Unmet need for chronic mental ill health: A population-based record linkage study
Main Article Content
Abstract
Introduction
It is widely recognised that many people with chronic mental ill health do not receive the treatment they need, with socio-demographic and socio-economic factors contributing to mental healthcare access and receipt of treatment. However, recently different perceptions have emerged following reports of over-treatment or over-medication of a proportion of the population presenting to primary care settings with poor mental health.
Objectives
This study aimed to investigate which factors are associated with the presence of poor mental health and the likelihood of being in receipt of pharmacological treatment for those who report chronic mental ill health.
Methods
The methodology was a record linkage of the 2011 Northern Ireland Census returns and a population-wide database of prescribed medications. The presence of a chronic mental health condition, as assessed through a self-reported question in the 2011 Census, was compared to electronic administrative records on psychotropic medication use in the twelve months following the Census.
Results
This study shows that those of non-white background (OR 0.38 95%CI 0.26-0.54), who were never married (OR 0.67 95%CI 061-0.73), unemployed (OR 0.65 95%CI 0.53-0.81) or living in a rural area (OR 0.88 95%CI 0.79-0.98) were less likely than their peers to receive medication for their poor mental health.
Conclusions
Some but not all of the observed variations may represent unmet need, but further studies are required to understand the knowledge of and attitudes towards healthcare services of groups (such as ethnic minorities and unemployed) who are identified as being less likely to receive treatment.
Highlights
- Self-reported chronic mental ill health varied by socio-demographic and socio-economic characteristics.
- Although a high proportion of people received psychotropic medication for their mental ill health, there were evident social patterns among those who do not receive treatment.
- Ethnic minorities, those whose were not married or unemployed were considerably more likely to experience unmet need.
Introduction
It is generally accepted that a significant proportion of people with mental ill health are under-diagnosed and untreated in primary care settings [1,2]. The presence of unmet need referring to a treatment gap between poor mental health status and treatment received by individuals with mental health problems has been previously highlighted at a United Kingdom [3,4] and Europe wide level [5]. There is a range of societal, cultural and individual factors that can influence treatment practices and lead to under-treating individuals. This could be linked to help-seeking behaviours, with men [6,7] and people residing in rural settings [8] being less likely to seek treatment. A similar effect has also been observed for ethnic minorities [9] and people living in socially deprived areas [4]. When focusing on medication rates, lower uptake of psychotropic prescriptions such as antidepressants has been identified among individuals of low educational and socio-economic status [10]. It is important to note that low patterns of psychotropic medication can lead to higher rates of distress due to untreated mental ill health problems [11].
However, the current rates of medication prescription for mental ill health across the world have recently become the subject of some debate. Perceptions have shifted suggesting that people with poor mental health are more likely to be over than under-medicated, though this varies across different healthcare systems. Reports of over-treatment [12] from some parts of the world such as the United States have called attention to the higher than recommended levels of medications and especially antidepressants [13]. Over-treatment can disproportionately affect individuals with mild to less severe symptoms, as they are more likely to receive antidepressants due to insufficient time for provision of alternative treatment options [14], despite the fact that medication might not be either recommended [15] or considered efficient treatment for those patients [16]. At population level, practices of over-prescription can potentially lead to worsening mental health problems instead of alleviating them [12].
In that context, the ability of general practitioners (GPs) to identify individuals with mental health conditions presenting in primary care settings, can potentially lead to either over- or underestimations of mental ill health, as 90% of individuals with mental health problems are managed within primary care settings [17]. Consequently, the accuracy of clinical diagnoses of mental health conditions in primary care can affect the likelihood of receiving treatment as GPs’ accuracy in identifying cases has been shown to vary [18].
Despite the potential impact of variations in medication practices, with both under-treatment and over-prescription being highlighted in the literature, there is still a lot of uncertainty stemming from methodological limitations of previous studies. With the exception of a number of Scandinavian studies [19,20], many earlier studies have been unable to accurately measure the level of treatment received on a community level. Additional issues include the use of self-reported measures of medication uptake [21] that might cause issues due to recall bias, non-representative samples linked to non-response rates [22] and cross-sectional study designs [5] that can only provide a snapshot of the population at one time-point. It has been further proposed that a more appropriate assessment of mental ill health would account for both the presence of a mental health condition and its overall impact on the individual [23].
If we make the assumption that prescription of psychotropic medications is the predominant form of treatment for mental illness, at least in Northern Ireland, then dissonances between those who report chronic poor mental health and those who receive psychotropic medications may indicate subgroups in the population whose needs are not being addressed by current approaches. The use of linked administrative data related to need and treatment across the entire population can overcome methodological hurdles and provide an indicator of areas of possible unmet need.
In this paper we therefore have two aims; (i) to establish the proportions and demographic and socio-economic characteristics of individuals reporting chronic poor mental health and (ii) for those with chronic poor mental health, to quantify the factors related to not being in receipt of psychotropic medication.
Methods
Data Sources
This study utilised data drawn from the Northern Ireland Longitudinal Study (NILS), which was formed in 2001 with a representative 28% sample of the NI population and is maintained by the Northern Ireland Statistics and Research Agency (NISRA). It comprises a linkage of the Northern Ireland Health Card Registration database (ensuring universal and free at-the-point-of-service healthcare) and Census returns, to which routinely collected vital statistics events such as births, deaths and migrations are linked on a regular basis. A description of the full cohort and the linkage methodology that was implemented is available elsewhere [24]. The NILS cohort was linked to a population-wide prescribing dataset through the use of the Health and Care Number, a unique identifier enabling linkage to other health service databases, including the centralised collation of primary care prescriptions (see below).
Study population
The study compared self-reports of chronic mental ill health from the 2011 Census returns (March 27th of that year) to psychotropic prescribing uptake for enumerated individuals in the twelve months following the Census. A total of 37,342 NILS members aged 16 and over were enumerated in the 2011 Census. People living in institutional settings (N=4,836) and those with missing data on the health conditions question (N=30,638) were excluded. For this analysis, a subset of 286,717 NILS members aged between twenty-five and seventy-four years at Census were selected as these represent the population group for whom socio-economic status can be reasonably stated (excluding those aged >25 & <74 years: N=56,151). All personal characteristics were drawn from the Census: these include age (in ten-year age-bands), sex and marital status (grouped as married; never married; and a single group classifying those widowed, separated and divorced). Northern Ireland is ethnically relatively homogenous, so only two ethnic groups were feasible (white; non-white). Additional socio-economic and area-level factors thought to be associated with mental health stigma were also included. Socio-economic status was determined using several indicators from the Census: these include educational attainment (grouped as no formal qualifications, intermediate level, degree-level); occupational social class derived using the National Statistics Socio-economic Classification (NS-SEC) [25] – with categories as shown in Table 1; housing tenure (owner-occupation, renting) and household car availability (two or more cars, one, no cars).
Assessment of chronic ill health in the Census
The 2011 Census included a question asking about the presence of “any of the following conditions which have lasted, or are expected to last, at least 12 months?” Respondents were asked to tick all applicable conditions from a list of ten (ranging from shortness of breath to sensory problems). For this study, we examined those respondents who ticked the category “an emotional, psychological or mental health condition (such as depression or schizophrenia)” . For brevity, we term this a chronic mental health problem. This measure has been used in a number of studies [26] providing similar estimates of population mental health with standardised measures such as the General Health Questionnaire (GHQ) and demonstrating a high degree of validity [27]. We also examined responses to another group - “shortness of breath or difficulty breathing (such as asthma)” - as a specificity comparison group for those seeking or receiving treatment for a respiratory problem (See Appendix). Breathing difficulty was chosen as it is quite common in the general population, can present across different age groups and is not considered a stigmatised condition. Although the 2011 Census included other chronic health conditions, these would not allow for a comparison between self-reported health conditions and medication uptake, as most physical conditions included are not treated pharmacologically at the point of contact (for example visual or hearing difficulties).
