Introduction

Healthcare workers have faced unprecedented professional and personal demands during the COVID-19 pandemic. Among medical providers in particular, concerns about infection, practical difficulties (e.g., in obtaining personal protective equipment), and exposure to high volumes of COVID-19 patients have contributed to increased psychological distress, including depression symptoms [1, 2]. This is compounded by the fact that many providers already reported elevated depression symptoms even before the pandemic began [3,4,5].

In spite of the growing need for mental healthcare, many medical providers who experience depression or other psychiatric concerns do not seek treatment. This may be due in part to stigma related to seeking help. For instance, among a sample of 2364 physicians, nearly half of those with a mental health concern avoided seeking treatment because they believed a mental health diagnosis would be embarrassing or shameful [4]. A survey of 1488 primary care providers (PCPs) found that almost three-fourths reported that fears of letting down colleagues and patients and reduced career opportunities were barriers to seeking help [3]. Further, many have reported that a culture of self-reliance and concerns about the professional repercussions of having a psychiatric diagnosis on one’s medical record present barriers to seeking mental healthcare [6]. Indeed, while stigma for seeking help has been reported in many populations, medical professionals in particular report disproportionately high rates of stigma [7].

Due to the stigma that medical providers experience for seeking mental healthcare and high distress from COVID-19, there has been increased interest and legislation to reduce stigma in this population (e.g., Dr. Lorna Breen Health Care Provider Protection Act) [8]. Such efforts are needed not only to reduce the burden of suffering among physicians, but also to increase the quality and efficiency of patient care, addressing both the Triple and Quadruple Aims [9, 10]. However, there is a lack of knowledge of effective ways to reduce stigma among medical personnel, in part because the basic processes by which stigma affects help-seeking in this population are unknown. In the general population, theoretical models have been successfully tested in which perceptions of public stigma become internalized in the form of self-stigma (i.e., predicted decreases in self-esteem if one were to seek help), which then results in negative attitudes toward help-seeking and lower help-seeking intentions [7, 11]. Additionally, these models have often focused on stigma related to seeking psychotherapy due to it being less commonly used than medication, particularly for depression [12, 13]. However, such models remain untested among medical providers. A better understanding of how stigma affects help-seeking in this population could lead to more effective help-seeking interventions.

The current study tested processes by which stigma impedes mental health help-seeking among medical providers. In particular, help-seeking among PCPs was examined due to especially high rates of depression and stigma reported by these professionals [3]. Additionally, this study focused specifically on associations of stigma with intentions to seek psychotherapy for depression. This was because stigma process models on seeking psychotherapy have been well characterized in the general population and because psychotherapy for depression is less commonly used in comparison with medication in spite of its similar effectiveness [13]. It was hypothesized that perceived public stigma for seeking psychotherapy would indirectly affect intentions to seek psychotherapy via two sequential mediators: self-stigma and then attitudes toward psychotherapy. Because PCPs’ personal experiences with psychotherapy may influence their referrals of depressed patients to psychotherapy [14, 15], exploratory analyses were also conducted to examine differences in PCPs’ intentions to seek personal psychotherapy versus intentions to refer patients to psychotherapy. Further, the association between intentions to seek personal psychotherapy and intentions to make a psychotherapy referral was explored.

Method

Licensed PCPs’ email addresses were obtained from the Idaho Board of Medicine, the Wyoming State Board of Nursing, and the Washington Medical Commission. Lists were also requested, but not obtained, from the boards of medicine in Montana, Wyoming, and Utah, and the boards of nursing in Idaho, Montana, Washington, and Utah. Data was collected during January–May 2020. PCPs were sent a recruitment email advertising a study on attitudes toward mental healthcare. Interested participants clicked on a link to access the survey online. They completed all study measures after providing informed consent. At the end of the survey, they provided demographic information and were directed to a debriefing page with the option to enter a raffle to win one of four $100 Amazon gift cards. This study was approved by the Institutional Review Board at the first author’s university before data collection.

