRC) case register: current status and recent enhancement of an electronic mental health record-derived data resource. BMJ Open 6(3):e008721
Cunningham H et al (2013) Getting more out of biomedical documents with GATE’s full lifecycle open source text analytics. PLoS Comput Biol 9(2):e1002854
Das-Munshi J et al (2017) Ethnicity and excess mortality in severe mental illness: a cohort study. Lancet Psychiatry 4(5):389–399
Noble M et al (2006) Measuring multiple deprivation at the small-area level. Environ Plann A 38(1):169–185
Smith T et al (2015) The English indices of deprivation 2015. Department for Communities and Local Government, London
Chilman N et al (2021) Text mining occupations from the mental health electronic health record: a natural language processing approach using records from the Clinical Record Interactive Search (CRIS) platform in south London, UK. BMJ Open 11(3):e042274
Morgan C et al (2017) Ethnicity and long-term course and outcome of psychotic disorders in a UK sample: the ÆSOP-10 study. Br J Psychiatry 211(2):88–94
Office for National Statistics (2019) Regional labour market statistics: HI00 Headline indicators for UK regions and countries. Office for National Statistics, London
Ramsay CE, Stewart T, Compton MT (2012) Unemployment among patients with newly diagnosed first-episode psychosis: prevalence and clinical correlates in a US sample. Soc Psychiatry Psychiatr Epidemiol 47(5):797–803
Hunt GE et al (2018) Prevalence of comorbid substance use in schizophrenia spectrum disorders in community and clinical settings, 1990–2017: systematic review and meta-analysis. Drug Alcohol Depend 191:234–258
Elinson L, Houck P, Pincus HA (2007) Working, receiving disability benefits, and access to mental health care in individuals with bipolar disorder. Bipolar Disord 9(1–2):158–165
Bauer M et al (2009) Frequency of subsyndromal symptoms and employment status in patients with bipolar disorder. Soc Psychiatry Psychiatr Epidemiol 44(7):515–522
Honkonen T et al (2007) Employment predictors for discharged schizophrenia patients. Soc Psychiatry Psychiatr Epidemiol 42(5):372–380
Miettunen J et al (2007) Socio-demographic and clinical predictors of occupational status in schizophrenic psychoses—follow-up within the Northern Finland 1966 Birth Cohort. Psychiatry Res 150(3):217–225
Chua YC et al (2019) First-episode psychosis and vocational outcomes: a predictive model. Schizophnia Res 211:63–68
Chang WC et al (2014) Clinical and cognitive predictors of vocational outcome in first-episode schizophrenia: a prospective 3 year follow-up study. Psychiatry Res 220(3):834–839
Tandberg M et al (2011) Neurocognition and occupational functioning in patients with first-episode psychosis: a 2-year follow-up study. Psychiatry Res 188(3):334–342
Martinez Camarillo S et al (2018) Occupational functioning and employment outcomes in bipolar disorder: a Mexican study with gender perspective. In: Bipolar disorders. Wiley, Hoboken, NJ
Tsang H et al (2000) Predictors of post-hospital employment status for psychiatric patients in Hong Kong: from perceptions of rehabilitation professionals to empirical evidence. Int J Soc Psychiatry 46(4):306–313
Kiejna A et al (2015) Predictors of vocational status in schizophrenia patients—results from the Polish Nationwide Survey. Int J Soc Psychiatry 61(8):824–831
Gilbert E, Marwaha S (2013) Predictors of employment in bipolar disorder: a systematic review. J Affect Disord 145(2):156–164
Chiripanhura B, Alun E, Ye Zhang T (2019) Labour market economic commentary: July 2019. Office for National Statistics, London
Li Y, Heath A (2020) Persisting disadvantages: a study of labour market dynamics of ethnic unemployment and earnings in the UK (2009–2015). J Ethnic Migr Stud 46(5):857–878
Amasowomwan F et al (2021) Good work: Employment support for black people with long-term health conditions. McPin Foundation, London
Okoroji C, Ibison Y, Robotham D (2023) Employment support for black people with long-term health conditions: a systematic narrative review of UK studies. J Mental Health:1–8
Royal College of Psychiatrists (2021) National Clinical Audit of Psychosis—Employment spotlight audit 2021. Healthcare Quality Improvement Partnership, London
National Institute for Health and Care Excellence (2015) Psychosis and schizophrenia in adults, Quality statement 5: supported employment programmes. https://www.nice.org.uk/guidance/qs80/chapter/quality-statement-5-supported-employment-programmes#:~:text=It%20is%20estimated%20that%20just,unemployed%20than%20the%20general%20population.