Area characteristics
Area-level characteristics were assessed using: (i) a measure of population density based on the NISRA Classification of Settlements [28], grouping individuals as living in urban, intermediate and rural settlements of >75,000 people, 2,500-75,000 and < 2,500 people respectively and (ii) area-level deprivation based on uptake of means-tested social security benefits [29] and calculated for 890 super output areas (SOA), each with an average population of approximately 2,000 people and subsequently grouped into quintiles ranging from least to most deprived.
Medication prescription
In Northern Ireland, the National Health Service provides universal health coverage free at the point of use, including prescriptions. Medications prescribed by GPs and dispensed from community pharmacies are (since 2009) collated centrally in an Enhanced Prescribing Database (EPD). This comprises of data on prescribed medicines, including information on the name of the dispensed medicine, its British National Formulary (BNF) category (a standard United Kingdom referencing system) and the date it was dispensed. For this study information on four categories of psychotropic medication was extracted: anxiolytic and antidepressant medications (BNF categories 4.1.2 and 4.1.3 respectively), anti-psychotic medication (oral and depot; BNF categories 4.2.1 and 4.2.2) and anti-mania medications (BNF categories 4.2.3). We also extracted information on respiratory medication (BNF Chapter 3, excluding antihistamine medication) to examine the uptake of medication for a common non-stigmatising physical health condition (in this case breathing difficulties) (See Appendix). This would act as a comparator to the receipt of psychiatric medication and help differentiate the socio-demographic correlates of stigma, which would be found only with the mental health medications from other explanations of low uptake, such as lower attendance at GPs or lack of knowledge of the system, which should be common to both types of medication. Mediation prescription use was assessed by summing the monthly usage, to include receipt of any prescription for medication thus covering every month during the twelve-month period following the Census. For depot injections, at least one such prescription during the same period was deemed sufficient.
NILS and the EPD were linked through the NI Health and Care Number by the data custodians, with all key matching fields and other potential identifiers removed before being made available to the researchers. All analysis was carried out within the NISRA secure setting. The study was approved by the Office for Research Ethics Committees Northern Ireland.
Analysis strategy
Descriptive statistics were used to describe the characteristics of the study population, as well as, the link between these characteristics and reporting a chronic mental health problem in the Census. These included individual (age, sex, ethnicity, marital status, education, social class), household (housing tenure, household car access) and area (locale of residence, area-level deprivation) level factors. Among those individuals that reported chronic mental ill health, variations in the receipt of a psychotropic medication prescription in the following twelve months was also explored in relation to these characteristics. Logistic regression models were implemented to investigate how the likelihood of reporting chronic mental ill health varied by the aforementioned population characteristics. Two steps of logistic regression models were presented, namely: i) a model specifying all population characteristic separately, each minimally adjusting for age/ sex and ii) a fully adjusted model, including all population characteristics described above. These two model structures were repeated to explore the characteristics associated with the likelihood of receiving a prescription in the twelve-month period among those reporting a chronic mental health condition. As both seeking and receiving pharmacological treatment have been shown to vary between individuals presenting with physical versus mental ill health, the analysis was repeated using chronic respiratory conditions as the outcome of interest (See Appendix). This provided a comparison aiming to identify whether uptake variation by socio-demographic characteristics was observed across individuals with physical and mental ill health.”
Results
Our cohort consisted of 23,803 individuals (8.3% of study population) who reported that they had a long-standing mental, emotional or psychological problem at the time of the Census. Table 1 shows that prevalence rates of chronic mental ill health were higher among individuals who were middle-aged (45-54 years: 10.6% and 55-64 years: 10.1%), female (9.5% versus male: 7.0%), of white ethnicity (8.4% versus non-white: 3.3%), separated/divorced/widowed (16.5% versus married: 5.5%) and with no academic qualifications (13.8%). In terms of household characteristics, lower social class (except for those unemployed: 8.3% versus never worked: 24.8%) as well as other socio-economic factors such as being in rented accommodation (16.7% versus owner-occupiers: 5.6%) and having no access to a car (20.3% versus \(\geq\) 2 cars: 4.1%) were linked to increased rates of reporting chronic poor mental health in the Census. A gradient association was observed between reporting chronic mental ill health and the level of area deprivation with the highest proportion being identified among those residing in the most deprived areas (14.2% versus least deprived:4.6%).
After adjustment for a range of individual, household and area-level factors, older individuals were less likely than younger individuals to report the presence of a chronic mental health condition at the time of the Census (65-74 years: OR=0.80: 95%CI=0.75-0.85), though that was not the case during the initial model (OR=1.00: 95%CI=0.95-1.06) suggesting the presence of a complex interplay between a range of socio-demographic factors and disclosure of chronic mental ill health. Mixed results were also observed in relation to the association between social class and chronic mental health status, with those unemployed shifting from being twice as likely as those in managerial positions to report mental ill health (OR=2.00: 95%CI=1.82-2.20) to being less likely to do so after full adjustment (OR=0.69: 95%CI=0.62-0.76). Those of non-white ethnic background were significantly less likely to report poor mental health (OR=0.38: 95%CI=00.32-0.45). People in rural and intermediate areas were less likely than their urban dwelling peers to report poor mental health but on further adjustment including adjustment for housing tenure and car ownership, the urban/rural differences were greatly attenuated.
