The ability of adult mental health services to meet clients' attachment needs: the development and implementation of an assessment measure.
Isabel Goodwin

Outline
This study was carried out by Isabel Goodwin as part of the training requirements for the doctoral programme in clinical psychology based at the School of Psychology at the University of Birmingham.
The study was supervised by Professor Ray Cochrane and Dr. Oliver Mason, also from the School of Psychology, and by Dr. Guy Holmes, Clinical Psychologist, working for Shropshire's Community and Mental Health NHS Trust.

The study examined the relevance of attachment theory to the relationships between mental health services and their clients. The aim of the study was to develop a questionnaire to assess the ability of adult mental health services to meet clients' attachment needs, that is, provide them with a secure and positive relationship.The questionnaire-the Service Attachment Questionnaire (SAQ)-was developed from data collected from focus groups of adult mental health service users in Shropshire. Two trials of the questionnaire were then carried out with service users of the Trust, to examine its reliability and validity.

Summary of Findings
The themes in the data generated by the focus groups of service users corresponded closely to those in the published literature and existing instruments in the field of adult attachment. This suggests that attachment is a relevant and important aspect of the relationships people have with adult mental health services.
The results from the clinical trials showed that people were reasonably satisfied with the relationships
provided by the adult mental health services, and that they perceived the service areas in the study as providing them with positive attachment relationships. This was the case for all service areas studied, though participants receiving inpatient services had lower response scores on the SAQ than participants from other service areas.
The Service Attachment Questionnaire (SAQ) showed a high level of internal reliability (that is, the items in the questionnaire appeared to be measuring the same thing) and good test-retest reliability (that is, peoples responses were reliably similar one month later).
The validity of the SAQ was demonstrated in two ways: (i) by positive correlations between peoples scores on the questionnaire and their own ratings of how much they felt they had improved as a result of coming into contact with the service, and how helpful they felt the service had been; (ii) via a statistical calculation (called factor analysis), which suggested that the SAQ has a single underlying dimension.

Recommendations
The Service Attachment Questionnaire is a reliable and user-oriented scale, which may usefully indicate how far service users perceive a mental health service as meeting their attachment needs (that is, providing them with a secure and positive relationship). The SAQ may be used by services to examine different areas of the service-which needs they may be successful in meeting, and which needs they may be failing to meet. Such information may provide guidance for services regarding the more specific attachment needs of clients receiving different types of mental health service (for example, inpatients, or clients in the community).

Background
Attachment theory suggests that the fundamental affectionate bonds and relationships that people develop as children with their parents are similarly present in adult relationships with other adults. Attachment behaviour involves an individual trying to get close to a person or persons they feel most comfortable with, promoting a feeling of security in the individual. Attachment relationships are described as either secure (or positive), or insecure (or negative-of which there are three identified sub-types). Such behaviour is most commonly activated in adults during times of illness or stress. A lot of research has been carried out in the field of mental health and illness, looking at the different types of attachment relationships people with mental health problems have.

Treatments such as psychological therapies have also been studied, to attempt to work out how they help people to have better, more secure relationships. It is estimated that around one third of the population have insecure attachment relationships, and that this leaves people vulnerable to developing mental health problems in times of stress. However, only a few studies have looked at the attachment relationships clients have with mental health service staff. Although many questionnaires and other measures exist, they tend to assess one-to-one relationships, with partners, friends or other family members, rather than relationships outside this group. This study looked at the ability of different parts of mental health services to meet the attachment relationship needs of clients, by developing a questionnaire based on the experiences of service users' with mental health services.

Further Details of Research

The Participants
Two focus groups of people using local mental health services were held. Each group was made up of three service users, a group facilitator and a co-facilitator. The groups were asked to discuss their experiences-both the helpful and unhelpful aspects of the relationships they had within the mental health services. The information from the focus groups (the data) was subjected to a qualitative analysis to extract themes and potential items for inclusion in the questionnaire. 154 people currently using adult mental health services participated in the first trial of the questionnaire. The participants were from inpatient (acute or rehabilitation) services, psychological therapy services, community mental health team services and day centre services. Of these, 54 participants also completed the second trial of the questionniare.

The Assessments and Measures
The Service Attachment Questionnaire (SAQ) was developed directly from the focus group data. The qualitative analysis produced six key themes, which formed the basis of the 6 subscale, 25 item questionnaire. The responses to each item were on a Likert scale of (1) Not at all; (2) Sometimes; (3) Quite Often, and (4) Always. The subscales were as follows: · Being attended to and listened to · Being there - consistency and continuity · Being given enough time - ending and leaving · Safe environment · Relationships which enable helpful talking · Human contact and comfort.

Items in the scale included, "I have somebody who listens attentively to me", and "I feel under pressure to get better and be discharged". Participants were also asked to give details of which service they were currently receiving, and the length and extent of current contact with that service. Self-ratings of how much they felt they had improved as a result of coming into contact with the service and how helpful they felt the service had been were also included. Both these items were assessed on a 5-point scale.

The Outcomes
The results of the qualitative part of the study indicate that attachment relationships are a relevant and important aspect of the relationships clients have with mental health services. The SAQ subsequently developed in the study proved a reliable and usable self-report measure that appears to assess the ability of mental health services to meet the attachment relationship needs of clients involved with different parts of the service.
The SAQ also showed acceptable temporal stability over a one month period. The actual service used in this study proved able to meet clients attachment needs with moderate success. That is, participants were reasonably satisfied with the relationships they were provided with by the service (the average score being 3 out of a maximum of 4 on the response scale). There were also positive correlations between participants' scores on the SAQ and their ratings of how much they felt they had improved and how helpful they felt the service had been (i.e. the higher a person's score, the higher their ratings of improvement and perception of helpfulness of the service).

The lowest correlation was between the subscale 'Being given enough time - ending and leaving' and participants' perception of how helpful the service had been. This may have been a reflection of the feared loss of the attachment figure which is likely to result in feelings of insecurity. The results also showed that average questionnaire scores for participants receiving inpatient services were significantly lower than for participants in other service areas, but that there were no corresponding differences between inpatient and other services in terms of how much participants felt they had improved and how helpful they felt the service had been. This finding may indicate that the attachment needs of these clients are more complex and specific than clients in other areas, and may not be elicited by a self-report measure.

Remaining Questions
As the questionnaire is one of the first of its kind, further evidence of its reliability and validity is important. Shropshire is a large, inland, rural county, and may differ in both client needs and service organisation from other areas of the country. Replication of the study in other locations and in other service areas (such as older clients or assertive outreach teams) may therefore be appropriate. Also further work with service users to clarify and confirm important aspects of the relationships provided in different service areas may afford extension or refinement of the SAQ.

Acknowledgements
Grateful thanks go to the focus group members and the participants who completed the Service Attachment Questionnaire. Also, to the management and staff of Shropshire's Community and Mental Health NHS Trust, particularly the Department of Psychological Therapies.

About the Study
Funding for the study was provided by the Psychological Therapies Department of Shropshire's Community and Mental Health NHS Trust, and the University of Birmingham. Ethical permission for the study was obtained from the Shropshire Ethical Review Committee, William Farr House, Shrewsbury.

Further Information
The Service Attachment Questionnaire (SAQ) is available from the author c/o the Psychology Consultancy Service, Shropshire's Community and Mental Health NHS Trust, Chaddeslode House, 130 Abbey Foregate, Shrewsbury. SY2 6AX.

A copy of the full-length research report is also available from the same address, and has also been submitted for publication to the British Journal of Medical Psychology. Any comments and feedback from interested persons are welcomed.