Masculinity norms and occupational role orientations in men treated for depression.

PURPOSE: A traditional male role orientation is considered to increase the risk of depression and preventing men from disclosing symptoms of mental illness and seeking professional help. Less is known about the variance of masculinity orientations in men already treated for depression and their role in the treatment process. In this study, patterns of masculinity norms and work role orientations will be identified among men treated for depression. Associations of these patterns with depressive symptoms, stigma and delay in professional help-seeking will be investigated.
METHODS: In a cross-sectional study, male role orientations (MRNS), work-related attitudes (AVEM), symptoms of mental disorders (PHQ), and attitudes related to stigma of mental illness (DSS) were assessed by standardized methods in a sample of 250 men treated for depression in general medical, psychiatric and psychotherapeutic services. Data were analyzed by means of latent profile analysis (LPA), by multinomial and linear regression models, and by path analysis.
RESULTS: The results of LPA revealed three latent classes of men treated for depression. Men assigned to class one reported a less traditional male role orientation, low professional ambitions and low coping capacities; men assigned to class two reported a traditional masculinity orientation, high professional ambitions but low coping capacities; men assigned to class three reported less traditional masculinity tended orientations, medium professional ambitions and high coping capacities. Men assigned to classes one and two to have more stigmatizing attitudes, longer periods of untreated illness and more severe symptoms of mental disorders, in comparison to men assigned to class three.
CONCLUSIONS: Overall, this study reveals that traditional masculinity norms and work-role orientations in men treated for depression are associated with a worse mental health status. Our study results also suggest that a slackening of traditional masculinity norms is associated with improved psychological well-being if it does not coincide with a complete distancing from professional ambitions and a lack of ability to cope with professional stress.

Introduction

While the worldwide depression prevalence among women is about twice that in men, suicide rates among men are significantly higher than among women [1-3]. In order to explain this paradoxical relation several authors hypothesize that a majority of men affected by depression regard their symptoms as incompatible with their masculine self-image and that they are therefore reluctant to seek help which in turn increases the risk of suicide [4-9]. This hypothesis is supported by study results indicating that adherence to a traditional masculine role orientation in depressed men was found to be related to increased self-stigma, which in turn worked as a mediator toward reduced willingness to seek help [10-11]. However, there is also evidence that the associations of traditional masculinity with adverse mental health behavior vary according to the specific set of masculine norms [9]. While adherence to masculine norms of self-reliance, power over women, and sexual dominance was found to be negatively associated with mental health, conformity to masculine norm of primacy of work was not related to any mental health-related outcome. Also, sexist and homophobic male attitudes appear to be associated with greater risk of poor mental health [11-12]. Recently, a prospective study on depression among male college students revealed that endorsement of self-reliance, playboy attitude and violent masculine norms increased the risk for depression, while adherence to winning and power over women was associated with lower depressive symptoms [13].

Several authors have hypothesized that gender-specific occupational role expectations and other workplace factors interact with the mental health impact of masculine role orientations [14, 15]. However, current findings about the associations between occupational role expectations and mental health are inconclusive. While studies from several countries [15] found that men working in male-dominated occupations or industries have a higher risk of depression than men working in occupations with a lower percentage of male workforce, results from an Australian national study on male health [16] revealed no direct associations between male workforce and mental health. A detailed analysis shows that it is not the male-dominated work environment per se but particular masculine attitudes among male workers that were associated with negative effects on their mental health [16]. While a high priority attributed to work was related to better mental health, other work-related attitudes such as self-reliance, dominance over others and winning had strong negative effects [16].

In sum recent research supports the hypothesis that adherence towards a traditional image of masculinity is related to lower male well-being and mental health [9, 11] and that this association is increased in male dominated occupational environments [15, 16]. However, in their consideration of what this implies for the development and the evaluation of treatment strategies, Seidler et al. [11] have criticized a one-sided negative focus of current masculinity concepts which, in their view, are reductionist and outdated given the changes in masculine images in modern societies. Furthermore, the authors state the need to investigate populations with clinically relevant depression [11].

The current state of research constitutes a weak basis for the development of new therapies for male depression. In particular, we do not know how different patterns of masculinity orientations are distributed in men treated for depression and how these patterns are associated with their clinical status. If a traditional masculinity orientation is a risk factor for depression, we would expect that a majority of male depressed patients hold such normative values. This expectation could be counteracted if men with traditional masculinity orientations were found to stigmatize people with mental disorders and, as a consequence, refuse professional help [17]. The interaction between traditional masculinity orientations, work related attitudes and depression symptom patterns is not well understood. Thus, it is currently unclear how noxious masculinity orientations can be dealt with in depression treatment.

In our study, we aim to address this research gap by identifying different patterns of masculinity and work role orientations in men treated for depression. We further want to investigate the associations of these patterns with stigmatization of mental illness, help-seeking behavior and the severity of psychiatric symptoms,

We expect that the results of our study will contribute to: 1. improving the knowledge about noxious and salutogenetic elements of masculinity and work-role orientations; 2. identifying specific target groups for male depression treatment with reference to types of masculinity and work role orientation; 3. developing adequate treatment techniques or settings for male depressed patients with reference to types of masculinity and work-role orientation.

Methods

Study design

This is a cross-sectional observational study on male persons who were treated for depression in psychiatric inpatient or outpatient services, with medication or psychotherapy or both at the time of the study.

Ethical approval

The study protocol was approved by the ethical board of the Medical Faculty at Ulm University under application number 202/15.

Sample

Inclusion criteria for study participation were: Male gender, age 18 to 65 years, being in treatment for depression in a psychiatric inpatient or outpatient service.

Recruitment process

In Germany outpatient treatment for depression is provided by family doctors (general practitioners), psychiatrists and psychological psychotherapists working in private practice while inpatient treatment is provided by general mental hospitals and by psychosomatic hospitals. Hospitals also provide treatment in outpatient clinics or day-hospitals. The treatment in all of these facilities is covered by the statutory health insurance for about 90% of the German population.

Psychiatric and psychosomatic hospitals provide inpatient and outpatient medical and psychotherapeutic treatment for all types of mental illness. Patients with psychotic disorders, addictions, severe depressive episodes and acute suicidal behavior are mainly treated in psychiatric hospitals. Psychosomatic hospitals are mainly concerned with diseases in which interactions between psychological and physical factors play the central role. These include eating disorders, somatoform disorders and depressive disorders. While psychological psychotherapists are only allowed to provide psychotherapy, office-based family doctors and psychiatrists can provide pharmacological treatment and (in case of adequate qualification) also psychotherapy.

In the attempt to reach male patients with different levels of depression severity and utilizing different types of treatment, we recuited study participants in all facilities mentioned above located in two major cities (Ulm/ Neu-Ulm and Augsburg) with an overall population of about 500.000 inhabitants and in four rural districts (Neu-Ulm, Günzburg, Dillingen and Donauwörth) with a total of about 500.000 inhabitants.

Mental health care professionals from the study regions were informed about the study by a symposium held at Ulm University before study onset. We also distributed information letters to relevant service providers by ordinary mail and e-mail. Furthermore, we initiated a media campaign in the local newspapers to inform about the aims of the study and invite men in current treatment for depression or burnout to participate in the study. Subjects who were interested could contact the study team by phone or e-mail.

Professionals working in outpatient facilities who consented to support the study were provided with information material which they were asked to distribute to eligible patients. The participant information pack included information about the study and an answering form together with a prepaid envelope addressed to the study center. Having returned the study forms, patients were contacted for an appointment by study workers. Patients who agreed to meet the study worker were asked to sign the informed consent form and to complete the study questionnaires. Patients in current inpatient treatment were asked by doctors or other clinical staff whether they were willing to meet the study worker for an information meeting. During the meeting potential study participants were informed about the aims and content of the study, and they were asked to sign the informed consent form and to complete the questionnaires. Male persons who answered to the press campaign were asked if they were currently being treated for depression and 18 to 65 years old. Where this was the case, patients were asked for an appointment with the study worker. During the meeting potential participants were informed about the aims and content of the study and asked to sign the informed consent form and to complete the study questionnaires.

