The results of the systematic review of the literature on components of institutional care have been published elsewhere [10 (link)]. Eight domains of care were identified: living conditions; interventions for schizophrenia; physical health; restraint and seclusion; staff training and support; therapeutic relationship; autonomy and service user involvement; and clinical governance. The results of the Delphi exercise have also been previously reported [11 (link)] and eleven domains of care were identified: social policy and human rights; social inclusion; self management and autonomy; therapeutic interventions; governance; staffing; staff attitudes; therapeutic environment; post-discharge care; carers; physical health care [11 (link)]. Collation of each country's care standards by HK and TT identified seven domains: living environment; mental and physical health; therapeutic relationship; service users' rights and autonomy; service user involvement; staff training and support; clinical governance. The project steering committee (PSC) reviewed these findings and agreed on nine domains for inclusion in the toolkit (Living Environment; Treatments and Interventions including restraint and seclusion; Therapeutic Environment; Self-management and Autonomy; Social Policy, Citizenship and Advocacy; Clinical Governance; Social Interface; Human Rights; and Recovery Based Practice). These were further reviewed and agreed by an international panel of experts in social care, mental health rehabilitation, recovery based practice, service user experience, disability rights, international mental health law, international mental health policy and care standard setting.
Toolkit items for assessment of these domains were generated by the UK centres. The toolkit was designed to be completed by the manager of the facility since we were aware, due to the complexity of their mental health problems, that only some service users would have the capacity to complete such a measure. However, service users' experiences of care were assessed in a later Phase of the project to investigate the association between unit manager toolkit ratings and service user reports. Where possible, toolkit items were worded to avoid revealing which answer would lead to a higher quality rating. A mix of question formats was used (Likert scales, ordered categories, quantitative responses, binary responses, lists of yes/no's summed to create quantitative responses, and vignettes that asked the respondent to generate answers which were "checklisted" by the researcher and summed to give a quantitative response). The varied format of questions aimed to increase the accuracy of responses by avoiding a response set and make the toolkit more interesting to complete. The draft toolkit was reviewed by the PSC and the international expert panel and further questions were added if there was evidence for their inclusion from Phase 1 or if they appeared highly relevant across countries.
The toolkit was translated in each country and back translated by someone independent of the project. Back translations were reviewed at the lead centre in the UK and amendments agreed with each country. The toolkit was piloted in each country in one or two facilities. A training session was attended by all researchers involved in data collection to ensure clarity of understanding of all items and their scoring.
Toolkit items for assessment of these domains were generated by the UK centres. The toolkit was designed to be completed by the manager of the facility since we were aware, due to the complexity of their mental health problems, that only some service users would have the capacity to complete such a measure. However, service users' experiences of care were assessed in a later Phase of the project to investigate the association between unit manager toolkit ratings and service user reports. Where possible, toolkit items were worded to avoid revealing which answer would lead to a higher quality rating. A mix of question formats was used (Likert scales, ordered categories, quantitative responses, binary responses, lists of yes/no's summed to create quantitative responses, and vignettes that asked the respondent to generate answers which were "checklisted" by the researcher and summed to give a quantitative response). The varied format of questions aimed to increase the accuracy of responses by avoiding a response set and make the toolkit more interesting to complete. The draft toolkit was reviewed by the PSC and the international expert panel and further questions were added if there was evidence for their inclusion from Phase 1 or if they appeared highly relevant across countries.
The toolkit was translated in each country and back translated by someone independent of the project. Back translations were reviewed at the lead centre in the UK and amendments agreed with each country. The toolkit was piloted in each country in one or two facilities. A training session was attended by all researchers involved in data collection to ensure clarity of understanding of all items and their scoring.
Free full text: Click here