No prescription in the 12 months following Census | Received prescription in the 12 months following Census | Rates of individuals receiving prescription in the 12 months following Census | Adjusted: age & sex |
Fully Adjusted$ |
|||
---|---|---|---|---|---|---|---|
N(%) | N (%) | % | OR (95% CI) | OR (95% CI) | |||
Cohort | 4,412 (100.0) | 19,391 (100.0) | 81.50% | ||||
25-34 | 946 (21.4) | 2,567 (13.2) | 73.10% | 1.00 | 1.00 | ||
35-44 | 1,047 (23.7) | 4,643 (23.9) | 81.60% | 1.65 (1.49-1.82) | 1.55 (1.39-1.72) | ||
Age | 45-54 | 1,121 (25.4) | 5,930 (30.6 | 84.10% | 2.02 (1.83-2.23) | 1.82 (1.63-2.03) | |
55-64 | 872 (19,8) | 4,450 (22.9) | 83.60% | 1.99 (1.79-2.21) | 1.71 (1.51-1.93) | ||
65-74 | 426 (9.7) | 1,801 (9.4) | 80.90% | 1.64 (1.44-1.86) | 1.37 (1.18-1.58) | ||
Male | 2,284 (51.8) | 7,336 (37.8) | 76.30% | 1.00 | 1.00 | ||
Sex | Female | 2,128 (48.2) | 12,055 (62.2) | 85.00% | 1.81 (1.70-1.94) | 1.78 (1.66-1.90) | |
White | 4,360 (98.8) | 19,314 (99.6) | 81.60% | 1.00 | 1.00 | ||
Ethnicity | Non-white | 55 (0.2) | 77 (0.4) | 59.70% | 0.36 (0.26-0.52) | 0.38 (0.26-0.54) | |
Married | 1,502 (34.0) | 7,837 (40.4) | 83.90% | 1.00 | 1.00 | ||
Marital status | Never married | 1,643 (37.2) | 5,313 (27.4) | 76.40% | 0.76 (0.69-0.82) | 0.67 (0.61-0.73) | |
Sep-Div-Wid | 1,267 (28.8) | 6,241 (32.2) | 83.10% | 0.86 (0.79-0.94) | 0.75 (0.68-0.82) | ||
Degree-level | 821 (18.6) | 3,097 (16.0 | 79.10% | 1.00 | 1.00 | ||
Education | Intermediate | 1,718 (38.9) | 7,195 (37.1) | 80.70% | 1.13 (1.03-1.25) | 1.06 (0.96-1.17) | |
No qualifications | 1,873 (42.5) | 9,099 (46.9) | 82.90% | 1.25 (1.13-1.37) | 1.14 (1.02-1.27) | ||
Managerial/Prof | 776 (17.6) | 3,065 (15.8 | 79.80% | 1.00 | 1.00 | ||
Intermediate | 608 (13.8) | 2,735 (14.1) | 81.80% | 1.11 (0.98-1.25) | 1.07 (0.95-1.22) | ||
Own account | 267 (6.0) | 912 (4.7) | 77.40% | 1.02 (0.87-1.20) | 0.96 (0.81-1.13) | ||
Social class/ | Semi-routine | 1,119 (25.4) | 5,440 (28.1) | 82.50% | 1.22 (1.10-1.35) | 1.13 (1.01-1.26) | |
Economic | Routine | 839 (19.0) | 4,243 (21.9) | 83.50% | 1.30 (1.17-1.45) | 1.16 (1.03-1.32) | |
activity | Never worked | 599 (13.5) | 2,466 (12.7) | 80.50% | 1.10 (0.97-1.24) | 1.01 (0.88-1.16) | |
Unemployed | 153 (3.5) | 369 (1.9) | 70.70% | 0.69 (0.56-0.85) | 0.65 (0.53-0.81) | ||
Student | 51 (1.2) | 161 (0.8) | 75.90% | 0.87 (0.63-1.21) | 0.84 (0.60-1.17) | ||
Housing tenure | Owner | 205,096 (78.0) | 12,242 (51.4) | 5.60% | 1.00 | 1.00 | |
Renting | 57,818 (22.0) | 11,561 (48.6) | 16.70% | 3.72 (3.62-3.83) | 1.75 (1.69-1.81) | ||
Two or more cars | 136,957 (52.1) | 5,906 (24.8) | 4.10% | 1.00 | 1.00 | ||
Household | One car | 95,006 (36.1) | 10,022 (42.1) | 9.50% | 2.62 (2.54-2.71) | 1.69 (1.62-1.75) | |
car access | No car | 30,891 (11.8) | 7,875 (33.1) | 20.30% | 6.42 (6.19-6.66) | 2.37 (2.26-2.48) | |
Urban | 90,917 (34.6) | 9,677 (40.7) | 9.60% | 1.00 | 1.00 | ||
Locale | Intermediate | 81,731 (31.1) | 7,732 (32.5) | 8.70% | 0.89 (0.86-0.92) | 0.95 (0.92-0.98) | |
of residence | Rural | 58,298 (22.2) | 3,798 (16.0) | 6.10% | 0.61 (0.59-0.63) | 0.98 (0.93-1.02) | |
Missing | 31,968 (12.1) | 2,596 (10.8) | 7.50% | 0.79 (0.75-0.83) | 1.05 (1.00-1.10) | ||
Least deprived | 52,903 (20.1) | 2,580 (10.8) | 4.60% | 1.00 | 1.00 | ||
Area-level | Less deprived | 56,497 (21.5) | 3,703 (15.6) | 6.20% | 1.37 (1.30-1.45) | 1.12 (1.06-1.18) | |
deprivation | Average | 53,841 (20.5) | 4,248 (17.8) | 7.30% | 1.67 (1.59-1.76) | 1.18 (1.12-1.24) | |
(SOA) | More deprived | 52,116 (19.8) | 5,647 (23.7) | 9.80% | 2.31 (2.20-2.43) | 1.28 (1.22-1.35) | |
Most deprived | 45,501 (17.3) | 7,497 (31.5) | 14.20% | 3.56 (3.40-3.73) | 1.36 (1.29-1.43) | ||
Missing | 1,796 (0.8) | 128 (0.6) | 6.70% | 1.53 (1.27-1.83) | 1.06 (0.87-1.29) |
A total of 19,391 (81.5%) cohort members were in receipt of psychotropic medication in the twelve months following the Census, leaving 18.5% with no prescription for their chronic mental health problems. Of those reporting chronic mental ill health, the higher uptake of pharmacological treatment in the following period was observed among females rather than males (85.0% and 76.3% respectively) and the middle-aged (45-54: 84.1% and 55-64: 83.6% respectively) compared to younger people (25-34: 73.1%). Although those in the 65-74 year age-group were less likely to report poor mental health, they were still more likely to receive medication in comparison to younger individuals (OR=1.37: 95%CI=1.18-1.58), potentially suggesting variations in presentation and reporting by patients or prescribing propensity by GPs. Lower psychotropic medication uptake was observed among the never married (76.4% versus married: 83.9%) who, along with the separated/widowed/divorced, were less likely to be in receipt of medication than those married (OR=0.67: 95%CI=0.61-0.73; and OR=0.75: 95%CI=0.68-0.82 respectively). This is in contrast to the higher rates of self-reported mental ill health among those two groups, suggesting an imbalance between need and treatment. Furthermore, much lower rates of pharmacological treatment were identified among individuals with a non-white background when compared with those of white ethnic background (59.7% and 81.6% respectively) which, after further adjustment for potential confounders, equated to an odds ratio of 0.38 (95%CI=0.26-0.54).
The relationship between prescribing and socio-economic status was not straightforward, though generally higher for those in the lower social strata. Those who were unemployed were also less likely to be prescribed psychotropic medication in comparison to those of managerial status (OR=0.65: 95%CI=0.53-0.81), while those in semi-routine and routine social classes (OR=1.13: 95%CI=1.01-1.26 and OR=1.16: 95%CI=1.03-1.32 respectively) were more likely than those in professional classes to be receiving medication. Prescription rates were also higher for those in rented accommodation than those in owner occupation (OR=1.28: 95%CI=1.18-1.38). Finally, the analysis showed that patients from rural areas were less likely to be in receipt of prescriptions when compared to urban areas (OR=0.88: 95%CI=0.79-0.98), with no difference between intermediate areas and urban residents.