All participants received a sum of 30 € for their participation in the study.

Assessment instruments and procedures

All assessments were conducted by means of standardized assessment instruments with validated psychometric properties.

Male Role Norms Scale (MRNS)

The Male Role Norms Scale (MRNS) is a self-rating questionnaire measuring respondents’ beliefs about male role norms by asking their agreement or disagreement with 57 statements on a 7-point Likert scale. While the MRNS was originally conceptualized to measure male role orientation along the four dimensions of “achieving status,” “cultivating independence and self-confidence,” “aggressiveness” and “antifemininity”, [18]. Thomson and Pleck [18] identified a three-factor structure including the dimensions “status,” “toughness” and “antifemininity.” The three-dimensional factorial structure was also confirmed for the German version of the MRNS [19, 20]. Cronbach’s alpha as reported for the German Version of the MRNS is .83 for status, .77 for toughness and .81 for antifemininity [19, 20].

Work-related behavior and experience scale (AVEM)

The Work-Related Behaviors and Experiences Pattern (AVEM = Arbeitsbzogene Verhaltens- und Erlebensmuster) is a self-rating questionnaire, measuring the following 11 dimensions:

1. subjective priority of work, 2. professional ambition, 3. over-commitment, 4. perfectionism, 5. professional distancing ability, 6. resignation tendency, 7. offensive problem management, 8. calmness/serenity, 9. experience of professional success, 10. life satisfaction, and 11. experience of social support. The AVEM consists of 66 items asking for agreement or disagreement with statements on a 5-point Likert scale. Cronbach’s alpha values between .79 and .83 are reported for the AVEM subscales [21].

Depression Stigma Scale (DSS)

The Depression Stigma Scale (DSS) has two subscales measuring negative attitudes about depression (DSS Personal Stigma) and assumptions about other people’s negative attitudes toward depression (DSS Perceived Stigma) [22, 23]. The original DSS consists of nine statements assessing the respondents’ own attitudes and nine statements assessing the respondents’ assumptions about the attitudes of other people by asking for agreement or disagreement on a 5-point Likert scale. For the purpose of our study, we added two items to assess the respondents’ gender-related attitudes (“depression is a typical women’s disease” and “it is unmanly to have depression”) and two items to assess the respondents’ assumptions about other people’s gender-related attitudes about depression (“most people think that depression is a typical women’s disease” and “most people think that it is unmanly to have depression”). Cronbach’s alphas for the original DSS total scale and its subscales were reported as 0.78, 0.76 and 0.82, respectively [24].

Patient Health Questionnaire (PHQ-SADS)

The Patient Health Questionnaire, Somatic Anxiety, and Depressive Symptoms (PHQ-SADS) includes 9 items for assessing depressive symptoms (PHQ-9), 7 items for assessing anxiety symptoms (GAD-7) and 15 items (PHQ-15) assessing somatic symptoms [24]. Test of the psychometric properties of the PHQ subscales revealed Cronbach’s alpha values of 0.89 for the PHQ-9, 0.92 for the GAD-7 and 0.80 for the PHQ-15 [24]. The PHQ-9 provides cut-off values for classifying minimal (1–4), mild (5–9), medium (6–14) and severe (15–27) depressive symptoms [24]. Since our sample includes only male participants we used the PHQ-15 without one item asking for menstruation complains.

Duration of Untreated Illness (DUI)

For estimating the Duration of Untreated Illness (DUI) we asked study participants in which year they had depressive symptoms for the first time and in which year they had first received professional treatment for these symptoms. We did not ask for the exact date of symptom or treatment onset. Therefore, years of treatment delay were calculated by (year of treatment onset–year of symptom onset).

Control variables

In order to control for confounding effects we included the following variables because of their established statistical associations with the measures used for the assessment of masculinity norms and work-role orientations [20, 21]: age, having a partner (no = 0; yes = 1), presence of children in the household (no = 0; yes = 1), higher education (general university entrance qualification = Abitur or university degree = 1; other = 0), unemployment (no = 0; yes = 1), household income above 3.000 € (no = 0; yes = 1), blue-collar work (n = 0; yes = 1); recruitment setting (1 = psychiatric hospital; 2 = psychosomatic hospital; 3 = family doctor; 4 = press invitation). The cut-off value of 3.000 € for household income was used because the mean net household income in Germany in 2016 was about 3.300 € [25].

Statistical analyses

We conducted a latent profile analysis (LPA) for the identification of latent patterns of associations between male role norms and work-related attitudes and experiences. LPA is the extension of latent class analysis (LCA) for continuous variables. LPA is based on the assumption that statistical associations between the characteristics x –x of a sampling unit i (e.g. a person) result from the sampling unit’s belonging to the category (class) k of a latent variable ϴ. The generalized probability model of the LPA for continuous variables is defined as: where the term f(x) indicates the class-specific density function of the distribution of the means, variances and covariances of the variables x –x, and π indicates the probability of membership of the sampling unit i in the latent class k [26].

Membership probabilities were estimated by means of a maximum likelihood approach with the expectation-maximization (EM) algorithm. Sampling units were assigned to latent classes based on maximal membership probability [26, 27].

Comparative model fit was tested by means of the Akaike information criterion (AIC) and the adjusted Bayes information criterion (ABIC), whereby lower values indicate better fit [28, 29]. The maximum number of classes was determined by means of the likelihood ratio test, indicating whether a model with k classes fits the data significantly better than a model with k– 1 classes [27]. The certainty of the class delineation is estimated by the entropy parameter, which ranges from zero to one. Values closer to one indicate better discrimination. LPA was computed using M-Plus 7.2 [30].

For the investigation of sociodemographic and mental health care characteristics related to latent class assignment, we computed a multinomial logistic regression model with assignment to the highest numbered class as reference category, including the control variables listed above.

We conducted analyses of variance (ANOVA) to test mean differences of the items included in the LPA and for symptoms of depression (PHQ-9), anxiety (GAD-7), somatization (PHQ-15), DSS personal stigma, DSS perceived stigma and DUI between latent classes. We tested the equality of variances by means of the Bartlett test and applied Bonferroni’s post hoc test in case of equal variances and Scheffe’s post hoc test in case of unequal variances for testing mean differences between latent classes.

We computed multiple linear regression models to estimate adjusted means by class membership for symptoms of depression (PHQ-9), anxiety (GAD-7), somatization (PHQ-15), DSS personal stigma, DSS perceived stigma and reported DUI controlling for age, presence of a partner, presence of children, education, unemployment, income, blue-collar work, duration of mental illness and recruitment setting. In addition, all models except that for reported DUI were controlled for reported DUI. We estimated the variance inflation factor (VIF) for each model variable to test for multicollinearity. An average VIF > 10 is regarded as indicating collinearity problems in a linear regression model [31].

We computed a path model to investigate the direct and indirect effects of latent class membership on stigmatization (total DSS score), the reported DUI and the severity of psychiatric symptoms (total PHQ-SADS score) using the membership to class 3 as reference category. We estimated robust standard errors for path coefficients in order to take account of the skew distribution of the variable indicating DUI. For testing the fit of the path model we estimated the Comparative Fit Index (CFI), the Tucker Lewis Index (TLI), The Root Mean Squared Error of Approximation (RMSEA) and the Standardized Root Mean Square Residual (SMR). A good model fit is indicated by CFI and TLI above 0.95 and RMSEA and SMR below 0.05 [32].