Jackson RG et al (2017) Natural language processing to extract symptoms of severe mental illness from clinical text: the Clinical Record Interactive Search Comprehensive Data Extraction (CRIS-CODE) project. BMJ Open 7(1):e012012
Morgan C et al (2014) Reappraising the long-term course and outcome of psychotic disorders: the AESOP-10 study. Psychol Med 44(13):2713
Tapfumaneyi A et al (2015) Predictors of vocational activity over the first year in inner-city early intervention in psychosis services. Early Interv Psychiatry 9(6):447–458
Das-Munshi J et al (2019) Ethnic density and other neighbourhood associations for mortality in severe mental illness: a retrospective cohort study with multi-level analysis from an urbanised and ethnically diverse location in the UK. Lancet Psychiatry 6(6):506–517
Patra BG et al (2021) Extracting social determinants of health from electronic health records using natural language processing: a systematic review. J Am Med Inform Assoc 28(12):2716–2727
Moscrop A et al (2020) If social determinants of health are so important, shouldn’t we ask patients about them? BMJ 371:m4150
Barnett P et al (2022) Interventions to improve social circumstances of people with mental health conditions: a rapid evidence synthesis. BMC Psychiatry 22(1):1–68
Acknowledgements
We appreciated the technical support from informatics personnel in the NIHR Biomedical Research Centre (BRC). We thank Dr Xingyi Song at the University of Sheffield, Dr Anna Kolliakou and Dr Angus Roberts at the BRC for their work developing the occupation application, which was used in this study.
Funding
This work was supported by the Clinical Record Interactive Search system which is funded by the National Institute for Health Research (NIHR) Maudsley Biomedical Research Centre at South London and Maudsley National Health Service (NHS) Foundation Trust and King’s College London and a joint infrastructure grant from Guy’s and St Thomas’ Charity and the Maudsley Charity. Specific occupation application development was supported by the Health Foundation working together with the Academy of Medical Sciences, in a clinician scientist fellowship held by JD-M. NC is funded by the Economic and Social Research Council for a PhD studentship under the London Interdisciplinary Social Science Doctoral Training Partnership (grand number ES/P000703/1). JD-M, SLH and CM are part supported by the ESRC Centre for Society and Mental Health at King’s College London (ESRC Reference: ES/S012567/1). JDM and SLH are partly supported by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King's College London. JDM and RS are supported by the National Institute for Health Research (NIHR) Applied Research Collaboration South London (NIHR ARC South London) at King’s College Hospital NHS Foundation Trust. RS is additionally supported by UKRI – Medical Research Council through the DATAMIND HDR UK Mental Health Data Hub (MRC reference: MR/W014386), and through the UK Prevention Research Partnership (Violence, Health and Society; MR-VO49879/1), an initiative funded by UK Research and Innovation Councils, the Department of Health and Social Care (England) and the UK devolved administrations, and leading health research charities. For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising.
Author information
Authors and Affiliations
Contributions
JD-M and NC conceived of the study and developed the study design with all authors. All analyses were led by NC with JD-M. NC and JD-M led interpretation of findings, and all authors contributed to the interpretation of findings for the work. NC led the first draft of the manuscript with JD-M. The manuscript was revised following contributions to revision from all authors. The final version of the paper has been approved by all authors and all authors agree to be accountable for all aspects of the work.
Corresponding author
Ethics declarations
Conflict of interest
RS declares research support received in the last 36 months from Janssen, GSK and Takeda. CM is the Editor-in-Chief of Social Psychiatry and Psychiatric Epidemiology. All other authors have no competing interests to declare that are relevant to the content of this article.
Ethical approval
The SLaM Case Register has been approved as a source of de-identified data for secondary analyses (Oxford Research Ethics Committee C, reference 18/SC/0372). This approval was extended in 2023 (reference 23/SC/0257). The study protocol was reviewed and approved by the patient-led CRIS oversight committee (reference: 21–051).
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Chilman, N., Laporte, D., Dorrington, S. et al. Understanding social and clinical associations with unemployment for people with schizophrenia and bipolar disorders: large-scale health records study. Soc Psychiatry Psychiatr Epidemiol 59, 1709–1719 (2024). https://doi.org/10.1007/s00127-024-02620-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00127-024-02620-6