No prescription in the 12 months following Census | Received prescription in the 12 months following Census | Rates of individuals receiving prescription in the 12 months following Census | Adjusted: age & sex |
Fully Adjusted$ |
|||
---|---|---|---|---|---|---|---|
N(%) | N (%) | % | OR (95% CI) | OR (95% CI) | |||
Cohort | 4,412 (100.0) | 19,391 (100.0) | 81.50% | ||||
25-34 | 946 (21.4) | 2,567 (13.2) | 73.10% | 1.00 | 1.00 | ||
Age | 35-44 | 1,047 (23.7) | 4,643 (23.9) | 81.60% | 1.65 (1.49-1.82) | 1.55 (1.39-1.72) | |
45-54 | 1,121 (25.4) | 5,930 (30.6 | 84.10% | 2.02 (1.83-2.23) | 1.82 (1.63-2.03) | ||
55-64 | 872 (19,8) | 4,450 (22.9) | 83.60% | 1.99 (1.79-2.21) | 1.71 (1.51-1.93) | ||
65-74 | 426 (9.7) | 1,801 (9.4) | 80.90% | 1.64 (1.44-1.86) | 1.37 (1.18-1.58) | ||
Male | 2,284 (51.8) | 7,336 (37.8) | 76.30% | 1.00 | 1.00 | ||
Sex | Female | 2,128 (48.2) | 12,055 (62.2) | 85.00% | 1.81 (1.70-1.94) | 1.78 (1.66-1.90) | |
White | 4,360 (98.8) | 19,314 (99.6) | 81.60% | 1.00 | 1.00 | ||
Ethnicity | Non-white | 55 (0.2) | 77 (0.4) | 59.70% | 0.36 (0.26-0.52) | 0.38 (0.26-0.54) | |
Married | 1,502 (34.0) | 7,837 (40.4) | 83.90% | 1.00 | 1.00 | ||
Marital | Never married | 1,643 (37.2) | 5,313 (27.4) | 76.40% | 0.76 (0.69-0.82) | 0.67 (0.61-0.73) | |
status | Sep-Div-Wid | 1,267 (28.8) | 6,241 (32.2) | 83.10% | 0.86 (0.79-0.94) | 0.75 (0.68-0.82) | |
Degree-level | 821 (18.6) | 3,097 (16.0 | 79.10% | 1.00 | 1.00 | ||
Education | Intermediate | 1,718 (38.9) | 7,195 (37.1) | 80.70% | 1.13 (1.03-1.25) | 1.06 (0.96-1.17) | |
No qualifications | 1,873 (42.5) | 9,099 (46.9) | 82.90% | 1.25 (1.13-1.37) | 1.14 (1.02-1.27) | ||
Managerial/Prof | 776 (17.6) | 3,065 (15.8 | 79.80% | 1.00 | 1.00 | ||
Intermediate | 608 (13.8) | 2,735 (14.1) | 81.80% | 1.11 (0.98-1.25) | 1.07 (0.95-1.22) | ||
Own account | 267 (6.0) | 912 (4.7) | 77.40% | 1.02 (0.87-1.20) | 0.96 (0.81-1.13) | ||
Social class/ | Semi-routine | 1,119 (25.4) | 5,440 (28.1) | 82.50% | 1.22 (1.10-1.35) | 1.13 (1.01-1.26) | |
Economic | Routine | 839 (19.0) | 4,243 (21.9) | 83.50% | 1.30 (1.17-1.45) | 1.16 (1.03-1.32) | |
activity | Never worked | 599 (13.5) | 2,466 (12.7) | 80.50% | 1.10 (0.97-1.24) | 1.01 (0.88-1.16) | |
Unemployed | 153 (3.5) | 369 (1.9) | 70.70% | 0.69 (0.56-0.85) | 0.65 (0.53-0.81) | ||
Student | 51 (1.2) | 161 (0.8) | 75.90% | 0.87 (0.63-1.21) | 0.84 (0.60-1.17) | ||
Housing | Owner | 2,378 (53.9) | 9,864 (50.9) | 80.60% | 1.00 | 1.00 | |
tenure | Renting | 2,034 (46.1) | 9,527 (49.1) | 82.40% | 1.21 (1.13-1.30) | 1.28 (1.18-1.38) | |
Two or more cars | 1,128 (25.6) | 4,778 (24.6) | 80.90% | 1.00 | 1.00 | ||
Household | One car | 1,836 (41.6) | 8,186 (42.2) | 81.70% | 1.07 (0.99-1.17) | 1.06 (0.97-1.16) | |
car access | No car | 1,448 (32.8) | 6,427 (33.1) | 81.60% | 1.10 (1.00-1.20) | 1.09 (0.97-1.23) | |
Urban | 1,775 (40.2) | 7,892 (40.7) | 81.60% | 1.00 | 1.00 | ||
Locale of | Intermediate | 1,327 (30.1) | 6,360 (32.8 | 82.30% | 1.06 (0.98-1.14) | 1.01 (0.93-1.10) | |
residence | Rural | 755 (17.1) | 3,043 (15.7) | 80.10% | 0.91 (0.82-0.99) | 0.88 (0.79-0.98) | |
Missing | 500 (11.3) | 2,096 (10.8) | 80.70% | 0.99 (0.89-1.11) | 0.97 (0.86-1.09) | ||
Least deprived | 524 (11.9) | 2,056 (10.6) | 79.70% | 1.00 | 1.00 | ||
Area-level | Less deprived | 682 (15.5) | 3,021 (15.6) | 81.60% | 1.13 (0.99-1.28) | 1.09 (0.96-1.25) | |
deprivation | Average | 779 (17.7) | 3,469 (17.9) | 81.70% | 1.15 (1.02-1.31) | 1.10 (0.97-1.25) | |
(SOA) | More deprived | 988 (22.4) | 4,659 (24.0) | 82.50% | 1.21 (1.08-1.37) | 1.10 (0.97-1.25) | |
Most deprived | 1,415 (32.0) | 6,082 (31.4) | 81.10% | 1.11 (1.00-1.25) | 0.98 (0.87-1.11) | ||
Missing | 24 (0.5) | 104 (0.5) | 81.50% | 1.11 (0.70-1.76) | 1.07 (0.67-1.71) |
The likelihood of receiving respiratory medication for the cohort of 25,983 people in the Census who reported chronic breathing difficulties was also explored (See Appendix). Overall, 82% of these people received at least one prescription in the following 12 months; this was shown to vary by socio-demographic factors though in a different manner to psychotropic medications. Uptake of respiratory medications exhibited a slightly stronger age gradient, perhaps reflecting the milder and more intermittent breathing difficulties (such as asthma) at younger ages. Prescriptions were also higher amongst females than males, though the difference (OR=1.21: 95%CI=1.15-1.28) was much lower than that observed among individuals with chronic mental ill health. The differences between white and non-white ethnic groups were also much lower for respiratory than for psychotropic medications (OR=0.78: 95%CI-0.56-1.07 and OR=0.38: 95%CI=0.26-0.54 respectively). Furthermore, no socio-economic, household or area-level variations were detected, indicating differing medication uptake patterns between individuals with chronic physical and mental health.
Discussion
This large population wide study sought to quantify variations in the prevalence of poor mental health as assessed by self-report in the 2011 Census and, through record linkage, to examine variations in treatment levels for those who reported poor mental health, as assessed by uptake of psychotropic medications in the 12 months following the Census. This study confirmed the established socio-demographic correlates and socio-economic gradients in mental health; namely that the likelihood of poor mental health was higher in women, the middle aged, those who were not married and those who were more socially disadvantaged. Self-reported levels of poor mental health were particularly low amongst non-white ethnic minorities (OR=0.38: 95%CI=0.32-0.45).
The present study also demonstrated that most people (82%) reporting chronic mental ill health received at least one prescription for psychotropic medication over the following twelve months. This is perhaps reassuring in that a very high proportion is being treated, but perhaps worrying given the heavy reliance on pharmacological therapy. However, the remaining 18% of people with mental ill health who did not receive any pharmacological treatment did not appear to be randomly distributed; this could be further supported by the fact that the observed social patterns were not present in those not receiving treatment for breathing difficulties. Individuals who were not married, unemployed or were living in rural communities were less likely to receive treatment for poor mental health. Much of this variation might be understood within the current paradigms of social support [30] or variations in primary care consultation rates [31], but as equivalent social patterns of non-treatment of respiratory medications for people reporting breathing difficulties were either non-existent or greatly attenuated (See Appendix), it could be suggested that stigma might also play a significant part [32].