Regression models and path analyses were computed using STATA 15 [33].

Results

In total, 265 men agreed to participate in the investigation. 15 persons were excluded from the analysis because they were not currently being treated for depression. From the 250 men included in the data analyses (see Table 1), 128 (51.2%) were recruited in psychiatric hospitals, 45 (18.0%) came from psychosomatic hospitals, 6 (2.4%) were asked by family doctors, and 71 (28.4%) study participants had responded to the press invitation. Of the participants who responded to the press invitation 5 (7.7%) were treated by family doctors, 28 (43.1%) were treated by office based psychiatrists, and 25 (38.5%) were treated by office based psychologists. Having ever been treated for depression in a psychiatric or psychosomatic hospital was reported by 31(43.7%) of the participants who responded to the press invitation.

Table 1

Sample characteristics by recruitment setting.

	Total	Psychiatric hospital	Psychosomatic hospital	Family doctor	Press invitation
N (%)	250 (100)	128 (51.2)	45 (18.0)	6 (2.4)	71 (28.4)
Age mean (SD)	46.6 (12.0)	46.7 (12.9)	43.2 (10.0)	36.3 (15.2)	49.4 (10.3)
Higher education n (%)	93 (37,2)	29 (22.7)	25 (55.6)	3 (50.0)	36 (50.7)
Monthly net household income above 3000€ n (%)	95 (38.0)	35 (27.3)	27 (60.0)	2 (33.3)	31 (43.7)
Living with partner n (%)	148 (59.2)	72 (56.3)	31 (68.9)	2 (33.3)	43 (60.6)
Living with children n (%)	149 (59.6)	84 (65.6)	24 (53.3)	1 (16.7)	40 (56.3)
Unemployed n (%)	33 (13.2)	19 (14.8)	5 (11.1)	0 (0.0)	9 (12.7)
Blue collar worker n (%)	64 (25.6)	40 (31.3)	7 (15.6)	0 (0.0)	17 (23.9)
Duration of illness in years mean (SD)	11.5 (10.2)	11.8 (10.2)	8.7 (8.0)	13.2 (10.5)	12.8 (11.3)
Duration of untreated illness (DUI) in years mean (SD)	3.7 (6.5)	3.1 (5.9)	4.7 (7.6)	3.3 (5.8)	4.2 (6.8)
Severity of depression (PHQ-9)
Minimal (1–4) n (%)	18 (7.2)	9 (7.0)	3 (6.7)	1 (16.7)	5 (7.4)
Mild (5–9) n (%)	47 (18.8)	23 (18.0)	8 (17.8)	0 (0.0)	16 (22.5)
Medium (10–14) n (%)	61 (24.4)	32 (25.0)	11 (24.4)	1 (16.7)	17 (23.9)
Severe (15–27) n (%)	124 (49.6)	64 (51.1)	23 (51.1)	4 (66.7)	33 (46.5)

On average, participants were 47 years old (SD = 12.0 years), 37.2% (n = 99) had a higher education, and 38.5% (n = 95) had a monthly household income above the average of the German households (3000 € and higher). About 59.2% (n = 148) lived with a partner and 59.6% (n = 149) lived, with children. While 25.6% (n = 64) of the participants were blue collar workers, 13.2% (n = 33) stated to be unemployed at the time of the interview.

On average, participants had first suffered symptoms of depression 11.5 years (SD = 10.2) earlier but the mean duration of untreated depression was 3.7 years (SD = 6.5 years).

About half of the sample reported symptoms of depression classified as severe based on the PHQ-9 cut-off value. With the exception of the few participants who were recruited from family doctors the distribution of the symptom severity varied only slightly across recruitment settings.

Results of Latent Profile Analysis (LPA)

Model fit parameters for LPAs with increasing numbers of classes are presented in Table 2. As indicated by the LRT, the three-class model fits the data significantly better than the two classes model, while the model fit does not significantly improve with more than three classes. The AIC, the ABIC and the entropy parameter improve with an increasing number of classes, but the change is greatest between the two- and three-class models.

Table 2

Model fit parameter for the LPA model selection.

Model	classes k	Log-likelihood	LRT p k-1	AIC	ABIC	Entropy
1	2	-10702.648	0.2720	21079.494	21094.603	0.821
2	3	-10496.747	0.0231	20809.894	20830.274	0.843
3	4	-10346.947	0.3802	20702.015	20727.665	0.847
4	5	-10233.093	0.3621	20642.187	20673.108	0.859

LRT = Log-likelihood ratio test; AIC = Akaike information criterion; ABIC = Adjusted Bayes Information Criteria

Based on the comparison of model fit parameters, we selected the three-class model for further consideration.

Fig 1 presents the latent class profiles of the mean distribution of the dimensions of adherence to traditional masculinity and occupational role orientation broken down by class membership in the three-class model (means and SD are presented in S2 Table). The first three categories represent the dimensions of the masculinity role norm scale (MRNS): status, toughness and anti-femininity. Comparisons of means between the three classes indicate that men assigned to class 2 (red line) have significantly higher values on all three dimensions indicating a more traditional male role orientation, while the values of those assigned to class 1 (blue line) and class 3 (green line) do not differ significantly from each other and indicate a less traditional male role orientation.

Fig 1

Latent class profile for masculinity norms and work role orientations.

Means and 95% confidence intervals.

The comparison of the dimensions representing job-related attitudes reveals that men assigned to class 2 reported the highest levels of job priority, occupational ambition, overcommitment and perfectionism, followed by those men assigned to class 3. Men assigned to class 1 reported the lowest level of these attitudes. Compared to men assigned to classes 1 and 3, men assigned to class 2 reported the lowest ability to distance themselves from their jobs while also showing the greatest tendency to give up when confronted with professional problems. In contrast, men assigned to class 3 reported the lowest likelihood of resignation and the highest level of “coping with professional problems in an offensive way”. Men assigned to class 3 reported the highest level of serenity in their professional context, and they also reported the highest level of “experiencing success, social support and life satisfaction” compared to those assigned to classes 1 and 2. While men assigned to class 2 had the lowest level of serenity, men assigned to class 1 reported the lowest level of experiencing job success. Men assigned to classes 1 and 2 did not differ with regard to life satisfaction or the experience of social support.

In summary LPA results indicate that:

Men assigned to class 1 reported a less pronounced adherence to traditional masculinity norms combined with low occupational ambitions and a low ability to manage professional stress and to experience professional success and recognition.

Men assigned to class 2 reported a strong adherence to traditional masculinity norms combined with high professional ambitions but a low ability to manage professional stress and to experience professional success and recognition.

Men assigned to class 3 reported a less pronounced adherence to traditional masculinity norms in combination with medium professional ambitions and a high ability to manage professional stress and to experience professional success and recognition.

Associations of latent class assignment with sociodemographic and treatment characteristics

The between class differences of sociodemographic characteristics and recruitment setting are provided in Table 3.

Table 3

Multi-nomial logit regression model of latent class assignment on sociodemographic characteristics, duration of illness and recruitment setting.

