Changes in attitudes toward seeking mental health services: A 40-year cross-temporal meta-analysis

Highlights

• We examined changes in attitudes toward seeking mental health services over time.

• Our cross-temporal meta-analysis included data from 6518 students over 40 years.

• Helpseeking attitudes have become reliably more negative from 1968 to 2008.

• This finding may reflect changing attitudes toward specialty mental health care.

Abstract

Although rates of treatment seeking for mental health problems are increasing, this increase is driven primarily by antidepressant medication use, and a majority of individuals with mental health problems remain untreated. Helpseeking attitudes are thought to be a key barrier to mental health service use, although little is known about whether such attitudes have changed over time. Research on this topic is mixed with respect to whether helpseeking attitudes have become more or less positive. The aim of the current study was to help clarify this issue using a cross-temporal meta-analysis of scores on Fischer and Turner's (1970) helpseeking attitude measure among university students (N = 6796) from 1968 to 2008. Results indicated that attitudes have become increasingly negative over time, r(44) = − 0.53, p < 0.01, with even stronger negative results when the data are weighted (w) for sample size and study variance, r(44) = − 0.63, p < .001. This disconcerting finding may reflect the greater emphasis of Fischer and Turner's scale toward helpseeking for psychotherapy. Such attitudes may be increasingly negative as a result of the unintended negative effects of efforts in recent decades to reduce stigma and market biological therapies by medicalizing mental health problems.

Introduction

Overall rates of treatment for mental disorders in developed nations are increasing over time. Trends in outpatient treatment for depression in the USA from the Medical Expenditure Surveys indicate that in 1987 less than one person (.73) per 100 received treatment and this increased to 2.33 in 1997 (Olfson et al., 2002), 2.37 in 1998 and 2.88 in 2007 (Marcus & Olfson, 2010). This evidence of increasing mental health service use, however, masks different trajectories in the use of pharmacotherapy and psychotherapy. Specifically, antidepressant medication use has increased sharply since the late 1980s (Olfson et al., 2002, Pratt et al., 2011) although this trend appears to be stabilizing since the late 1990s (Marcus & Olfson, 2010). In contrast, the proportion of people receiving outpatient psychotherapy for depression in the Medical Expenditure Surveys declined by 28% over 20 years. Among people treated for depression, 71% received psychotherapy in 1987, 60.2% in 1997 (Olfson et al., 2002), 53.6% in 1998 and 43.1% in 2007 (Marcus & Olfson, 2010).

Similar trends have been reported for the treatment of mental disorders in general via comparisons of the US National Comorbidity Survey (NCS), collected between 1990 and 1992, and its replication (NCS-R), collected between 2001 and 2003 (Kessler et al., 2005). Among individuals with a mood, anxiety, or substance disorder, 20.3% received treatment in the early 1990s whereas 32.9% received treatment in the early 2000s. By far the greatest increase in service use in these surveys occurred in the general medical sector, where treatment rates were 2.59 times as high in 2001–2003 as in 1990–1992. High rates of consultations from general practitioners (53%) have also been reported in Australian samples of young people (e.g., Reavley, Yap, Wright, & Jorm, 2011). The increase in rates of treatment over time has mostly been attributed to the expansion of psychotropic medications provided by the general medical sector.

Unfortunately, despite evidence of increasing rates of pharmacological treatment for mental health problems, the majority of mental health services do not meet evidence-based guidelines, especially in the general medical sector. In the NCS minimally adequate evidence-based treatment was defined as pharmacotherapy that: lasted ≥ 2 months, was appropriate for the focal disorder, and included > 4 visits to any type of physician. For psychotherapy, minimally adequate evidence-based treatment was defined as: ≥ 8 visits with any health care or human services professional lasting an average of ≥ 30 minutes. Of the treated patients in this survey with disorders, 48.3% received treatment that met minimal evidence-based standards from mental health professionals and only 12.7% received minimally adequate treatment from general medical providers (Wang et al., 2005). The authors concluded that despite increases in treatment rates over time, there remain serious problems with mental health care in the United States with the use of mental health services remaining disturbingly low, most patients not receiving any care for mental disorders, and only one third of those who do seek professional help receiving treatment that meets minimal standards of adequacy.

