Refining Black men’s depression measurement using participatory approaches: a concept mapping study

To date, researchers and clinicians have relied on validated measures and clinically-derived handbooks, such as the Diagnostic and Statistical Manual of Mental Disorders (DSM-V), to diagnose depression. However, studies assessing the reliability and validity of these measures in Black male populations have yielded mixed results [3, 17, 18]. A small, but robust, body of research finds that Black and male populations articulate depressive symptoms in ways that diverge from the DSM criteria. In a review by Sydney Hankerson and colleagues [19], they synthesized evidence on racial and gender biases that contribute to misdiagnosis of depression among African American men. Of note, they found studies that describe that African Americans with major depressive disorder (MDD) are more likely to somaticize their emotional problems and, thus, may result in fewer MDD diagnoses in the clinical setting. Researchers also reported that the language used to describe depression in focus groups comprised of African Americans diverged from DSM characterizations, such as irritability, negative thought processes, hopelessness, loneliness, and social isolation, among others [20, 21]. Additionally, existing measures to capture male depression (e.g. Gotland Male Depression Scale, Male Depression risk scales, etc.) were developed and validated in non-Black samples, which omits the complex interplay of racialized and gendered experiences that may manifest in Black men’s descriptions of depressive symptoms [22‐24]. Collectively, these studies suggest that the DSM criteria of which most measures are derived may fail to identify (1) culturally nuanced symptoms and subsequently reduce opportunities for expanded set of symptom experiences as a function of the disorder and (2) how race- and gender-related factors intersect to inform Black men’s depression experience. Our study seeks to mitigate this gap using concept mapping, a participatory methodology [25].

Previously described limitations in depression measurement highlight the importance of engaging Black men directly to determine their depression experience. Concept mapping, a structured mixed methods approach to organize and operationalize stakeholders’ views on a particular research topic, is one such method that provides further evidence towards clarifying stakeholder-driven narratives about health [26]. The method combines qualitative data collection procedures and quantitative analytic approaches to create a pictorial representation of stakeholder inputs about a topic. In this study, we employed concept mapping to generate conceptualizations of how depressive symptoms were described in a community-based sample of Black men and their stakeholders. By employing a concept mapping approach to refine our understanding of the depression experience, Black men and their stakeholders are directly involved in the critical preliminary stages of scale refinement processes.

Our guiding hypothesis is that Black men’s marginalized social status in the United States fundamentally shapes their depression symptoms, and ultimately, the ways in which they conceptualize the depression experience. Specifically, we apply concept mapping to understand (1) the integration of race- and gender-related factors in Black men’s depression experience, and (2) the utility of concept mapping methodology to expand depression scale-refinement processes beyond existing DSM-V diagnostic criteria. By refining the conceptualization of depression with Black men and their stakeholders, we bridge important insights from extant research that posits that language around depression may diverge from DSM-V classifications of major depressive disorder [20, 27]. We assert that improved measurement of depression among Black men will provide more comprehensive opportunities for the early detection of depressive symptoms in community and clinical settings, thus mitigating the risk of morbidity and premature mortality associated with misdiagnosed or undetected symptoms. The goal of the study and our associated methodology is to provide a launch point for improved depression measurement, which may in turn, support better detection, diagnosis, and earlier entry points for Black men to seek mental healthcare.

Participants were eligible for this study if they were at least 18 years old, a current resident in the Research Triangle area of North Carolina, English-speaking, and self-identified as a Black man, Black woman, or primary care provider of any racial or ethnic background. Participants for the concept mapping process were recruited using a non-random purposive sample, which is the preferred sampling frame for concept mapping to achieve heterogeneity in the reflected items generated from participants [28]. Black men, women, and primary care providers were recruited to participate in different phases of the concept mapping study; Black men were involved throughout all three phases of the study, whereas Black women and providers were only involved in the initial brainstorming phase. This process was conducted in order to expand the pool of brainstormed items related to depression through the lens of key relational and therapeutic stakeholders in Black men’s health [27, 29, 30].

Black men and women stakeholder groups were recruited through targeted flyers distributed through community organizations and listservs of interest (e.g. fraternities, male-centered organizations, etc.), academic and community-based networks, such as gender-specific student organizations and male-centered mentoring organizations in predominantly Black neighborhoods, to generate a diverse range of eligible participants. Additional Black men were recruited using snowball sampling from previous participants. Primary care providers were recruited using direct contacts from Community Advisory Board (CAB) members and a recruitment services support team at University of North Carolina at Chapel Hill. Participants were compensated up to $50 for their participation in two in-person concept mapping sessions ($25 per session).