N (%)	Class 1 85 (34.0)	Class 2 58 (23.2)	Class 3 107 (42.8)	RR1) Class 1 vs Class 3 (p)	RR1) Class 2 vs Class 3 (p)
Age mean (SD)	48.5 (11.4)	45.5 (11.9)	45.7 (12.4)	1.03 (0.148)	0.99 (0.524)
Higher education n (%)	22 (25.9)	16 (27.6)	55 (51.4)	0.28 (0.003)	0.27 (0.006)
Monthly net household income above 3000€ n (%)	20 (23.5)	22 (37.9)	53 (49.5)	0.56 (0.170)	0.98 (0.964)
Living with partner n (%)	36 (42.4)	33 (56.9)	79 (73.8)	0.21 (0.009	0.35 (0.012)
Living with children n (%)	48 (56.5)	41 (70.7)	60 (56.1)	1.58 (0.296)	3.67 (0.005)
Unemployed n (%)	18 (21.2)	6 (10.3)	9 (8.4)	2.95 (0.045)	1.10 (0.871)
Blue collar worker n (%)	28 (32.9)	18 (31.0)	18 (16.8)	1.35 (0.492)	1.98 (0.136)
Duration of illness in years mean (SD)	13.5 (11.3)	12.1 (11.3)	9.6 (8.3)	1.05 (0.075)	1.01 (0.737)
Duration of untreated illness (DUI) in years mean (SD)	3.3 (5.8)	5.7 (8.7)	3.0 (5.3)	0.94 (0.150)	1.05 (0.219)
Recruitment setting
Psychiatric hospital n (%)	40 (47.6)	27 (46.6)	61 (57.0)	Ref.	Ref.
Psychosomatic hospital n (%)	14 (16.5)	13 (22.4)	18 (16.8)	4.43 (0.005)	3.68 (0.013)
Family doctor n (%)	2 (2.4)	1 (1.7)	3 (2.8)	2.09 (0.470)	1.32 (0.831)
Press invitation n (%)	29 (34.1)	17 (29.3)	25 (23.4)	3.64 (0.004)	3.06 (0.018)

1) Multi-nomial logit regression model with assignment to Class 3 as reference category.

As revealed by the risk ratios (RR) resulting from the multinomial logistic regression model (Table 3) the probability of assignment to latent class 1 in comparison to assignment to class 3 was lower for participants who had a higher education, and for those who lived with a partner, but higher for participants who were unemployed. In comparison to participants who were recruited in psychiatric hospitals those who were recruited in psychosomatic hospitals or those who were recruited by press invitation had a higher probability to become assigned to class 1 rather than class 3.

The probability of assignment to class 2 vs. class 3 was lower for participants with a higher income and for those who lived with a partner, but higher for those who had children. In comparison to participants who were recruited in psychiatric hospitals those who were recruited from psychosomatic hospitals or those who were recruited by press invitation had a higher probability to become assigned to class 2 rather than class 3. A pseudo-R2 of 0.16 indicates a sufficient fit of the multinomial logit model.

Symptoms of mental disorder, stigmatization and duration of untreated illness by latent class membership

The adjusted means (see S3 Table for raw means) for the PHQ subscales in Fig 2 indicate that participants assigned to class 3 had significantly fewer symptoms of anxiety and depression than those assigned to classes 1 and 2 while the means of somatization symptoms differed only between classes 2 and 3. Results of the VIF estimation reveal an average VIF of 1.47 for all regression models which is far below the critical value of 10. The highest VIF was estimated for reported duration of depressive illness with 2.18 which is also far below the critical value of 10.

Fig 2

Patient Health Questionnaire (PHQ-SADS) subscales by latent class assignment.

Means and 95% confidence intervals adjusted for age, partner, children, higher education, unemployment, bluecollar worker, income, duration of illness, duration of untreated illness, recruitment setting.

Path analyses

Fig 3 reveals the standardized path coefficients (ß) and robust standard errors (se) of latent class membership in classes 1 and 2 vs. membership in class 3 on stigmatizing attitudes, reported DUI and symptoms of mental disorder.

Fig 3

Standardized path coefficients and robust standard errors of the effects of assignment to latent Class1 or latent Class 2 vs. latent Class 3 (reference category) on stigmatizing attitudes (stigma = DSS total sum score), the duration of untreated illness (DUI), and symptoms of mental disorder (PHQ = PHQ-SADS total sum score).

As indicated by path coefficients, study participants assigned to class 1 reported significantly more stigmatizing attitudes (ß = 0.28; se = 0.06; p < 0.001) and significantly more severe symptoms of mental disorder (ß = 0.33; se = 0.056; p < 0.001) compared to those assigned to class 3. The path coefficient between class 1 membership and DUI is not significant (ß = -0.022; se = 0.063; p = 0.729).

Study participants assigned to class 2 reported significantly more stigmatizing attitudes (ß = 0.43; se = 0.06; p < 0.001) and more severe symptoms of mental disorder (ß = 0.4; se = 0.067; p = 0.001) than those assigned to class 3, while the path from class 2 to DUI (ß = 0.13; se = 0.072; p = 0.091) indicates no significant effect.

While stigmatizing attitudes were significantly related to an increased DUI (ß = 0.1; se = 0.016; p < 0.001) and to increased severity of symptoms (ß = 0.17; se = 0.06; p < 0.01) the path coefficient from DUI to symptoms of mental disorder (ß = -0.07; se = 0.06; p < 0.261) indicate no significant effect.

As indicated by the decomposition of direct and indirect effects, membership to class 1 (ß = 0.02; p < 0.001) and membership to class 2 (ß = 0.044; p = 0.001) were both indirectly related to a longer DUI compared to membership to class 3.

The model explains 3.8% of the variance of DUI but 16% of the variance of stigmatizing attitudes and 24% of the variance of symptoms of mental disorder. The model fit parameters (CFI = 1.0; TLI = 1.0; RMSEA = 0.000; SRMR = 0.006) indicate a very close fit of the model with the empirical covariance structure.

Discussion

Our findings confirm previous results that the adherence to traditional masculinity concepts is directly associated with more stigmatizing attitudes against depressive illness and increased symptoms of mental disorder [10, 11, 34, 35]. However, our results also indicate that men treated for depression are not a homogenous group with respect to their adherence to traditional masculinity norms and occupational role orientations. Furthermore, our findings reveal that a less pronounced adherence to traditional masculinity is only associated with better mental health and less stigmatizing attitudes if it coincides with a medium level of occupational ambition and adequate abilities to cope with occupational stress and to perceive recognition at work.

These results underline the interconnectedness between masculinity concepts and occupational role orientations for male mental health [36]. In western societies professional success and social status are core elements of the traditional male gender role [37]. The workplace therefore can become both an important generator of feelings of personal accomplishment and social recognition but also a source of emotional strain and the risk of failure [38-40]. Our data suggest that those study participants who adhere most strongly to the traditional masculine role orientations have the highest levels of professional ambition and the lowest capacity to cope with occupational stress. Research on occupational mental health revealed that job over-commitment is an intrinsic risk factor increasing the negative effects of job-strain on mental health [39].

Under such circumstances keeping a distance from occupational ambitions could ease emotional strain and fear of failure [39]. Further research should investigate, whether this relief comes at the price of losing crucial emotional rewards, e.g. experiencing accomplishment and success. Both pronounced work role adherence and radical distance appear to be related to poorer emotional well-being and more stigmatizing attitudes (when compared with medium levels of occupational ambition and effort).

With regard to reported DUI, we found only indirect effects of traditional masculinity and work-role orientations indicating that these effects were mediated by stigmatizing attitudes. This is in accordance with the finding of a recent review article of 27 studies [41] that negative attitudes toward people with mental illness were associated with lower willingness to seek professional help. On the basis of an elaborated theoretical model Schomerus et al. [17] found that with increasing level of stigmatizing attitudes people with untreated mental illness were more reluctant to identify themselves as being mentally ill and to accept the need for mental health care.

Our finding, of no association between reported DUI and symptom severity is in accordance with previous studies revealing that longer DUI is related to reduced treatment effectiveness and a lower likelihood of remission but not to symptom severity and other clinical features [42-46].

Limitations

Our study sample includes only men currently treated for depression. Therefore, we cannot draw any conclusions about the representativity of our study results. Therefore, the finding that only 23% of study participants had a traditional masculinity orientation may reflect that men with a lower adherence to traditional masculinity norms are more prepared to seek professional help.