An extensive body of research has attempted to better understand barriers to mental health care. These can broadly be categorized into knowledge based, structural, and attitudinal¹ barriers (Thompson, Hunt, & Issakidis, 2004). Examples of knowledge based barriers include not recognizing symptoms of mental disorders and not knowing where to get help. Structural barriers include financial cost, difficulty with transport and lack of available services. Attitudinal barriers include stigma related concerns, fears or embarrassment about revealing personal details, beliefs that one should handle problems by oneself, and beliefs about the probability that a service will help resolve a problem. Typically when asked, individuals describe a range of barriers to seeking help with varying degrees of influence on their helpseeking (e.g., Wells, Robins, Bushnell, Jarosz, & Oakley-Browne, 1994).

Attitudinal barriers have arguably been most consistently related to intentions to seek help and actual service utilization (e.g., Gulliver et al., 2010, Jagdeo et al., 2009, Mojtabai et al., 2002, Pescosolido et al., 2010, Wells et al., 1994). Analysis of the NCS found that among those with a mental disorder and perceived need for help, the major determinants of helpseeking were attitudinal and sociodemographic (Mojtabai et al., 2002). Negative attitudes toward seeking help for mental health problems are prevalent. A comparison of Canadian and US population surveys found that 18% and 24% had negative attitudes, respectively. Rates of negative attitudes were significantly higher among younger adults aged 15 to 24, with 25% of younger Canadians and 35% of younger Americans holding negative attitudes toward seeking mental health services (Jagdeo et al., 2009).

Some of the most frequently identified attitudinal barriers to seeking mental health care include people's wishes to handle problems on their own (Gulliver et al., 2010, Rickwood et al., 2007, Wells et al., 1994, Wetherell et al., 2004, Wilson and Deane, 2012), thinking that the problem will go away by itself (Sareen et al., 2007, Thompson et al., 2004), doubts about the perceived benefits of helpseeking (e.g., Rickwood et al., 2007, Rughani et al., 2011, ten Have et al., 2010) and concerns about stigma associated with seeking mental health services (Gulliver et al., 2010, Jorm et al., 2007, Pescosolido et al., 2010).

The overwhelming evidence pointing to attitudes as a significant barrier to mental health services has resulted in extensive efforts over the past two decades from clinicians, policy makers, and researchers to reduce negative attitudes, and in particular stigma, toward mental illness. These stigma reduction efforts in the US and internationally have, in part, attempted to medicalize mental illness by focusing on its neurobiological etiology (Corrigan and Watson, 2004, Pescosolido et al., 2010, Phelan, 2005). These efforts have also coincided with aggressive direct-to-consumer advertising of psychotropic medications that further emphasized that mental disorders are biological diseases in need of biological treatments (Rosenthal, Berndt, Donohue, Frank, & Epstein, 2002). Unfortunately, although these attempts to educate the public and improve mental health systems have increased mental health literacy, including the endorsement of neurobiological causes and treatments of mental disorders, they have not resulted in reductions in public stigma toward people with mental illness. In fact, studies in the US (Pescosolido et al., 2010), Germany (Angermeyer, Holzinger, & Matschinger, 2009), Turkey (Bag, Yilmaz, & Kirpinar, 2006), Germany, Russia, and Mongolia (Dietrich, Beck, Bujantugs, Kenzine, Matschinger, & Angermeyer, 2004) have reported that endorsements of neurobiological causes of mental illness are associated with increases in the desire for social distance from those with such problems. An empirically supported possible explanation for these findings (Phelan, 2005) involves perceptions of mental illnesses as having genetic causes, resulting in them being seen as more fixed, unchangeable, serious, and persistent. As a result, people with these disorders are seen as fundamentally different from those without them.

Despite evidence that attitudes toward people with mental illness appear to be either stable or perhaps worsening, less is known about changes in attitudes toward seeking mental health services over time. Currin, Hayslip, and Temple (2011) surveyed perceptions of mental health and mental health services among older adults in 1977 (N = 90), 1991 (N = 101), and 2000 (N = 99), and among younger adults in 1991 (N = 131) and 2000 (N = 147). The authors created their own measures and administered them at all three time points, and also used Fischer and Turner's (1970) Attitudes Toward Seeking Professional Psychological Help Scale (ATSPPHS) with the 1991 and 2001 samples. This study found increasingly positive perceptions among the older sample from 1977 to 1991, whereas both the older and younger samples demonstrated significantly more negative helpseeking attitudes from 1991 to 2001 using the reliable and valid ATSPPHS.