Rates of psychotropic prescriptions were higher in middle-aged individuals when compared to their younger counterparts, which is in accordance with previous reports [33]. However, it is unclear if this indicates under-treatment of young people, perhaps related to lower levels of presentation to primary care services by younger individuals [34] or a higher prevalence of non-pharmacological treatment at younger ages. Treatment rates for mental ill health were also substantially higher for women, perhaps because they are more likely to recognise and report their mental health conditions, and subsequently seek and receive mental health treatment [35]. However, small gender differences were also evident for receipt of respiratory medications (See Appendix), suggesting that differential access to healthcare services may also be important. The lower treatment rates in men is likely to be multifactorial including being less likely to recognise and acknowledge a significant mental health problem and to subsequently seek help [1], experiencing psychosocial barriers linked to stigmatising beliefs about mental ill health, and compounded by lower primary care attendance rates [36].
It is notable that while married individuals were less likely to report chronic mental ill health, married people were more likely than their non-married peers to receive psychotropic medication for their poor mental health. This is surprising given the higher rates of health service utilisation by widowed and divorced people [37]. However, marriage may be associated with a range of health promoting activities, perhaps arising from increased partner pressures or because of the perceived obligation to maintain health or of exposure to wider social influences [30]. This may explain the higher reported levels of cholesterol screening amongst married men and women in the US [38] and of breast screening in women in the UK [39].
The study confirmed the complex association between mental health and social status as medication use was generally and moderately higher among those more socially disadvantaged [10,40], though this was not the case for those who were unemployed at the time of the Census. It may be that the high prevalence of mental ill health reduces stigma and increases likelihood of recognition and propensity to seek treatment, though Holman, using the data from the British Social Attitudes survey, found that the relationship between social disadvantage and stigma was not straightforward and varied by sex and according to the measure of disadvantage [41]. The higher primary care attendance rates of the more disadvantaged [42], which is mostly explained by morbidity rates [43], may also present additional opportunities for diagnosis and treatment, though the alternative explanation that more affluent sufferers may be preferentially seeking and accessing psychological treatments cannot be discounted.
Ethnic minorities, referring to individuals of non-white ethnic background in this study, exhibited different self-reporting and treatment patterns. Although they were only about one third as likely as the white population to report chronic poor mental health (OR=0.38: 95%CI=0.32-0.45), those who were unwell were only about one third as likely to receive psychotropic medications in the follow-up period (OR=0.38: 95%CI=0.26-0.54). However, the interpretation of this is not straightforward and the explanations may be as various as the different constituent groups. On one hand the lower likelihood of treatment could be seen as evidence of the lower rates of healthcare access as previously reported among individuals from Pakistani, Indian and Bangladeshi backgrounds [44] and this is possibly supported by the slightly (though non-significant) lower likelihood of receiving medication for breathing difficulties. It is known that reduced language proficiency can act as a significant barrier to accessing available primary care services, and to receiving an appropriate diagnosis and treatment. This is perhaps more important for mental health issues [45], but unfortunately, proficiency in English was not available for this study. The lower medication rates could also signal the impediment of culturally-related mental health stigma among non-Western ethnic groups such as Chinese patients [32], and this might explain the much lower likelihood of being on psychotropic medications. However, the apparent treatment shortfall could be also be attributable to a greater use of alternative or complimentary medicine rather than traditional pharmacological treatment, which is known to be common amongst individuals from Non-European countries [46,47].
Finally, individuals in rural areas were less likely to be in receipt of pharmacological treatment, potentially highlighting the presence of different medication practices or help-seeking behaviours between residents of rural and urban areas. There is evidence that the prevalence of poor mental health is higher in cities though whether this is due to social drift or factors inherent to cities is unclear [8]. In the current study the prevalence of chronic poor mental health was considerably higher in cities in models adjusted only for age and sex but this association largely disappeared in the fully adjusted models; these may represent over adjustment as both housing tenure and car ownership have a recognised urban/rural bias. The higher mental health treatment rates in cities may be a consequence of a greater availability of community mental health services or psychiatric facilities in different areas; though difference in the attitudes of rural residents should also be considered as they might have a preference for self-reliance and thus do not engage with health services for a mental health problem [48]. It is noteworthy that for individuals with chronic breathing problems there was no link between area of residence and the likelihood to receive appropriate treatment; which might support the presence of a complex interplay between rurality, access to services and stigma avoidance of mental healthcare services that does not apply to physical ill health.
Limitations
This study has some significant strengths and limitations. It is a large population-based record-linkage study that utilised the 2011 Census, which provided prevalence estimates of chronic mental health status, and a centralised electronic database of prescribed medications, which included standardised categories of all medications prescribed by GPs and dispensed from community pharmacies across Northern Ireland. This approach obviated the selection biases inherent in many population surveys and provided an accurate and independent picture of treatment patterns. However, the use of the chronic mental health Census question might cause concern as it is obviously based on self-report, though the validity of the measure has been previously explored [27] and variations in mental ill health rates identified in this study, including the link between household and area deprivation, were in accordance with previous evidence on neighbourhood quality and mental health [49].
The study relies on receipt of medications prescribed by GPs and dispensed from community pharmacies, excluding medication prescribed in hospital settings, and this may have underestimated some further information biases as a small proportion of patients will have received a prescription but not have taken it to the pharmacist or not have used the dispensed medication. Further analysis into different psychotropic medication was limited by small numbers of individuals prescribed specific types medication for a more severe condition (such as schizophrenia). A notable omission is a lack of information on the use of non-pharmacological treatment such as the recommended Cognitive Behavioural Therapy [15]. Unfortunately, data on these treatment pathways were not readily available. However, the availability of these treatments is known to be limited in Northern Ireland and associated with significant delays [50], and in any event, it is recognised that even amongst these patients a high percentage will be also treated pharmacologically. We think it is likely that the inclusion of psychological therapies could perhaps have further accentuated rather than reduced the observed treatment patterns.
Finally, it is acknowledged that we do not have a true measure of need in terms of type or severity of ill health and therefore of the appropriateness of treatment. It is therefore possible that some of the differences between groups could represent over-treatment. Mitchell et al. [18] have indicated that GPs have a diagnostic sensitivity for depression of approximately 50% of “true cases” with false positives outnumbering missed cases, leading to more over-treatment than under-diagnosis [12].
Implications
This study suggests that social and demographic factors are closely but moderately associated with the likelihood of treatment for poor mental health on a population level. However, overcoming the associated barriers in help-seeking behaviours will not be easy. Where access to primary care is likely to be an issue perhaps an increased awareness amongst GPs would increase the potential for case-finding amongst these infrequent attenders when they do eventually present at primary care. It is evident that much more needs to be done to reduce the stigma in society associated with mental health, but it is less clear what works. Notwithstanding that many public education and social marketing campaigns, such as the “Time to Change” campaign [51,52], have been implemented to reduce stigma and its impact on mental health help-seeking, it remains an obstacle to improving quality of life for people with mental ill health [53] and overcoming barriers among specific socio-demographic groups on a population level [54]. Perhaps these campaigns also need to be more targeted. Campaigns already recognise that there is a need to focus on men, (with examples such as Beyondblue [55] and Spur Projects [56]), but there is possibly a need for additional focus on ethnic minority sub-populations and on people living in rural communities. Finally, there is a need to better understand the conceptualisation, reporting and treatment preferences of the different ethnic minorities. Further qualitative studies should try to unbundle the obvious heterogeneity within this community and determine the role of language and knowledge of the processes of health services, as well as, the extent to which the current lower levels of medication represent a preference for alternative treatment approaches or true under-treatment.