Measures

Clinical Vignette

Participants first completed a clinical vignette. This consisted of a brief (123 words) written description of a patient who met minimum criteria for a moderate major depressive episode according to the International Classification of Diseases-10 [16]. After reading the vignette, PCPs were asked a single question on how likely they would be to refer the patient to psychotherapy on a scale of 1 (not at all likely) to 7 (very likely): “How likely would you be to refer this patient to psychotherapy?” This score was used as the indicator of intentions to make a psychotherapy referral. Similar referral vignettes have been used with PCPs previously [17].

Intentions to Seek Psychotherapy

After completing the vignette, participants answered one question on how likely they would be to seek psychotherapy for themselves if they were experiencing depression on a scale of 1 (not at all likely) to 7 (very likely). Specifically, they were asked, “If you were to experience symptoms of depression, how likely would you be to seek psychotherapy?” The score for this question constituted the indicator of intentions to seek psychotherapy. Similar single-item measures of intentions have been found to prospectively predict psychotherapy use in past research [18].

Perceived Public Stigma for Seeking Psychotherapy

The Perceptions of Stigmatization by Others for Seeking Help scale (PSOSH) [19] is a five-item self-report questionnaire that measures beliefs about how negatively others would view the participant for seeking psychotherapy (e.g., “If you sought counseling services for this issue, to what degree do you believe that the people you interact with would react negatively to you?”). Responses are given on a scale of 1 to 5, and total scores are divided by 5. Possible total scores range from 1 to 5, with higher scores indicating higher perceptions of public stigma. The PSOSH has demonstrated good internal consistency (\(\alpha\) = 0.88) and good 3-week test–retest reliability (r = 0.82) and is moderately correlated with other measures of stigma related to seeking psychological help [19]. In the current study, an internal consistency of \(\alpha\) = 0.90 was found.

Self-stigma for Seeking Psychotherapy

The Self-Stigma of Seeking Help scale (SSOSH) [20] is a 10-item questionnaire that measures beliefs about the extent to which seeking professional psychological help would negatively impact one’s self-esteem (e.g., “I would feel inadequate if I went to a therapist for psychological help.”). Responses are given on a Likert scale of 1 to 5 and total scores are divided by 10, with possible scores ranging from 1 to 5. Higher scores indicate higher levels self-stigma. The measure has demonstrated good internal consistency (\(\alpha\) = 0.89) and 2-month test–retest reliability (r = 0.71) [20]. It has been found to be positively correlated with scales measuring anticipated risks of seeking psychological help, negative attitudes toward seeking psychotherapy, and lower willingness to seek help, and it has demonstrated good discriminative validity with measures of perceived public stigma [19, 20]. Internal consistency was \(\alpha\) = 0.83 in the current study.

Attitudes Toward Seeking Psychotherapy

The Mental Help Seeking Attitudes Scale (MHSAS) [21] contains nine items measuring positive and negative appraisals of receiving professional psychological help. Because this study examined attitudes toward seeking psychotherapy specifically, we modified the wording of the MHSAS to ask only about seeking psychotherapy. It began with the stem, “If I had a mental health concern, seeking help from a psychotherapist would be…” followed by nine sets of opposing adjective pairs (e.g., useless, useful). We provided participants with the definition of “psychotherapist” as, “psychologists, clinical social workers, counselors, or any other mental health professional that you might seek psychotherapy or counseling from.” Responses on the MHSAS are given on a 7-point scale with the adjective pairs serving as anchors at each extreme of the scale; total scores are divided by 9. Possible scores range from 1 to 7, with higher scores indicating more positive attitudes. The MHSAS has demonstrated excellent internal consistency (\(\alpha\) = 0.94) and good 3-week test–retest reliability (r = 0.76) [21]. It has also been negatively correlated with perceived public stigma and self-stigma for seeking psychotherapy, indicating good concurrent validity [21]. An internal consistency of \(\alpha\) = 0.90 was found in the current study.

Data Analysis

Mediation analyses were conducted with the PROCESS macro for SPSS v27 using 5000 bootstrap samples to test the main hypothesis regarding indirect effects of perceived public stigma on help-seeking intentions. With this method, indirect effects with 95% confidence intervals that do not include zero are considered significant. Additionally, associations of demographic characteristics with help-seeking intentions were tested, and significant correlates were included as covariates in the model. For the exploratory analyses, a paired-samples t-test was conducted to compare PCPs’ intentions to refer a patient with depression to psychotherapy with their intentions to seek personal psychotherapy for depression. Last, bivariate correlation was used to test whether personal help-seeking intentions were associated with intentions to refer the vignette patient to psychotherapy.