In contrast to this evidence of worsening helpseeking attitudes in recent decades, comparison of the NCS (N = 5388) with the NCS-R (N = 4319) revealed that helpseeking attitudes improved over the decade between these cross-sectional surveys (Mojtabai, 2007). Three items assessed attitudes: willingness to go for professional help, comfort talking about personal problems, and embarrassment if friends knew about helpseeking. The percentage reporting being at least “somewhat” embarrassed reduced from 42.5% (1990–1992) to 35.5% (2001–2003). The improvement in attitudes was strongest for younger participants (18–24 years). Similarly, participants who read vignettes about individuals with mental disorders as part of the General Social Surveys (GSS) increasingly endorsed treatment from a general medical doctor, a psychiatrist, a mental hospital, and/or prescription medications from 1996 to the 2006 survey (Pescosolido et al., 2010). The 1998 and 2006 GSS also assessed opinions about the benefits and risks of psychiatric medications and willingness to take them in hypothetical situations. Results from these surveys clearly demonstrated increasingly favorable attitudes toward drug treatment over time (Mojtabai, 2009).

What might account for this conflicting evidence of worsening helpseeking attitudes in Currin et al.'s (2011) study versus improving helpseeking attitudes in the NCS and GSS? One possible explanation may have to do with their samples; Currin et al. (2011) surveyed younger and older community dwelling Texans, whereas participants from the NCS and GSS were nationally representative adults. A second possible explanation is measurement reliability. Although the NCS and GSS could be considered more reliable because their large samples reduce random measurement error, the space limitations in these population-based surveys requires very brief (between 1 and 3 items) measures of helpseeking attitudes, potentially reducing reliability. Currin et al.'s (2011) 1991 to 2000 comparison employed the most commonly used self-report measure of helpseeking attitudes in the literature, the 29-item ATSPPHS (Fischer & Turner, 1970), which has proven reliability and validity. ten Have et al. (2010) note that there is, “no universal standard for how attitudes toward mental health help-seeking can best be measured” (p. 160) but that research is needed to demonstrate the reliability and validity of brief attitude questions typically used in epidemiological population-based surveys.

A third possible explanation for the conflicting findings is that they reflected differing attitudes toward pharmacotherapy versus talk therapy. The attitude questions in the GSS referred explicitly to psychiatric medications and specific sources that typically use those pharmacological treatments (i.e., general medical doctors, psychiatrists, and mental hospitals). The attitude questions in the NCS did not refer to specific mental health providers or types of treatment, but it is likely that respondents interpreted them as referring to mental health treatment within the specialty and perhaps especially within the general medical sector, where increasing numbers of people are receiving mental health care over time (Angermeyer et al., 2009, Kessler et al., 2005). Increasing positive attitudes over time within the GSS and NCS may therefore reflect improvements in attitudes toward pharmacotherapy. In contrast, the ATSPPHS refers specifically to psychiatrists, psychologists, psychological counseling, psychotherapy, psychiatric treatment, and treatment in mental hospitals. And 14 of the 29 questions from this scale refer to talk therapy or discussing emotional difficulties. Although this scale wasn't developed as a measure of attitudes toward talk therapy, that is what it appears to predominantly measure. Increasingly negative attitudes over time on the ATSPPHS may therefore reflect worsening attitudes toward specialty mental health care in general, and/or psychotherapy more specifically. The results from two community-based surveys of public opinion of mental health service providers support this hypothesis. Angermeyer, Matschinger, and Riedel-Heller (1999) and Jorm et al. (1997) both found that for depression, the public perceived psychiatrists and psychologists as being less helpful than general practitioners.

The aim of the current study was to provide a more reliable test of whether and to what extent helpseeking attitudes, as measured by the ATSPPHS (Fischer & Turner, 1970), have changed over time in university student populations. Analyzing trends over time in attitudes toward mental health services may help explain the increasing rates of pharmacotherapy use and declining rates of psychotherapy use in the general population.