Online supplement 1 Additional methodological detail
Scoping
The initial scoping and mapping to prioritise and specify sub-questions and relevant study types was undertaken by a multidisciplinary research team which included members from psychiatry, psychology, sociology, epidemiology and health policy and those with experience of using mental health services.
Data sources
The reviews that were checked for relevant references were Corrigan (2004), Corrigan and Rüsch (2002), Gary (2005), Kushner and Sher (1991), Schomerus and Angermeyer (2008), and Thornicroft, 2008.
Selection of studies
Titles and abstracts resulting from the search were screened by one author (OS/FM/TG) and a random 10% sample was checked by another author (SC), which indicated satisfactory agreement (96%).
Data extraction, analysis and synthesis
Where more than one measure of help-seeking was used, a heuristic was applied to select the measure for analysis: if the measures differed in being attitudinal, intentional or behavioural, we selected the one most proximal to behaviour; if they differed by help-seeking from different types of provider (e.g. general practitioner or psychiatrist) we selected the measure relating to providers who care for those more severely unwell. When converting the measure of effect to a standardised effect size (Cohen’s d), where the study reported a standardised regression coefficient, this was first converted to a Pearson r according to published guidance (Peterson and Brown, 2005). If this was not possible from the available data, authors were contacted for further information. Narrative synthesis was undertaken due to substantial methodological and clinical heterogeneity between studies (Popay et al., 2006). The extracted median effect sizes were interpreted as small, medium or large in accordance with published guidance (Cohen, 1992). When studies reported results for multiple stigma measures, we selected the measure with the median effect size (Grimshaw et al., 2003), except in the subgroup analysis by stigma type where analysis was at the outcome level.
Stigma-related barriers were classified based on categorisations in previous research (Clement et al., 2012a) and grouped into Shame/embarrassment, Negative social judgement, Disclosure concerns/confidentiality, Employment-related discrimination and General stigma/other stigma barriers. The wording of the stigma-related barriers and number, or percentage, of participants reporting each barrier were extracted.
Subgroup and sensitivity analyses
The subgroup comparison on comparing studies with samples receiving care with studies with other samples was a post-hoc comparison undertaken on the suggestion of one the reviewers of the paper. Within the association studies, subgroup analyses were also undertaken by type of stigma measure and comparing attitude/intention studies vs. studies with retrospective behavioural indicators of help-seeking vs. prospective studies. The latter was unplanned and added because it was thought important to investigate whether these methodological issues impacted the results. For the association studies subgroups were compared by examination of the interpreted effect sizes; and for the barriers studies by visual inspection of data patterns. For the qualitative process studies, given the large number of subthemes identified, a pre-analysis decision was made to restrict the subgroup analysis to subthemes identified in at least 10 studies, and subgroups with at least five studies. As it was difficult to make judgements about patterns using visual inspection in what remained a large matrix of data, chi square / Fisher’s exact tests for the qualitative process studies subgroup analysis were used.
References
Clement S, Brohan E, Jeffery D, Henderson C, Hatch SL, Thornicroft G (2012b). Development and psychometric properties the Barriers to Access to Care Evaluation scale (BACE) related to people with mental ill health. BMC Psychiatry 12, 36.
Cohen J (1992). A power primer. Psychological Bulletin 112 (1), 155-159.
Corrigan P (2004). How stigma interferes with mental health care. American Psychologist 59, 614-625.
Corrigan P, Rüsch N (2002). Mental illness stereotypes and clinical care: Do people avoid treatment because of stigma? Psychiatric Rehabilitation Skills 6(3), 312-334.
Gary F (2005). Stigma: Barrier to mental health care among ethnic minorities. Issues in Mental Health Nursing 26, 979-999.
Grimshaw J, McAuley LM, Bero LA, Grilli R, Oxman AD, Ramsay C, Vale L, Zwarenstein M (2003). Systematic reviews of the effectiveness of quality improvement strategies and programmes. Quality and Safety in Health Care 12, 298-303.
Kushner M, Sher K (1991). The Relation of Treatment Fearfulness and Psychological Service Utilization: An Overview. Professional Psychology - Research & Practice 22, 196-203.
Peterson RA, Brown SP (2005). On the use of beta coefficients in meta-analysis. Journal of Applied Psychology 90, 175–181.
Popay J, Roberts H, Sowden A, Petticrew M, Arai L, Rodgers M, Britten N, Roen K, Duffy S (2006). Guidance on the conduct of narrative synthesis in systematic reviews. ESRC methods programme: University of Lancaster, UK.
Pope C, Mays NJP (2007). Synthesising qualitative and quantitative health evidence: A guide to methods. Open University Press: Maidenhead.
Schomerus G, Angermeyer M (2008). Stigma and its impact on help-seeking for mental disorders: What do we know? Epidemiologia e Psichiatria Sociale 17, 31-37.
Thornicroft G (2008). Stigma and discrimination limit access to mental health care. Epidemiologia e Psichiatria Sociale 17, 14-19.