Conclusion
In this population-based study of individuals experiencing chronic mental ill health, the presence of social patterns in pharmacological treatment practices was identified as the likelihood of being in receipt of psychotropic medication varied by gender, ethnicity, marital and employment status. This study emphasises the need to focus on current prescribing practices and their link to healthcare inequalities across the population.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.
Authors’ contributions
FT, DOR and MR were responsible for the design of the study. FT performed the analysis, supervised by MR and DOR. All authors contributed to the intellectual content of the paper as well as offering revisions on drafts.
Appendix
Reported breathing problems at census | Received prescription in the 12 months following census | Adjusted: age & sex |
Fully Adjusted $ |
|||
---|---|---|---|---|---|---|
N(%) | N (%) | % | OR (95% CI) | |||
Cohort | 25,983 (9.1) | 21,306 (82.0) | ||||
25-34 | 3,386 (5.5) | 1,210 (35.7) | 1.00 | 1.00 | ||
35-44 | 4,032 (6.1) | 1,834 (45.5) | 1.50 (1.37-1.65) | 1.42 (1.29-1.57) | ||
Age | 45-54 | 5,421 (8.1) | 2,712 (50.0) | 1.80 (1.64-1.96) | 1.62 (1.47-1.79) | |
55-64 | 6,488 (12.3) | 3,319 (51.2) | 1.89 (1.74-2.06) | 1.66 (1.51-1.83) | ||
65-74 | 6,656 (17.0) | 3,407 (51.2) | 1.89 (1.74-2.06) | 1.64 (1.48-1.81) | ||
Male | 11,716 (8.5) | 5,318 (45.4) | 1.00 | 1.00 | ||
Sex | Female | 14,267 (9.6) | 7,164 (50.2) | 1.22 (1.16-1.28) | 1.21 (1.15-1.28) | |
White | 25,823 (9.1) | 12,419 (48.1) | 1.00 | 1.00 | ||
Ethnicity | Non-white | 160 (4.1) | 63 (39.4) | 0.76 (0.55-1.04) | 0.78 (0.56-1.07) | |
Married | 13,737 (8.1) | 6,781 (49.4) | 1.00 | 1.00 | ||
Marital status | Never married | 5,413 (7.6) | 2,240 (41.4) | 0.89 (0.83-0.95) | 0.87 (0.81-0.94) | |
Sep-Div-Wid | 6,833 (15.0) | 3,461 (50.7) | 0.99 (0.93-1.05) | 0.96 (0.90-1.02) | ||
Degree-level | 5,436 (6.0) | 2,436 (44.8) | 1.00 | 1.00 | ||
Education | Intermediate | 8,555 (7.4) | 3,884 (45.4) | 1.02 (0.96-1.10) | 1.03 (0.95-1.11) | |
No qualifications | 11,992 (15.1) | 6,162 (51.4) | 1.14 (1.07-1.22) | 1.15 (1.05-1.24) | ||
Managerial/Prof | 5,341 (6.2) | 2,456 (46.0) | 1.00 | 1.00 | ||
Intermediate | 3,821 (7.7) | 1,810 (47.4) | 1.01 (0.92-1.09) | 0.99 (0.90-1.08) | ||
Own account | 1,686 (7.5) | 766 (45.4) | 0.97 (0.87-1.09) | 0.93 (0.83-1.05) | ||
Social class/ | Semi-routine | 6,854 (10.3) | 3,317 (48.4) | 1.03 (0.96-1.11) | 0.98 (0.90-1.06) | |
Economic activity | Routine | 5,477 (13.4) | 2,817 (51.4) | 1.13 (1.05-1.22) | 1.05 (0.95-1.14) | |
Never worked | 2,137 (17.3) | 1,037 (48.5) | 1.02 (0.92-1.13) | 0.94 (0.83-1.05) | ||
Unemployed | 464 (7.4) | 193 (41.6) | 0.91 (0.75-1.10) | 0.87 (0.72-1.07) | ||
Student | 203 (8.0) | 86 (42.4) | 0.98 (0.74-1.31) | 0.95 (0.71-1.27) | ||
Housing tenure | Owner | 16,537 (7.6) | 7,923 (47.9) | 1.00 | 1.00 | |
Renting | 9,446 (13.6) | 4,559 (48.3) | 1.03 (0.98-1.08) | 1.01 (0.95-1.08) | ||
Two or more cars | 8,755 (6.1) | 4,106 (46.9) | 1.00 | 1.00 | ||
Household car access | One car | 11,203 (10.7) | 5,326 (48.3) | 0.99 (0.94-1.05) | 0.97 (0.91-1.04) | |
No car | 6,205 (16.0) | 3,050 (49.2) | 1.01 (0.95-1.08) | 0.98 (0.90-1.07) | ||
Urban | 10,481 (10.4) | 5,086 (48.5) | 1.00 | 1.00 | ||
Intermediate | 8,359 (9.3) | 4,055 (48.5) | 1.01 (0.96-1.07) | 1.02 (0.96-1.09) | ||
Locale of residence | Rural | 4,545 (7.3) | 2,134 (47.0) | 0.96 (0.89-1.03) | 0.98 (0.91-1.06) | |
Missing | 2,598 (7.5) | 1,207 (46.5) | 0.97 (0.89-1.06) | 0.99 (0.90-1.08) | ||
Least deprived | 3,752 (6.7) | 1,774 (47.3) | 1.00 | 1.00 | ||
Area-level deprivation (SOA) | Less deprived | 4,559 (7.6) | 2,114 (46.4) | 0.96 (0.88-1.04) | 0.95 (0.87-1.04) | |
Average | 4,760 (8.2) | 2,260 (47.5) | 1.00 (0.92-1.09) | 0.99 (0.90-1.08) | ||
More deprived | 5,628 (9.7) | 2,705 (48.1) | 1.01 (0.93-1.10) | 0.99 (0.91-1.07) | ||
Most deprived | 7,132 (13.5) | 3,554 (49.8) | 1.07 (0.99-1.16) | 1.05 (0.96-1.15) | ||
Missing | 152 (7.9) | 75 (49.3) | 1.03 (0.74-1.43) | 1.04 (0.74-1.45) |
Acknowledgements
The help provided by the staff of the Northern Ireland Longitudinal Study (NILS) and the NILS Research Support Unit is acknowledged. The NILS is funded by the Health and Social Care Research and Development Division of the Public Health Agency (HSC R&D Division) and NISRA. The NILS-RSU is funded by the ESRC and the Northern Ireland Government. The authors alone are responsible for the interpretation of the data and any views or opinions presented are solely those of the author and do not necessarily represent those of NISRA/NILS.
Statement on conflicts of interest
The authors declare no conflicts of interest.
Abbreviations
BNF | British National Formulary |
---|---|
EPD | Enhanced Prescribing Database |
GHQ | General Health Questionnaire |
GPs | General Practitioners |
NI | Northern Ireland |
NILS | Northern Ireland Longitudinal Study |
NISRA | Northern Ireland Statistics and Research Agency |
NS-SEC | National Statistics Socio-economic Classification |
SOA | Super Output Area |
Funding Statement
The authors received no specific funding for this work.