Results

Participants (N = 112) consisted of Doctors of Medicine (MD; 67.0%), Doctors of Osteopathy (DO; 15.2%), Nurse Practitioners (NP; 14.3%), Doctors of Nursing Practice (DNP; 2.7%), and Doctors of Philosophy (PhD; 0.9%). Most identified as White (93.8%; Hispanic = 0.9%; Multicultural/Mixed Race = 0.9%; Asian = 0.9%; 3.6% declined to answer) and as male (55.4%; female = 44.6%), with a mean age of 49.16 years (SD = 11.77). This was a higher percentage of White and female PCPs than were in the two states with provider demographics available (Idaho and Washington) [22, 23]. However, the mean age was consistent with that of PCPs in these states [22, 23]. Participants reported working in a variety of primary care settings, including private practice (47.3%), hospital clinics (21.4%), community health centers (6.3%), Federally Qualified Health Centers (6.3%), university clinics (3.6%), and other settings (15.1%). They reported providing an average of 112.59 h (SD = 65.32) of direct patient care per month and had held their most advanced medical degree for M = 19.63 years (SD = 12.25). In contrast with past studies [3, 4], over half (61.6%) reported previous psychotherapy use.

Before conducting the main analyses, associations of demographic variables with intentions to seek psychotherapy were tested. Past psychotherapy use (t[110] = 2.83, p = 0.006), female gender (t[110] = 2.00, p = 0.048), and younger age (β = − 0.27, t = − 2.90, p = 0.004) were associated with stronger intentions to seek psychotherapy. As such, these were included as covariates in the mediation analysis. After controlling for these variables, the indirect effect of perceived public stigma (M[SD] = 2.02[0.64]) on intentions to seek psychotherapy (M[SD] = 5.38[0.1.57]) via self-stigma (M[SD] = 1.64[0.75]) and attitudes (M[SD] = 6.07[1.24]) was significant (estimate: − 0.03, 95% CI: [− 0.073, − 0.003]; see Fig. 1). While the direct effect of perceived public stigma on attitudes was still significant in this model, the direct effect on intentions was non-significant, demonstrating full mediation. This indicates that perceived public stigma predicts higher self-stigma, which then predicts more negative attitudes toward psychotherapy and finally, weaker intentions to seek psychotherapy for depression.

Fig. 1
figure 1

Indirect effect of perceived public stigma on intentions to seek psychotherapy

Full size image

Exploratory analyses on psychotherapy referrals were then conducted. PCPs reported stronger intentions to refer the depressed vignette patient to psychotherapy (M[SD] = 6.28[1.09]) than to seek personal psychotherapy for depression, with a medium effect (t(111) = 6.97, p < 0.001, d = 0.66). Additionally, PCPs’ intentions to seek personal psychotherapy and their intentions to refer the vignette patient to psychotherapy were moderately correlated (r = 0.51, p < 0.01). This indicates that PCPs who reported weaker intentions to seek personal psychotherapy for depression also reported a lower likelihood of referring the vignette patient to psychotherapy.

Discussion

The current study tested mechanisms by which stigma affects help-seeking for depression among PCPs. It also explored the association of intentions to seek psychotherapy with likelihood to refer patients with depression to psychotherapy. Results were consistent with stigma process models previously tested [7, 11]. Specifically, these results suggest that among PCPs, perceived public stigma (e.g., beliefs that seeking psychotherapy is shameful) becomes internalized as self-stigma (i.e., beliefs that one’s self-esteem would be reduced by seeking psychotherapy). In turn, self-stigma may lead to negative attitudes toward seeking psychotherapy, such as beliefs that it will be unhelpful or limit professional opportunities. Finally, negative attitudes may then impede intentions to seek psychotherapy when needed. Further, results of this study suggest that while PCPs could be more likely to refer patients to psychotherapy than to seek psychotherapy themselves, those with lower personal help-seeking intentions might also be less likely to refer patients to psychotherapy. Taken together, this indicates that stigma continues to be a barrier to PCPs receiving mental healthcare during COVID-19. This is problematic since providers’ mental health needs have increased during this period. Additionally, stigma may indirectly influence PCPs’ clinical decisions when working with patients with depression.

