Research: Rural mental health
Working on a grant from the National Mental Health Partnership (now called the Mental Health Network), we were given the task of investigating rural mental health services in England and Wales. This included provision of those services, as well as access to them (perceived and actual). We also examined specific aspects of rural mental illness, including suicide in those engaged in rural occupations, and piloted novel methods of mental health therapies in rural communities. This work was conducted as part of a two-year research programme, supported by Hampshire Partnership NHS Trust, under the supervision of Dr Alain Gregoire. A summary of that research is explored here. The full report, as provided to the National Mental Health Partnership is available on request. The programme was divided into three strands: Access to services; models of care; and the rural population.
Access to services
Survey of NHS Trusts
Questionnaires were sent to an executive at 35 NHS Mental Health Trust; 26 were returned. Information was requested regarding the ‘rurality’ of the Trust (based on proportion of population served by the Trust, who lived in rural locations. Information was also sought on the level of service provided in respect of Community Mental Health Teams (CMHTs), Child and Adolescent Mental Health Services (CAMHS), gender segregated inpatient facilities, Crisis Intervention, Assertive Outreach, and Early Intervention. The Trust executives were also asked to comment on accessibility to those services. The responses suggested that rural populations appear to have consistently poorer access to mental health services than do urban ones. Even where Trusts provided service to rural areas, access to those services was perceived as incomplete. These data, albeit from a limited source, suggest that access to mental health services in rural locations is more restricted than it is in urban areas.
Survey of 'NGO' perceptions
We sought the views of a wide sample of voluntary and non-government organisations (NGOs) with rural responsibilities. We examined their perceptions of mental health services, particularly in respect of service access and provision for their members. Questions focused on factors relating to access (to inpatient services, specialist teams or clinics, alternative methods of care, self-help networks, drop-in centres, etc), and availability (of transport, home care teams, support from urban teams, etc). Other factors related to feelings of isolation, whether local GPs were equipped to treat mental illness, and satisfaction that mental health teams understood rural needs. Sixty questionnaires were analysed. Overall, perceptions are generally more negative for those NGOs and voluntary organisations that provide services in rural settings. This was especially seen in more remote areas, and in respect of availability of specialist staff visits, frequency of home visits, access to inpatient facilities, access to alternatives care options, and the perception that mental health workers understand the specific needs of rural clients. This provides yet further evidence that access to mental health services is perceived to be poorer in rural settings. Furthermore, issues concerning the quality of service tend to be more poorly perceived in rural contexts.
Models of care
Case Management in Rural Primary Care
One of the most consistent findings in our rural mental health research is getting access to services, particularly in remote areas that are poorly served with public transport. As we saw, local GPs are not always best equipped to deal with mental illness. One solution is take services to the patient. This is all very well in urban settings, where CMHT teams can visit several patients in one day. In rural areas, patients are more sparsely located, so home teams have to spend much more time travelling. This is costly and inefficient. A potential answer lies in the way in which care is delivered. There is growing use of evidence in the benefits of treating patients by telephone. The calls can be made by qualified staff, specifically trained in this method of treatment. We undertook a comparison of case management against ‘treatment as usual’ for depressed patients in primary care. Patients were recruited from New Forest GP practices, and randomised to the one of the study conditions. Patients were excluded if they showed psychotic symptoms, or if they appeared to be at risk for suicide. Any patient whose illness was severe enough to warrant referral to secondary care was also excluded. Case management (CM) was provided by a Community Psychiatric Nurse (CPN), and involved weekly telephone calls (lasting about ten-minutes). ‘Treatment as usual’ (TAU) was standard care provided by the patient’s GP. Outcomes were measured using questionnaires that examined psychiatric symptoms, quality of life, and perceived satisfaction with treatment. Satisfaction focused on four domains: satisfaction with information (explanation of treatment plan, information about that treatment, information about illness); satisfaction with the process (convenience of appointments, amount of contact with professionals); satisfaction with support (health professional’s understanding of patient’s difficulties, support from those professionals, opportunities to communicate feelings, quality of contact with professionals); and satisfaction with outcome (effectiveness of treatment, overall satisfaction with care, confidence to use service again). Symptom improvement was measured by the Hospital Anxiety and Depression Scale (HADS; Zigmond & Snaith, 1983). The patient’s perceptions regarding quality of life were measured by the SF-36 (Ware, 1992). Thirty-nine patients were randomised into each group. Measurements were taken at baseline, prior to treatment, and at weeks 6 and 12 post-treatment.