References
-
McManus S, Bebbington P, Jenkins R, Brugha T. Mental health and wellbeing in England: Adult Psychiatric Morbidity Survey 2014. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/556596/apms-2014-full-rpt.pdf Accessed April 2018.
-
Mann N. Unmet need in mental health provision. BMJ 2017;358:j364.
-
Bebbington P (2000). The need for psychiatric treatment in the general population. In: Unmet Need in Psychiatry. Cambridge:Cambridge University Press; 2000. p. 85-96.
-
McConnell P, Bebbington P, McClelland R, Gillespie K, Houghton S. Prevalence of psychiatric disorder and the need for psychiatric care in Northern Ireland: population study in the District of Derry. Br J Psychiatry 2002;181:214-219. 10.1192/bjp.181.3.214
10.1192/bjp.181.3.214 -
Alonso J, Codony M, Kovess V, Angermeyer MC, Katz SJ, Haro JM, De Girolamo G, De Graaf R, Demyttenaere K, Vilagut G, Almansa J, Lépine JP, Brugha TS. Population level of unmet need for mental healthcare in Europe. Br J Psychiatry 2007;190:299-306. 10.1192/bjp.bp.106.022004
10.1192/bjp.bp.106.022004 -
Fleury MJ, Ngui AN, Bamvita JM, Grenier G, Caron J. Predictors of Healthcare Service Utilization for Mental Health Reasons. Int J Environ Res Public Health 2014;11:10559-10586. 10.3390/ijerph111010559
10.3390/ijerph111010559 -
Hansen AH, Hoye A. Gender differences in the use of psychiatric outpatient specialist services in Tromsø, Norway are dependent on age: a population-based cross-sectional survey. BMC Health Serv Res 2015;15:477. 10.1186/s12913-015-1146-z
10.1186/s12913-015-1146-z -
Sundquist K, Frank G, Sundquist J. Urbanisation and incidence of psychosis and depression: follow up study of 4.4 million women and men in Sweden. Br J Psychiatry 2004;184:293–8. 10.1192/bjp.184.4.293
10.1192/bjp.184.4.293 -
Cabassa LJ, Zayas LH, Hansen MC. Latino adults' access to mental health care: a review of epidemiological studies. Adm Policy Ment Health 2006;33(3):316-330. 10.1007/s10488-006-0040-8
10.1007/s10488-006-0040-8 -
Kivimaki M, Gunnell D, Lawlor DA, Davey Smith G, Pentti J, Virtanen M, Elovainio M, Klaukka T, Vahtera J. Social inequalities in antidepressant treatment and mortality: A longitudinal register study. Psychol Med 2007;37(3):373-382. 10.1017/S0033291706009457
10.1017/S0033291706009457 -
Kessler RC, Demler O, Frank RG, Olfson M, Pincus HA, Walters EE, et al. Prevalence and treatment of mental disorders, 1990 to 2003. N Engl J M 2005;352:2515–2523. 10.1056/NEJMsa043266
10.1056/NEJMsa043266 -
Dowrick C, Frances A. Medicalising unhappiness: new classification of depression risks more patients being put on drug treatment from which they will not benefit. BMJ 2013;347:f7140. https://doi.org/10.1136/bmj.f7140
10.1136/bmj.f7140 -
Mojtabai R, Olfson M. National trends in long-term use of antidepressant medications: results from the U.S. National Health and Nutrition Examination Survey. J Clin Psychiatry 2014;75(2):169-177. 10.4088/JCP.13m08443
10.4088/JCP.13m08443 -
Olfson M, Blanco C, Marcus SC. Treatment of adult depression in the United States. JAMA Int Med 2016;176(10):1482-1491. 10.1001/jamainternmed.2016.5057
10.1001/jamainternmed.2016.5057 -
National Collaborating Centre for Mental Health. Depression: the treatment and management of depression in adults: NICE clinical guideline [CG90]. Available at: https://www.nice.org.uk/guidance/cg90 . Accessed April 2018.
-
Barbui C, Cipriani A, Patel V, Ayuso-Mateos JL, van Ommeren M. Efficacy of antidepressants and benzodiazepines in minor depression: systematic review and meta-analysis. Br J Psychiatry 2011;198(1):11-16. 10.1192/bjp.bp.109.076448
10.1192/bjp.bp.109.076448 -
NHS England. The five year forward view for mental health: A report from the Independent Mental Health Taskforce to the NHS in England. Available at: https://www.england.nhs.uk/wp-content/uploads/2016/02/Mental-Health-Taskforce-FYFV-final.pdf . Accessed April 2018.
-
Mitchell AJ, Vaze A, Rao S. Clinical diagnosis of depression in primary care: a meta-analysis. Lancet 2009;374:609-19. 10.1016/S0140-6736(09)60879-5
10.1016/S0140-6736(09)60879-5 -
Crump C, Sundquist K, Sundquist J, Winkleby M. Neighborhood deprivation and psychiatric medication prescription: a Swedish national multilevel study. Ann Epidemiol 2011;21(4):231–237. 10.1016/j.annepidem.2011.01.005
10.1016/j.annepidem.2011.01.005 -
Hollander AC, Bruce D, Burström B, Ekblad S. Gender-related mental health differences between refugees and non-refugee immigrants--a cross-sectional register-based study. BMC Public Health 2011;11:180. 10.1186/1471-2458-11-180
10.1186/1471-2458-11-180 -
Drieling RL, LaCroix AZ, Beresford SAA, Boudreau DM, Kooperberg C, Heckbert SR. Validity of Self-Reported Medication Use Compared With Pharmacy Records in a Cohort of Older Women: Findings From the Women’s Health Initiative. Am J Epidemiol 2016;184(3):233-238. 10.1093/aje/kwv446
10.1093/aje/kwv446 -
De Graaf R, Bijl RV, Smit F, Ravelli A, Vollebergh WA. Psychiatric and sociodemographic predictors of attrition in a longitudinal study: the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Am J Epidemiol 2000;152:1039-1047. 10.1093/aje/152.11.1039
10.1093/aje/152.11.1039 -
Narrow WE, Rae DS, Robins LN, Regier DA. Revised prevalence estimates of mental disorders in the United States: using a clinical significance criterion to reconcile 2 surveys’ estimates. Arch Gen Psychiatry 2002;59:115-123. https://doi.org/archpsyc.59.2.115
-
O’Reilly D, Rosato M, Catney G, Johnston F, Brolly M. Cohort description: the Northern Ireland Longitudinal Study (NILS). Int J Epidemiol 2012;41:634–41. 10.1093/ije/dyq271
10.1093/ije/dyq271 -
Rose D, Pevalin D. A researcher’s guide to the National Statistics Socio-economic Classification. London: Sage Publications; 2002.
-
Tseliou F, Maguire A, Donnelly M, O'Reilly D. The impact of childhood residential mobility on mental health outcomes in adolescence and early adulthood: a record linkage study. J Epidemiol Community Health 2016a;70(3):278-85. 10.1136/jech-2015-206123
10.1136/jech-2015-206123 -
Tseliou F, Maguire A, Donnelly M, O'Reilly D. The influence of mobility on mental health status in young people: The role of area-level deprivation. Health Place 2016b;42:96-103. https://doi.org/10.1016/j.healthplace.2016.09.001
10.1016/j.healthplace.2016.09.001 -
Northern Ireland Statistics and Research Agency. Northern Ireland Multiple Deprivation Measure 2010. Available at: https://www.nisra.gov.uk/sites/nisra.gov.uk/files/publications/NIMDM_2010_Report_0.pdf . Accessed April 2018.