These findings should be interpreted in the context of several limitations. Notably, this study was observational and cross-sectional. Thus, while the results are consistent with previous stigma process models, causal inferences cannot be made. Additionally, this study’s sample identified almost entirely as White, and a large proportion reported past psychotherapy use. This may limit generalizability to more diverse samples of PCPs and those without past psychotherapy experience. Moreover, this study only examined attitudes and stigma related to seeking psychotherapy for depression. Stigma may have different effects on PCPs’ intentions to seek psychiatric medications, and future research is needed to examine medication-related stigma in this population. Stigma and attitudes may also differ when seeking help for conditions besides depression, such as substance use or schizophrenia spectrum disorders. Further, although stigma appears to predict PCPs’ personal help-seeking intentions, our analyses did not show a relationship of stigma with referral intentions. Thus, the correlation between intentions to seek personal psychotherapy and to refer the vignette patient to psychotherapy could be due to factors unrelated to stigma, such as comfort with prescribing antidepressants, local availability of psychotherapy providers, or judgements that a specific patient is better suited for medication than psychotherapy. Finally, referrals to psychotherapy were examined using a vignette rather than naturalistic measures of referrals, such as chart reviews.

In spite of these limitations, this study provides the first theory-driven investigation of stigma and seeking psychotherapy among PCPs, particularly in the context of COVID-19. As such, it highlights multiple help-seeking intervention targets. For example, increased openness about mental health struggles from physicians in leadership positions may help decrease concerns about letting down colleagues and perceptions that seeking care is shameful (i.e., addressing perceived public stigma). Additionally, removing structural barriers, such as licensing questions requiring disclosure of mental health conditions, could mitigate concerns about treatments such as psychotherapy negatively impacting professional opportunities (i.e., addressing negative attitudes toward psychotherapy) [24]. Further, fostering a culture of safety, vulnerability, and self-reflection within medical training and practice settings may help challenge self-stigmatizing beliefs about seeking psychotherapy [25]. Notably, the fact that the effect of perceived public stigma was fully mediated by self-stigma and attitudes indicates that efforts to reduce stigma must ultimately address self-stigma and attitudes toward psychotherapy in order to influence intentions to seek this treatment.

In the future, research could test contexts in which stigma-related interventions are most effective among PCPs. For instance, availability of mental healthcare providers may interact with stigma to affect PCPs’ personal psychotherapy use and psychotherapy referrals, particularly in rural locations. The characteristics of PCPs’ caseloads could also alter the effectiveness of stigma interventions. For example, it may be more difficult for providers to access psychotherapy if they experience irregular schedules or have increased hours of direct service provision, even after stigma-related interventions have been successfully implemented. Additionally, structural factors, such as the quality of employee assistance programs, may affect stigma and help-seeking directly (i.e., by decreasing the cost of psychotherapy) and indirectly (i.e., by showing agency support of personal mental healthcare). Further, increased coordination with mental healthcare agencies or integration of on-site behavioral healthcare personnel might enhance efforts to reduce stigma and facilitate both personal psychotherapy use and psychotherapy referrals [26].

Relatedly, future research could clarify whether stigma related to psychotherapy influences behavioral healthcare processes in primary care settings. For instance, studies could examine whether PCPs inadvertently convey their own perceptions of stigma for seeking psychotherapy when discussing treatment options with patients. Future studies could also test whether PCPs’ endorsements of stigma and personal help-seeking intentions are related to psychotherapy referrals for other conditions besides depression, such as elevated suicide risk. In addition, research could identify how PCPs can reduce patients’ perceptions of stigma for seeking psychotherapy when making referrals. This may be especially relevant to PCPs who already hold positive attitudes and low perceptions of stigma for seeking psychotherapy, such as those in the present study.