Our results showed encouraging data. Patient satisfaction with treatment was significantly greater overall for CM group than the TAU group; this was equally apparent at weeks 6 and 12. Satisfaction was better for the CM group across all domains, and within each of the domains. CM patients were more likely to complete treatment than TAU patients. Improvements were seen in reported depression for both groups, but only the CM group showed sufficient improvement that could be considered to be clinically relevant. Improvements in reported anxiety were only seen in the CM group. Overall, case managed treatment was associated with more positive outcomes, greater satisfaction and generally better illness improvement perceptions than those receiving treatment as usual. This study was undertaken in rural primary care and may provide a useful model for treatment options in rural settings. We are planning to explore the data further for multivariate analyses, with a view to submitting the paper for publication.
In a similar vein to the case managed study, we piloted a study using telephone conference facilities on a group of women with postnatal depression (PND) who lived in rural areas. We worked in collaboration with Rural Minds (a sub-group of Mind, the mental charity). Teleconference care was designed to employ structured one-hour telephone calls, facilitated by a health visitor and CPN, delivered to several PND mothers at the same time. The programme of sessions was based on Cognitive Behavioural Therapy (CBT) principles. This has a number of benefits, particularly in rural settings: mental health professionals reduce the need to travel to their clients, and many patients can be treated simultaneously, providing greater efficiency. There is also evidence that suggests group-therapy is very beneficial in the treatment of mental illness. Eighteen PND mothers were recruited from a rural secondary care setting to the telephone conference support group; this was conducted over 8 sessions. Outcome measures were examined using the Edinburgh Postnatal Depression Scale (EPDS; Cox et al., 1987) and a satisfaction questionnaire. This was examined at weeks 4 and 8 of the programme. PND mothers in the programme were also compared to a waiting-list control group.
Our initial results were very positive. By the end of the sessions, PND mothers in the teleconference programme showed significant improvements in mood, as shown by the EPDS, progressing to sub-clinical ratings. PND mothers on the waiting list consistently showed EPDS scores indicative of current depression. PND mothers in the teleconference programme also reported very favourable satisfaction. Similar to RCT of case managed depression, this method of delivering psychological treatment by telephone may represent a cost effective alternative for treating mental illness in rural settings. Rural Minds have since applied these methods in rural settings across England and Wales.
The rural population
Suicide among Agricultural Workers
Farmers account for the largest numbers of suicides amongst any single occupational group in the UK, and it is the second most important cause of death in young farmers after accidents. The methods typically used by farmers have changed in recent years. For a long a time firearms were the most common; since The Firearm (Amendment) Act (1988) hanging is now the most likely method. Farmers’ methods tend to be more ‘lethal’ and more impulsive, often without any form of prior communication. Farmer suicides are typically associated with their work, financial troubles, legal problems, physical health and relationships. Evidence suggests that the most common factor is mental illness, usually depression. This is despite rates of mental illness being lower in farmers than in the general population. Farmers tend to internalise their problems and avoid seeking help. The economic climate for farmers has been less than favourable for some time, not helped by threats from foot-and-mouth, bovine tuberculosis, and Bovine Spongiform Encephalopathy.
Although some of these findings may be generalisable to others in rural occupations, or to those living in rural areas, very little has been written about that; we sought to address that gap in evidence. Overall, rates of suicide in England and Wales appear to be lower in rural areas, compared to urban ones. However, this pattern may be changing; some evidence suggests increasing rates. Moreover, while there is evidence that the usual risk factors for suicide are higher in urban areas, suicide may be a more likely endpoint in rural groups. We undertook an investigation of Coroners’ inquest files to examine these factors. We focused on data from Coroners across several districts in southern England (where there was an open verdict or one of suicide), spanning 1984 to 2005. We focused on those victims living in a ‘rural location’. We also on ‘victims’ who were engaged in a rural occupation (farmer, farm worker, gardens and horticulture, country vet, animal workers, and other skilled rural occupations, such as thatching, etc.); we called these ‘rurally employed’. Coroners files were scrutinised, from which we elicited data on a wide range of predisposing factors (relationships, type of home, working status, occupation, suicide method, involvement of alcohol, where found, fatal substances, current and historical illness (physical and mental), communication of intent, history of previous attempts or self-harm, recent contact with health professionals (GP and secondary care), psychiatric diagnoses, and psychiatric inpatient history.