The recruitment of participants via press invitation could result in a further limitation of generlizability due to self-selection bias associated with peoples’ previous experiences with treatment or other clinical or personal characteristics.

In addition, our measurement of DUI is based on self-report only and therefore susceptible to recall bias.

Finally, due to the cross-sectional design our study does not allow to make any conclusions about causal effects or processes of change.

Conclusions

Overall, this study reveals that masculinity norms and work-role orientations are relevant among men treated for depression. Our results suggest that any slackening of traditional masculinity norms may be related to better psychological well-being when it not coincides with a complete distancing from work-role ambitions and/or deficits in coping with occupational stress. Longitudinal studies are needed to clarify whether helping depressed men (a) to take a distance from traditional masculinity norms, (b) to enhance a balanced work-role orientation, (c) to develop effective stress coping, and (d) to reduce stigmatizing attitudes could improve the effectiveness of depression treatment.

Depression Stigma Scale (DSS) subscales and duration of untreated illness (DUI) by latent class assignment.

Means and 95% confidence intervals adjusted for age, partner, education, unemployment, income, duration of illness, duration of untreated illness (only in the DSS models), recruitment setting.

(DOCX)

Click here for additional data file.

Cronbach’s alpha of the study instruments in the current sample.

(DOCX)

Click here for additional data file.

Means and standard deviations of the dimensions of the Male Role Norms Scale (MRNS) and the work related behavior and experience scale (AVEM) by latent class assignment.

(DOCX)

Click here for additional data file.

Raw mean differences of PHQ-SADS subscales, DSS subscales and DUI by latent class assignment.

(DOCX)

Click here for additional data file.

(TXT)

Click here for additional data file.

(DAT)

Click here for additional data file.

(STS)

Click here for additional data file.

13 Dec 2019

PONE-D-19-27414

Why do men hesitate to seek professional treatment for depression? The influences of masculinity norms and Stigmatizing attitudes on the delay of the search for professional help by men with depression

PLOS ONE

Dear Dr. Kilian,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by January 7, 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Stephan Doering, M.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Introduction

The question of the role of masculinity in delaying treatment seeking is an important one, however the study as presented does not focus sufficiently on this. In both the introduction and the discussion the focus is much more on male role orientations and work roles in relation to depression. The problem with that focus is that the sample are all in treatment for depression and there is no non-depressed comparison group. Thus, all the study can really examine is different configurations of masculinity that can be discerned among depressed men. The authors need to make a much stronger case about why this warrants attention, and re-title the paper to reflect that focus. I believe the question of masculinity and delay in help-seeking, and the role of stigma in that, is of much more interest and if the paper could be refocused around that it would be of more value.

The inclusion of ‘burnout’ is at no point explained – is it a proxy for poor mental health? Or a condition in its own right? Or a recruiting convenience? It needs to be discussed in the introduction and a rational provided for combining a burnout sample with a depressed sample.

Methods.

The paper doesn’t report on the qualitative element of the study so I don’t think it necessary to present the paper as based on a mixed-methods design.

Can the cities or towns where the study took place be named in addition to the country and the years of the study.

The inclusion of ‘burnout’ as an inclusion criteria needs much more explanation – i.e. what is ‘burnout’, how is it diagnosed, what is the relationship between burnout and depression – basically we need sufficient information to be able to judge that combining these two samples provides a suitable sample for the analysis undertaken and the conclusions drawn.

We need more information on participants recruited through newspaper advertisements – how were they screened.

We need more information on the format of data collection – was it self-complete questionnaires, was it by face-to-face interview? Was it conducted by clinic staff or by the researchers etc.

Delay in psychiatric treatment – notes that delay is calculated based on “the year in which a psychiatric treatment for depressive symptoms was used for the first time”. In table 1 delay is reported as ‘months untreated’ – need to explain how those months were calculated i.e. from January in the year treatment was initiated?

Also, if receiving treatment for burnout is an eligibility criterion is it possible that some of the men might never have been treated for depression?

Overall, we need more clarity around the diagnostic status of the sample at recruitment. This also needs to be reported in Table 1.

Results.

It would be good to see descriptive statistics for the various measures in Table 1 as well. And as mentioned above, diagnostic status (at minimum if they are being treated for depression or for burnout’.

The text describing figure 5 appears to have errors – the second sentence compares class 2 to class 2. There are also some spelling errors in that paragraph.

Discussion

The discussion would benefit from being more focused on the question that the paper describes in the title and abstract – the relationship between masculine norms and work role and the delaying of treatment seeking, and the putative role of stigma in that relationship. The focus is predominantly on differences in adherence to traditional masculinity and occupation role orientations in a depressed/burnout sample, and the authors have not made a strong argument about why this is important to study.

Much of the discussion then describes the relationship of masculinity and work role orientation which is not relevant to the original research question. Moreover the assumption made “that traditional masculine role orientation is not harmful per se to mental health, but depends on its functional and contextual meaning in the work environment” cannot be supported by the study as the sample is comprised of only men with depression.

Reviewer #2: Overall, the paper paper is focused on an important gap in the literature and makes a contribution to the field. However, there were some limitations which must be seriously taken into account. These include generalizability of the findings without more information about recruitment process and sample, types of occupations of the participants, as well as the recall bias that may be affecting one of the main outcomes (delay in treatment). The study can only provide information on men who ultimately seek care, and there may be a much wider pool of men who never seek treatment at all, who may not be represented in these results and this should be acknowledged.

There is some convoluted writing and significant typos. It may be useful to have the paper copy edited (perhaps professionally) for standard English when revising the paper.

Abstract:

• More detail is needed on the methods. How were patients recruited? Convenience sample? What measures were used? Are they validated measures? Was the study quantitative or qualitative? What was the participation rate? Some of this information can be provided solely in the manuscript, but more detail is needed in the abstract as well.

• Last sentence of the conclusions should read: “in the development of strategies to reduce the delay…and to tailor the psychiatric treatment…”

Introduction:

• The introduction needs to be more focused on the importance of the work environment earlier. As the introduction currently reads, it is not clear immediately that the focus will be occupational ‘identity’ or norms.

• Page 3, last sentence of the first paragraph: “negatively associated with depressive symptoms” is unclear. Can say, “was associated with lower depressive symptoms,” or other language to make the association clearer.

• Page 4, first sentence: Please specify what the ‘control populations’ were. Individuals without depression?

Methods:

• The fact that this is a mixed-methods study needs to be in the abstract.

• Please specify whether the measures were validated measures from the very beginning (first paragraph of study design section).

• There are multiple typos in the methods section.

• When you say that a ‘sub-sample of study participants were selected on the bases if the results’ please specify what the basis of this decision was. Will help the reader understand generalizability of the results. Who was chosen to be interviewed?

• Why was this age rage chosen? Please justify the rationale. Was the idea to choose only working-age men? If so, please specify.

• For some readers, such as myself, we are unfamiliar with the term ‘psychosomatic hospitals.’ Please provide a bit more background on these.

• The paper should specify how many patients came from each of the different settings, as again, this will affect generalizability.

• Please provide more info on the recruitment process. Were patients asked to participate if they were having depressive symptoms, if they were in treatment for depression, etc. What did the recruitment messaging say?

• It says that eligible patients were contacted by doctors or clinical staff about willingness to participate. Was this the case even for participants who came in via the press release?

• Cronbach’s alpha numbers are part of results, not methods. Please move these to the results section.

• Please provide the full name for all acronyms (such as AVEM on page 6).

• Some questions were added to validated scales. This would be unvalidated measures unless validated elsewhere. Please specify if that was the case.

• For ‘delay of treatment,’ which is an important outcome in this study, please provide further information on how this was assessed. There seems to be a lot of room for recall bias, which is inherently problematic and should be acknowledged in the limitations section.