Data was collected from October 2017 to January 2018 in Durham, North Carolina in three phases: brainstorming; sorting; and interpretation. The lead author facilitated each phase and audio recorded sessions with stakeholders. The brainstorming phase consisted of group discussions during which participants were asked to brainstorm or “free list” ideas related to the focal question: “What are distinct characteristics of depression among Black men?” Two brainstorming sessions were conducted in-person with Black men and Black women, respectively. Primary care providers completed the brainstorming activity online due to time constraints. During the sorting phase, participants individually organized the list of brainstormed items into piles based on conceptual similarity. Participants labeled piles independently based on how they perceived the items related to each other. At baseline, participants completed a brief survey comprised of demographic information (e.g. age, occupation, etc.) and an assessment of depressive symptoms using the Center for Epidemiological Studies Depression (CES-D) 12-item scale [31]. The research protocol for our concept mapping study was approved by the Institutional Review Board at University of North Carolina at Chapel Hill and all study procedures were performed in accordance with relevant guidelines and regulations set forth by the institution. Informed written consent to participate in the study was obtained from all participants prior to the study initiation.

To maintain community perspectives throughout the study, we assembled an advisory board to guide the design, implementation, and analytic phases of our study. We convened a CAB of three members, two Black men and a Black woman primary care provider, who were able to offer a community-driven perspective of the research questions and process. CAB members were selected based on their familiarity with Black men’s health, ability to participate in periodic conference calls, interest in the research questions, and willingness to engage community leaders on the research topic. Recruitment of CAB members was conducted by convenience from community-based research studies focused on Black men’s health by research staff. Once the study launched, the CAB assisted with recruitment efforts, helped refine brainstormed items, and participated in the interpretation session at the conclusion of the study. CAB members were offered $60 for participation by the study’s primary investigator (first author).

We conducted a content analysis to compare concept map clusters to psychometric constructs identified through established factor analyses studies of several well-validated measures of depression. Namely, we compared our findings to the following measures: the 20-item Center for Epidemiologic Studies Depression Scale (CES-D) [31], the 9-item Patient Health Questionnaire (PHQ-9) [32], and the 21-item Beck’s Depression Inventory (BDI) [33]. Our content analysis compared each concept map cluster of the finalized concept map to confirmatory factor analysis results of the four selected depression scales. We analyzed both item wording and factor analyses for similarities in wording or phrasing of the construct. Concept mapping constructs were considered to be similar if equivalent content or wording was captured in the confirmatory factor analyses and resulting items in each depression scale.

Figure 1 illustrates the final cluster map derived from data collection during the study’s sorting phase. The final cluster map consists of six conceptual clusters labeled as (1) Physical States; (2) Emotional States; (3) Internal Conflict; (4) Communication with Others; (5) Social Pressures; and (6) Diminished Drive. Each cluster label represents the common theme of the statements in that particular cluster and was named by study participants during the interpretation phase. Each point on the cluster map corresponds to the statements reflected in Table 1.

The “physical states” cluster is an eight-item cluster that includes statements describing physical manifestations of negative affect. Statements in this cluster included somatic symptoms or clinical diagnoses, such as high blood pressure, heart palpitations, binge eating, and insomnia. “Emotional States” is a 16-item cluster that describes both the internalizing and externalizing aspects associated with depressed mood, such as anger and feeling hopeless, and included statements related to feeling stagnant, fatigued, or frustrated. The “communication with others” cluster consists of 18 brainstormed statements that describe challenges of maintaining interpersonal communication across multiple relationship domains, including family, friends, and romantic partners, as Black men deal with depressive symptoms. Statements in this cluster included being unable to communicate properly with others, isolated from others, and having unclear or limited communication. One participant described statements in this cluster as “what people withhold” when dealing with depressive symptoms. Of note, this cluster not only included verbal communication, but also elements of non-verbal communication that may be emotionally driven, such as limited facial expression or blunted affect. The “social pressures” cluster included ten items related to the strains associated with achieving traditional male gender roles. Statements in this cluster included adherence to norms of competition with other men and maintaining notions of success and power. The “internal conflict” cluster consisted of 11 brainstormed statements that named internalized sentiments such as feeling “off” and pessimistic. Notably, this cluster also includes items related to internalized standings in the Black community (e.g. not feeling supported in the Black community). Finally, “diminished drive” includes five brainstormed statements related to the unfinished business and untapped potential of Black men. Statements in this cluster included not able to complete tasks and not able to complete life goals, such as school and employment. Male participants continually highlighted gender norms related to not being able to provide for your family throughout the brainstorming phase as a key catalyst of depressive symptoms for Black men.

Table 2 illustrates results from our content analysis that compared concept mapping results with items and constructs identified in existing depression scales. We found that none of the existing measures included in our content analysis describe all of the clusters generated by our concept mapping study. Notably, all but one cluster, “social pressures”, was reflected in one or more of the four depression scales. Items and concepts related to “physical states” and “emotional states” as well as “diminished drive” clusters were described in factor analyses studies for each of the four scales.