-
Northern Ireland Statistics and Research Agency. Report of the Inter-Departmental Urban-Rural Definition Group. Statistical Classification and Delineation of Settlements 2005. Available at: http://www.ninis2.nisra.gov.uk/public/documents/ur_report.pdf . Accessed April 2018.
-
Cotten SR. Marital status and mental health revisited: Examining the importance of risk factors and resources. Fam. Relations An Interdiscip. J. Appl. Fam. Stud 1999;48:225–233. 10.2307/585631
10.2307/585631 -
Stirling, AM, Wilson P, McConnachie A. Deprivation, psychological distress, and consultation length in general practice. Br J Gen Pract 2001;51:456–460. Available from: http://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC1314026&blobtype=pdf
-
Chung KF, Wong MC. Experience of stigma among Chinese mental health patients in Hong Kong. Psychiatr. Bull 2004;28:451–454. 10.1177/0020764018815926
10.1177/0020764018815926 -
EU contribution to the World Mental Health Surveys Initiative (EU-WMH). Social inequalities in mental health and in unmet need for mental health care in Europe. 2011. Available at: http://www.eu-wmh.org/PDF/INEQUALITIES_Report.pdf Accessed April 2018.
-
Mackenzie CS, Gekoski WL, Knox VJ. Age, gender, and the underutilization of mental health services: The influence of help-seeking attitudes. Aging Ment Health 2007;10(6):574-582. 10.1080/13607860600641200
10.1080/13607860600641200 -
Oliver MI, Pearson N, Coe N, Gunnell D. Help-seeking behaviour in men and women with common mental health problems: cross-sectional study. Br J Psych 2005;186(4):297-301. 10.1192/bjp.186.4.297
10.1192/bjp.186.4.297 -
Ojeda VD, Bergstresser SM. Gender, Race-Ethnicity, and Psychosocial Barriers to Mental Health Care: An Examination of Perceptions and Attitudes among Adults Reporting Unmet Need. J. Health Soc Behav 2008;49:317–334. 10.1177/002214650804900306
10.1177/002214650804900306 -
Joung IM, van der Meer JB, Mackenbach JP. Marital status and health care utilization. Int J Epidemiol 1995;24:569-75. 10.1093/ije/24.3.569
10.1093/ije/24.3.569 -
Stimpson JP, Wilson FA. Cholesterol screening by marital status and sex in the United States. Prev Chronic Dis 2009;6(2):A55. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2687861/pdf/PCD62A55.pdf
-
Coyle C, Kinnear H, Rosato M, Mairs A, Hall C, O’Reilly D. Do women who intermittently attend breast screening differ who attend every invitation and those who never attend? J Med Screen. 2014;21(2):98-103. 10.1177%2F0969141314533677
10.1177%2F0969141314533677 -
Perry MJ. The relationship between social class and mental disorder. J Prim Prev 1996;17:17-30. 10.1007/BF02262736
10.1007/BF02262736 -
Holman D. Exploring the relationship between social class, mental illness stigma and mental health literacy using British national survey data. Health 2015;19(4):413–429. 10.1177/1363459314554316
10.1177/1363459314554316 -
Cookson R, Propper C, Asaria M, Raine R. Socio-economic inequalities in health care in England. Fisc Stud 2016;37(3-4):371-403. 10.1111/j.1475-5890.2016.12109
10.1111/j.1475-5890.2016.12109 -
Muldoon L, Rayner J, Dahrouge S. Patient poverty and workload in primary care: study of prescription drug benefit recipients in community health centres. Can Fam Physician 2013;59(4):384-90. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3625088/pdf/0590384.pdf
-
King’s Fund. Access to health care and minority ethnic groups. 2006. Available at: https://www.kingsfund.org.uk/sites/default/files/field/field_publication_file/access-to-health-care-minority-ethnic-groups-briefing-kings-fund-february-2006.pdf Accessed April 2018.
-
Brisset C, Leanza Y, Rosenberg E, Vissandjee B, Kirmayer LJ, Muckle G, Xenocostas S, Laforce H. Language barriers in mental health care: A survey of primary care practitioners. J Immigr Minor Health 2014;16:1238–1246. 10.1007/s10903-013-9971-9
10.1007/s10903-013-9971-9 -
Thorens G, Gex-Fabry M, Zullino DF, Eytan A. Attitudes toward psychopharmacology among hospitalized patients from diverse ethno-cultural backgrounds. BMC Psychiatry 2008;8:55. 10.1186%2F1471-244X-8-55
10.1186%2F1471-244X-8-55 -
Rhee TG, Evans RL, McAlpine DD, Johnson P. Racial/Ethnic differences in the use of complementary and alternative medicine in US adults with moderate mental distress: Results from the 2012 National Health Interview Survey. J Prim Care Community Health 2017;8(2):43-45. 10.1177/2150131916671229
10.1177/2150131916671229 -
Judd F, Jackson H, Komiti A, Murray G, Fraser C, Grieve A, Gomez R. Help-seeking by rural residents for mental health problems: the importance of agrarian values. Aust N Z J Psychiatry. 2009;40:769-776. 10.1080/j.1440-1614.2006.01882.x
10.1080/j.1440-1614.2006.01882.x -
Jones-Rounds ML, Evans GW, Braubach M. The interactive effects of housing and neighbourhood quality on psychological well-being. J Epidemiol Community Health 2014;68(2):171–175. 10.1136/jech-2013-202431
10.1136/jech-2013-202431 -
Betts J, Thompson J. Mental Health in Northern Ireland: Overview, Strategies, Policies, Care Pathways, CAMHS and Barriers to Accessing Services. Available at: http://www.niassembly.gov.uk/globalassets/documents/raise/publications/2016-2021/2017/health/0817.pdf . Accessed March 2019.
-
Smith M. Anti-stigma campaigns: time to change. Br J Psychiatry 2013;202:49-50. 10.1192/bjp.bp.113.126813
10.1192/bjp.bp.113.126813 -
Evans-Lacko S, Corker E, Williams P, Henderson C,Thornicroft G. Effect of Time to Change anti-stigma campaign on trends in mental-illness-related public stigma among the English population in 2003-13: an analysis of survey data. Lancet Psychiatry 2014;1(2):121-128. 10.1016/S2215-0366(14)70243-3
10.1016/S2215-0366(14)70243-3 -
Sartorius N. Stigma: what can psychiatrists do about it? Lancet 1998;352:1058-9. 10.1016/S0140-6736(98)08008-8
10.1016/S0140-6736(98)08008-8 -
Henderson C, Robinson E, Evans-Lacko S, Thornicroft G. Relationships between anti-stigma programme awareness, disclosure comfort and intended help-seeking regarding a mental health problem. Br J Psychiatry 2017;211(5):316-322. 10.1192/bjp.bp.116.195867
10.1192/bjp.bp.116.195867 -
Beyondblue. Men’s program. 2010. Available at: http://www.beyondblue.org.au/about-us/programs/mens-program . Accessed March 2019
-
Spur Projects. A bold new approach to suicide prevention. 2014. Available at: http://www.spurprojects.com.au/ Accessed March 2019