Rural population results
Rurality of residence
Rural victims significantly more likely to be older, married/cohabiting, and to have had recent contact with psychiatric services than urban victims. Rural victims were less likely to be single/divorced, unemployed, and have a history of previous deliberate self harm than urban victims. Rural methods of suicide were more likely to involve firearms or car exhaust; urban methods were more likely to include overdose, jumping from heights, or a deliberate vehicle accident. Rural victims were less likely to communicate intent. Rural victims were more likely to be found within the grounds of their home, while those from urban areas were more likely to be found in their homes. Where suicides were the result of poisoning (drugs or chemicals) rural victims were more likely to use a substance available at work; urban victims were more likely to use a prescription drug.
Rurality of occupation
Rurally employed victims were significantly more likely to be male, engaged in full-time work, have a history of mental illness, to be currently receiving psychiatric medication, and to have had recent contact with GP or psychiatrist than non-rurally employed victims. Rurally employed victims were significantly less likely to have history of physical illness, have a history of previous history of attempted suicide, and less likely to communicate intent. Mental illness for rurally employed victims was more likely to involve psychoses; for non-rurally employed victims depression was more likely. Rurally employed victims were more likely to use firearms or hanging to complete suicide; their non-rural counterparts were more likely to use overdoses, car exhaust, jumping from heights, or deliberate vehicle accidents. Rurally employed were more likely to be found within the grounds of their home (or at work), while non-rurally employed victims were more likely to be found in their homes, or in a public place. Where suicides were the result of poisoning, rurally employed victims were more likely to use weed killer or animal drugs (probably obtained from work); non-rurally employed victims were more likely to use a prescription drug.
Further analyses, especially data relating more specifically to farmers, can be found in the full report. The implications of methods and precipitating factors are also discussed. Overall, it would appear that rural suicide victims are from more stable backgrounds than urban victims. These factors have not been explored in previous suicide studies, but appear to reflect demographics that are generally associated with rural life. Rural suicides tend to be ‘more effective’, reflecting intent and access to more lethal methods, and yet rural suicides are less likely to be communicated beforehand. These suicides are also more likely to be in a place where the victim is less likely to be found.
Adult Mental Health Service Mapping & Minimum Data Set
The Mental Health Service Mapping (MHSM) and Minimum Data Set (MDS) is a collaborative project between the Department of Health (DH), National Institute for Mental Health in England (NIMHE), and University of Durham/Public Health Observatory. We have examined a range of variables from both data sets in relation to rurality of PCTs and Mental Health Trusts. To achieve this we have applied an established definition of rurality to the census demographic data for all PCTs and Trusts.
In our full report we review previous studies that examine perceived inequality in mental health services in rural settings. However, very little work has been undertaken to formally examine this, particularly in the UK. American studies have confirmed that rural access to psychiatrists, psychologists, psychiatric nurses and social workers is very poor. Research in the UK has explored potential contributing factors to perceived (or actual) inequalities, including poverty, social isolation, cultural differences, and social stigma. One possible reason for the under-resourcing of mental health services to rural areas might relate to ‘known’ prevalence rates. Evidence suggests that urban prevalence of mental illness exceeded rural by ratio of (roughly) 3 to 2. Such differences appear to remain even after accounting for social and demographic factors, such as age, unemployment, home ownership, marital status, and social support. Rates of consultation for mental illness have been found to be fewer in rural areas than urban ones for males. However, this may reflect geographic isolation and social stigma as much as true illness prevalence.
Evidence that mental illness rates are lower in rural settings may seem compelling, but there are several problems. The reported prevalence may not represent true rates. Also, published studies have used very inconsistent measures of ‘rurality’. Furthermore, some of the predisposing factors for mental illness are often misused in rural research. Poverty is a good example; farming communities have seen some of the worst economic downturns of any in the UK. It has been suggested that measures of social deprivation fail to address rural disadvantage. While employment, homeownership, and access to a car may be increased in rural areas, rural residents may feel more isolated than their urban counterparts when deprived of such things; car ownership is perhaps the most pertinent. We sought to examine mental health services in England and Wales more comprehensively than previous studies, and use a specific measure of rurality.