• The information on the LPA could potentially be included in an appendix if allowed, rather than in the body of the paper.

• ‘one-way analyses of variance’= ANOVA and should be specified.

Results:

• Many of the covariates were not described in the methods. How is ‘higher education’ defined? College degree? More than college degree? Duration of illness is lacking units. Is this years, months, etc. Why was the cut-off of >3000 euros used? Please provide the rationale for this decision.

• Page 13, last sentence: the sentence structure is very convoluted, which is obscuring the meaning of the sentence. Do you mean, ‘participants in latent class 1 were less likely to have a partner…than participants in latent class 3’?

• Page 15, last paragraph, second sentence: class 2 is compared to class 2.

• The authors seem to be presenting an overwhelming amount of results that are hard to interpret. I would consider trying to figure out what the main results and take-away findings are and moving all others to an appendix section.

• Many typos throughout the results section.

Discussion:

• The first sentence of the discussion section was not well set up in the introduction. It was unclear that this was the hypothesis that the authors were challenging.

• The bullets in the discussion section are more re-hashing of the results and are actually much clearer summaries than the ones provided in the actual results section.

• ‘Sex’ and ‘gender’ are used interchangeably throughout the paper, when the authors are in fact referring to ‘gender’ only throughout.

• Page 19: there should be something in the discussion about how the info from the study could be used in practice.

Limitations section:

• Self-selection of participants is an important limitation, as it limits the generalizability of results. Can only speak for patients IN treatment who were willing to participate in the study. Cannot speak for men who do not seek treatment at all or who are unwilling to participate in the study. Generalizability should be listed as a limitation specifically.

• It is also a limitation that delay is all self-report, which is highly subjected to recall bias. Unclear how severity of illness, duration of treatment, etc may impact recall of treatment delay.

• A major limitation is that type of occupation is not included in the study results. What type of work did the men included in the study do? Blue collar vs white collar? This seems like it could strongly affect or influence the findings.

Conclusions:

• Can unfortunately not speak about the ‘treatment process,’ only ‘seeking treatment’ or ‘access’ to treatment.

• The last sentence, while true, seems unrelated to the study. More could be said specifically about the impact or use of the findings.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

20 Feb 2020

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Introduction

The question of the role of masculinity in delaying treatment seeking is an important one, however the study as presented does not focus sufficiently on this. In both the introduction and the discussion the focus is much more on male role orientations and work roles in relation to depression. The problem with that focus is that the sample are all in treatment for depression and there is no non-depressed comparison group. Thus, all the study can really examine is different configurations of masculinity that can be discerned among depressed men. The authors need to make a much stronger case about why this warrants attention, and re-title the paper to reflect that focus. I believe the question of masculinity and delay in help-seeking, and the role of stigma in that, is of much more interest and if the paper could be refocused around that it would be of more value.

Authors' reply

We agree with the reviewer that the main focus of the manuscript is the association between masculinity and work-role orientation with psychiatric symptoms, stigmatization and the delay of treatment in men already treated for depression. In our view the investigation of this question is important because it can improve our understanding of how relevant aspects of masculinity are in the treatment process.

We changed the title of the manuscript to make our intention more clear and we tried to focus the introduction.

Reviewer

The inclusion of ‘burnout’ is at no point explained – is it a proxy for poor mental health? Or a condition in its own right? Or a recruiting convenience? It needs to be discussed in the introduction and a rational provided for combining a burnout sample with a depressed sample.

Authors' reply

The main reason for including people with burnout was that we suspected that particularly by family doctors the diagnosis of burnout will often be used when they guess that patients refuse to accept the diagnosis of depression. We therefore expected that by including patients treated for burnout we would get access to patients who are reluctant to accept a diagnosis of a mental disorder. However, due to the fact that only very few patients (n = 15) treated for burnout participated in the study, we decided to exclude these patients from our analyses to get a more homogenous sample. We re-calculated all statistical analyses on the basis of the new sample.

Reviewer

Methods.

The paper doesn’t report on the qualitative element of the study so I don’t think it necessary to present the paper as based on a mixed-methods design.

Authors' reply

The study was conceptualized as a mixed-method study in the sense that the results of the standardized quantitative part of the investigation should be used as the basis for the selection of study participants for further qualitative inquiries. However, we agree with the reviewer that the current manuscript reports only results of the standardized quantitative part of the study. Therefore, we followed the recommendation not denoting the study as mixed-method in this manuscript.

Reviewer

Can the cities or towns where the study took place be named in addition to the country and the years of the study.

Authors' reply

According to the recommendation of reviewer 1 we named the main cities and described the study region more detailed in the method section.

Reviewer

The inclusion of ‘burnout’ as an inclusion criteria needs much more explanation – i.e. what is ‘burnout’, how is it diagnosed, what is the relationship between burnout and depression – basically we need sufficient information to be able to judge that combining these two samples provides a suitable sample for the analysis undertaken and the conclusions drawn.

Authors' reply

See above.

Reviewer

We need more information on participants recruited through newspaper advertisements – how were they screened.

Authors' reply

We described the recruitment process more detailed and we present an extra table with the description of sample characteristics broken down by recruitment setting.

Reviewer

We need more information on the format of data collection – was it self-complete questionnaires, was it by face-to-face interview? Was it conducted by clinic staff or by the researchers etc.

Authors' reply

We described the data collection more detailed in the method section.

Reviewer

Delay in psychiatric treatment – notes that delay is calculated based on “the year in which a psychiatric treatment for depressive symptoms was used for the first time”. In table 1 delay is reported as ‘months untreated’ – need to explain how those months were calculated i.e. from January in the year treatment was initiated?

Authors' reply

We asked patient in which year they had depressive symptoms for the first time and in which year they had their first professional treatment contact and computed the delay of treatment by calculating the difference between both years. We agree with the evaluator that the use of months as a unit of treatment delay suggests a higher measurement accuracy than was actually implemented. We therefore describe the assessment of the variable more detailed and changed the unit of measurement into years.

Reviewer

Also, if receiving treatment for burnout is an eligibility criterion is it possible that some of the men might never have been treated for depression?

Authors' reply

See above.

Reviewer

Overall, we need more clarity around the diagnostic status of the sample at recruitment. This also needs to be reported in Table 1.

Authors' reply

We reported the clinical status of the study participants by means of the PHQ-9 cut of values for the severity of depression in table 1.

Reviewer

Results.

It would be good to see descriptive statistics for the various measures in Table 1 as well. And as mentioned above, diagnostic status (at minimum if they are being treated for depression or for burnout’.

Authors' reply

In addition to the PHQ-9 cut-off values we reported the means and standard deviations of all PHQ subscales broken down by latent-class in table 2.

reviewer

The text describing figure 5 appears to have errors – the second sentence compares class 2 to class 2. There are also some spelling errors in that paragraph.

Authors' reply

We are sorry for this error and corrected the formulation.

Reviewer

Discussion

The discussion would benefit from being more focused on the question that the paper describes in the title and abstract – the relationship between masculine norms and work role and the delaying of treatment seeking, and the putative role of stigma in that relationship. The focus is predominantly on differences in adherence to traditional masculinity and occupation role orientations in a depressed/burnout sample, and the authors have not made a strong argument about why this is important to study.

Authors' reply

We agree that our study is mainly focused on the relationship of masculinity and work-role orientations with the psychological well-being of depressed patients. We therefore tried to underline the clinical relevance of these associations for depression therapy in the introduction and in the discussion.

Reviewer

Much of the discussion then describes the relationship of masculinity and work role orientation which is not relevant to the original research question. Moreover the assumption made “that traditional masculine role orientation is not harmful per se to mental health, but depends on its functional and contextual meaning in the work environment” cannot be supported by the study as the sample is comprised of only men with depression.

Authors' reply

We agree that our study design and the study sample does not allow general conclusions about the effects of masculinity norms and work-role orientations on the etiology of depressive symptoms and we therefore focused our conclusion to the population of men with depression which is represented by our sample

Reviewer 2

Reviewer #2: Overall, the paper paper is focused on an important gap in the literature and makes a contribution to the field. However, there were some limitations which must be seriously taken into account. These include generalizability of the findings without more information about recruitment process and sample, types of occupations of the participants, as well as the recall bias that may be affecting one of the main outcomes (delay in treatment). The study can only provide information on men who ultimately seek care, and there may be a much wider pool of men who never seek treatment at all, who may not be represented in these results and this should be acknowledged.

Authors' reply

We addressed the limitation of generalizability in the limitation section.

Reviewer

There is some convoluted writing and significant typos. It may be useful to have the paper copy edited (perhaps professionally) for standard English when revising the paper.

Authors' reply

We have completely revised the language of the manuscript, including professional editing

Reviewer

Abstract:

• More detail is needed on the methods. How were patients recruited? Convenience sample? What measures were used? Are they validated measures? Was the study quantitative or qualitative? What was the participation rate? Some of this information can be provided solely in the manuscript, but more detail is needed in the abstract as well.

Authors' reply

We described the study methods in the abstract more detailed.

Reviewer

• Last sentence of the conclusions should read: “in the development of strategies to reduce the delay…and to tailor the psychiatric treatment…”

Authors' reply

We thank the reviewer for this advice and made the recommended changes.

Reviewer

Introduction:

• The introduction needs to be more focused on the importance of the work environment earlier. As the introduction currently reads, it is not clear immediately that the focus will be occupational ‘identity’ or norms.

Authors' reply

We reformulated the introduction to make the focus on masculinity and work-role orientations more clear.

reviewer

• Page 3, last sentence of the first paragraph: “negatively associated with depressive symptoms” is unclear. Can say, “was associated with lower depressive symptoms,” or other language to make the association clearer.

Authors' reply

We thank the reviewer for this advice and made the recommended changes.

Reviewer

• Page 4, first sentence: Please specify what the ‘control populations’ were. Individuals without depression?

Authors' reply

We added the missing information about the control population.

Reviewer

Methods:

• The fact that this is a mixed-methods study needs to be in the abstract.

Authors' reply

Since the manuscript provides only results from the quantitative part of the study, we followed the recommendation of reviewer 1 and deleted the note on the mixed-method character of the study from the manuscript.

Reviewer

• Please specify whether the measures were validated measures from the very beginning (first paragraph of study design section).

Authors' reply

Done.

Reviewer

• There are multiple typos in the methods section.

Authors' reply

We regret the negligence and have tried to correct all mistakes.

Reviewer

• When you say that a ‘sub-sample of study participants were selected on the bases if the results’ please specify what the basis of this decision was. Will help the reader understand generalizability of the results. Who was chosen to be interviewed?

Authors' reply

For only quantitative data were used in this manuscript and the qualitative part of the study had no influence on the quantitative results, we followed the recommendation of reviewer 1 and deleted all notes on the qualitative part of the study from this manuscript.

Reviewer

• Why was this age rage chosen? Please justify the rationale. Was the idea to choose only working-age men? If so, please specify.

Authors' reply

Yes, because our particular interest in work-role orientations we included only men below the usual age of retirement, which in Germany is 65 years.

Reviewer

• For some readers, such as myself, we are unfamiliar with the term ‘psychosomatic hospitals.’ Please provide a bit more background on these.

Authors' reply

We specified the difference between psychiatric and psychosomatic hospitals in the method section.

Reviewer

• The paper should specify how many patients came from each of the different settings, as again, this will affect generalizability.

Authors' reply

We presented detailed information about the recruitment in table 1.

Reviewer

• Please provide more info on the recruitment process. Were patients asked to participate if they were having depressive symptoms, if they were in treatment for depression, etc. What did the recruitment messaging say?

Authors' reply

We presented a more detailed description of the recruitment process and the press invitation in the method section.

Reviewer

• It says that eligible patients were contacted by doctors or clinical staff about willingness to participate. Was this the case even for participants who came in via the press release?

Authors' reply

We added the information that patients who reported on the press release were directly contacted by the study workers.

Reviewer

• Cronbach’s alpha numbers are part of results, not methods. Please move these to the results section.

Authors' reply

Cronbach’s Alpha computed with our data are now presented in the result section.

Reviewer

• Please provide the full name for all acronyms (such as AVEM on page 6).

Authors' reply

Done.

Reviewer

• Some questions were added to validated scales. This would be unvalidated measures unless validated elsewhere. Please specify if that was the case.

Authors' reply

We computed Cronbach’s Alpha for all measures used in the study. The psychometric properties of the DSS subscales with the additional items were better than those reported in the literature.

Reviewer

• For ‘delay of treatment,’ which is an important outcome in this study, please provide further information on how this was assessed. There seems to be a lot of room for recall bias, which is inherently problematic and should be acknowledged in the limitations section.

Authors' reply

We agree that recall bias is a problem and we addressed this in the limitation section.

Reviewer

• The information on the LPA could potentially be included in an appendix if allowed, rather than in the body of the paper.

Authors' reply

We would like to leave the information in the methods section because the use of LPA is not so common in medical science.

Reviewer

• ‘one-way analyses of variance’= ANOVA and should be specified.

Authors' reply

In order to reduce the amount of results presented, we no longer refer to ANOVA results.

Reviewer

Results:

• Many of the covariates were not described in the methods. How is ‘higher education’ defined? College degree? More than college degree? Duration of illness is lacking units. Is this years, months, etc. Why was the cut-off of >3000 euros used? Please provide the rationale for this decision.

Authors' reply

We provided more detailed information on covariates in the statistical method section.

The 3000 € cut-off was used because it is the average net household income in Germany at the year of the study was about 3.400 € and the mean and the median net household income category in our study were 2000-3000 €

Reviewer

• Page 13, last sentence: the sentence structure is very convoluted, which is obscuring the meaning of the sentence. Do you mean, ‘participants in latent class 1 were less likely to have a partner…than participants in latent class 3’?

Authors' reply

We have reformulated the sentence

Reviewer

• Page 15, last paragraph, second sentence: class 2 is compared to class 2.

Authors' reply

We corrected this error

Reviewer

• The authors seem to be presenting an overwhelming amount of results that are hard to interpret. I would consider trying to figure out what the main results and take-away findings are and moving all others to an appendix section.

Authors' reply

We agree that the presented results are in part redundant. However, as a consequence of our decision to exclude participants who were treated for burnout and not for depression we also decided to present only the results for the PHQ scales and delete the results related to symptoms of burnout.

Reviewer

• Many typos throughout the results section.

Authors' reply

We did a new language editing of the whole manuscript

Reviewer

Discussion:

• The first sentence of the discussion section was not well set up in the introduction. It was unclear that this was the hypothesis that the authors were challenging.

Authors' reply

We reformulated the introduction and also the first sentence of the discussion

Reviewer

• The bullets in the discussion section are more re-hashing of the results and are actually much clearer summaries than the ones provided in the actual results section.

Authors' reply

We regard the summary of the results of the latent profile analysis as useful for helping the reader to follow the discussion.

Reviewer

• ‘Sex’ and ‘gender’ are used interchangeably throughout the paper, when the authors are in fact referring to ‘gender’ only throughout.

Authors' reply

We changed “sex” into “gender” if appropriate

Reviewer

• Page 19: there should be something in the discussion about how the info from the study could be used in practice.