Our definition of rurality was based on the methods used by the Countryside Agency to describe wards, which focuses on factors such as population density, economic activity and employment of the population, use of public transport, and ethnicity. The Minimum Data Set (MDS) and Mental Health Service Mapping (MHSM) are reported at several levels, but we focused on Mental Health Trusts (MHT) and Primary Care Trusts (PCT). The Countryside Agency data for wards was mapped onto information about MHT and PCT ward boundaries to define rural instances of these healthcare divisions. Complete details of our calculation of rural vs. urban MHT and PCT can be found in our full report. In general, MHT data represents service provision; whereas PCT data represent service receipt. Using our calculations, 20 of the 83 MHTs (24%) and 77 of the 303 PCTs (25%) were classified as ‘rural’. Using similar methods, 137 PCTs (45%) and 31 MHTs (37%) were determined to represent ‘inner city’ areas.
The Mental Health Service Mapping (MHSM) reports mental health service provision according to health service organisation areas (such as MHT and PCT). It focuses on types of service, reporting the number of teams, caseloads, staffing levels, bed numbers, and other variables such as the number of drop-in centres available. Staffing levels are described in terms of whole time equivalents (WTE), rather than the actual number of personnel. The Minimum Data Set (MDS) focuses on patient access to services. Data represent the ‘spell of care’ for each patient in health care provider area (such as MHT and PCT). The data are reported in two formats: patient-centred, which focuses on the amount of care a patient receives (number of days in hospital, number of consultant contacts, etc.); and care-centred, which focuses on the number of services the MHT has provided (total beds days, total consultant contacts, etc.). We have reported the data at the care-centred level. The MDS reports data on hospital bed usage (including type: intensive care, medium secure, standard), admissions, discharges, readmissions, length of stay, outpatient attendances, day care attendance, sectioning, and extent of ‘non-NHS care’. Within these, further details are provided: care complexity, care team type, clinician contact, age of patient, diagnosis, ethnicity, gender, and marital status.
MHSM and MDS results
MHSM by MHT
There were significantly fewer doctors, nurses, social workers and ‘other therapists’ in rural MHTs, compared to non-rural. Meanwhile, there were significantly more doctors, nurses, psychologists, social workers, ‘other therapists’ and volunteer workers in inner city MHTs, compared to non-inner city. This was reflected across ‘all services’ and within specific services, such as assertive outreach, community mental health teams (CMHT), community rehabilitation, and psychotherapy.
MHSM by PCT
There were significantly fewer social workers in rural PCTs, compared to non-rural, and significantly more psychologists, social workers, social care and development (SCD) workers, and ‘other therapists’ in inner city PCTs, compared to non-inner city. This was reflected across ‘all services’ and within specific services, such as assertive outreach, CMHTs, crisis resolution, early intervention, community rehabilitation, day care, and psychotherapy.
MDS by MHT
There were fewer ‘bed days’ (per 100k population) supplied by rural MHTs, compared to non-rural, and a greater number provided by inner city MHTs, compared to non-inner city. This was particularly the case with ‘intensive’ beds. In non-rural and inner city MHTs the supply of psychiatric beds was more likely to be intensive or medium secure; in rural MHTs psychiatric beds were likely to be ‘standard’. Compared to non-rural and inner city MHTs, admissions to rural MHTs were less likely to involve substance abuse. Outside of inner cities, admissions were more likely to relate to mood disorders. There was no difference in lengths of stay between any of the groups examined. Total contacts (per 100k population) by mental health professionals were lower in rural (vs. non-rural) MHTs, although social worker contacts were significantly higher in rural MHTs. Rural MHTs provided fewer health professional contacts in key services, such as old age psychiatry, substance abuse, assertive outreach, crisis resolution, early psychosis intervention, and general adult.
MDS by PCT: There were fewer ‘bed days’ (per 100k population) supplied by rural PCTs, compared to non-rural, especially in respect of standard and intensive beds. There were more ‘bed days’ (per 100k population) supplied by inner city PCTs (for all beds). In non-rural and inner city PCTs the supply of psychiatric beds was more likely to be intensive or medium secure; in rural PCTs psychiatric beds were likely to be ‘standard’. Compared to non-rural and inner city MHTs, admissions to rural MHTs were less likely to involve substance abuse, alcohol abuse, and psychoses. Outside of inner cities, admissions were more likely to relate to mood disorders. Compared to non-rural and inner city PCTs, significantly fewer patients from rural PCTs were contacted within 90 days of discharge, and had significantly fewer outpatient appointments. Total contacts (per 100k population) by mental health professionals were lower in rural (vs. non-rural) PCTs. Rural PCTs provided fewer health professional contacts from clinical psychologists, CPNs, physiotherapists, occupational therapists and psychotherapists. Patients from rural PCTs received fewer key services, such as crisis resolution, old age psychiatry teams, assertive outreach, and substance abuse teams.