Authors' reply

We discussed this point more detailed

Reviewer

Limitations section:

• Self-selection of participants is an important limitation, as it limits the generalizability of results. Can only speak for patients IN treatment who were willing to participate in the study. Cannot speak for men who do not seek treatment at all or who are unwilling to participate in the study. Generalizability should be listed as a limitation specifically.

• It is also a limitation that delay is all self-report, which is highly subjected to recall bias. Unclear how severity of illness, duration of treatment, etc may impact recall of treatment delay.

Authors' reply

We discussed these limitations in detail in the limitation section

Reviewer

• A major limitation is that type of occupation is not included in the study results. What type of work did the men included in the study do? Blue collar vs white collar? This seems like it could strongly affect or influence the findings.

Authors' reply

We included blue collar vs. white collar work as a control variable

reviewer

Conclusions:

• Can unfortunately not speak about the ‘treatment process,’ only ‘seeking treatment’ or ‘access’ to treatment.

• The last sentence, while true, seems unrelated to the study. More could be said specifically about the impact or use of the findings.

Authors' reply

We reformulated the conclusion to keep closer to the presented results

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: response to reviewers.docx

Click here for additional data file.

1 Apr 2020

PONE-D-19-27414R1

Masculinity norms and occupational role orientation in men treated for depression.

PLOS ONE

Dear Mr. Kilian,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by April 30, 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Stephan Doering, M.D.

Academic Editor

PLOS ONE

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: Overall:

An important topic, important to help improve services to this high-risk group for suicide. Improved from last version of the manuscript. However, the manuscript still feels long, with redundancies between tables and text and could be streamlined further. There is a tendency to assume causation rather than correlation, so this should be reviewed and changed accordingly throughout the manuscript. An important contribution to the paper, but still needs revisions.

Abstract:

-there are a few typos throughout

-the conclusions should be revised to reflect the findings of the paper. While the conclusions bring up an important point, they are not directly related to the study findings. They assume causation, rather than correlation, between traditional masculinities and adequate coping strategies.

Introduction:

-pg. 4: overall the first paragraph is hard to follow. ‘Positive’ and ‘negative’ are sometimes used to refer to the direction of the association and sometimes to quality of emotions. For example, the paragraph uses ‘positive health’, ‘positive depressive symptoms’ and ‘positively related.’ Ways to steer from these words, or to use them consistently, would be useful.

-pg. 4, 2nd paragraph: ‘the question remains how this knowledge can be used to prevent or treat depressive illness in men.’ While this is true, this is not addressed in the current study. This research is not addressing that issue, and this should either be moved to the discussion or removed.

-pg 5: similarly, ‘the etiology and the course of male depression.’ This study only looks at correlations, not etiology or causation. Additionally, ‘it is currently unclear how to identify and how to deal with noxious masculinity orientations in the treatment of depression.’ While this is true, again, this will not be addressed with the current study. This is not an aim that this research can address and should be in discussion as next steps.

Methods:

-pg. 6: study design. What does ‘treatment’ for depression entail? Do you have data on men treated with meds vs therapy vs both? This could speak to severity of symptoms.

-pg 7: there is a distinction drawn between psychiatric and psychosomatic hospitals. Are these different or the same thing?

-pg. 9: what does the acronym ‘AVEM’ stand for?

-pg. 10: good clarification of duration of untreated illness from prior draft

-pg. 11, I worry that the analyses are overcontrolling by controlling for both household income and blue-collar work. Similarly, by controlling for recruitment setting, I wonder if this controlling for severity of symptoms as well. Were there major differences controlling and not controlling for those factors in the models?

-pg 12: anxiety (PHQ-7)= anxiety (GAD-7), not PHQ-7.

-pg. 12: similar worry about overadjustment with unemployment and income and blue-collar work all included in the model. Duration of mental illness should be ‘reported duration’

Results:

-pg. 13: of the individuals who responded via press invitation, where did they receive their treatment for depression?

-there are inconsistencies in rounding throughout the paper. Sometimes ‘59%’, other times 37.2%. Rounding should be consistent throughout the paper based on journal guidelines.

-pg. 13: I am not sure what ‘Abitur’ is. Please explain for an international audience.

-can psychometric properties of the study instruments in the current sample be moved to an appendix to shorten the paper which currently feels too long?

-I wonder if the class differences (pg 16) could be summarized in a table for ease of interpretation. A simple 3 x 3 table of the 3 classes, and ‘status’ ‘toughness’ and ‘anti-femininity.’

-pg. 18 and on, the RR and p-values are already listed in the table, so I would remove them in the text for clarity.

-pg. 19: ‘class 3 have higher symptoms of depression…” Should this be ‘lower’?

-there are many mistakes in the text compared to the tables. It points to a lack of attention to detail.

-there is a lot of repetition between the tables and the text. Can cut out one or the other to shorten the manuscript.

-figure 3 could be included as a supplement, rather than in the main body of the manuscript.

-can summarize the path analyses, starting in pg 21, as the findings were very similar to those presented previously.

-I would structure the findings similarly, so that if you are comparing class 1 to class 3, you then compare class 2 to class 3 (as opposed to class 3 to class 2). Similar structure will help in the interpretation of so many results.

-last 3 paragraphs of the results are repetitive and can be removed or moved to a supplement

Discussion:

-the summaries included in the discussion are good, but better belong in the results than in the discussion.

-the paragraph that begins with ‘Nevertheless, our findings’, pg. 23, should be the first paragraph of the discussion and should begin with ‘Our findings.’ The prior paragraph includes important information that seems to be more appropriate in a ‘limitations’ section. You should begin with a discussion of findings, putting them in context, then include the limitations of the current study.

-some of the discussion points are assuming causation, rather than correlation of the study findings. For example, pg. 24, 1st paragraph, ‘our data suggest that this relief may come with the price of losing crucial emotional rewards.’ Can’t draw this conclusion from the current study design, can’t assume cause and effect.

Limitations:

-memory bias, may be better termed as ‘recall bias’ which is widely used in the literature.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

24 Apr 2020

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

6. Review Comments to the Author

Reviewer #1: (No Response)

Reviewer #2: Overall:

An important topic, important to help improve services to this high-risk group for suicide. Improved from last version of the manuscript. However, the manuscript still feels long, with redundancies between tables and text and could be streamlined further. There is a tendency to assume causation rather than correlation, so this should be reviewed and changed accordingly throughout the manuscript. An important contribution to the paper, but still needs revisions.

Authors' reply

• We thank the reviewer for the general positive assessment and advice for the further improvement of the paper.

Abstract:

-there are a few typos throughout

-the conclusions should be revised to reflect the findings of the paper. While the conclusions bring up an important point, they are not directly related to the study findings. They assume causation, rather than correlation, between traditional masculinities and adequate coping strategies.

Authors' reply

• We apologize for the typos and corrected them throughout the text.

• We revised the conclusion section of the abstract as advised

Introduction:

-pg. 4: overall the first paragraph is hard to follow. ‘Positive’ and ‘negative’ are sometimes used to refer to the direction of the association and sometimes to quality of emotions. For example, the paragraph uses ‘positive health’, ‘positive depressive symptoms’ and ‘positively related.’ Ways to steer from these words, or to use them consistently, would be useful.

-pg. 4, 2nd paragraph: ‘the question remains how this knowledge can be used to prevent or treat depressive illness in men.’ While this is true, this is not addressed in the current study. This research is not addressing that issue, and this should either be moved to the discussion or removed.

-pg 5: similarly, ‘the etiology and the course of male depression.’ This study only looks at correlations, not etiology or causation. Additionally, ‘it is currently unclear how to identify and how to deal with noxious masculinity orientations in the treatment of depression.’ While this is true, again, this will not be addressed with the current study. This is not an aim that this research can address and should be in discussion as next steps.

Authors' reply

• We thank the reviewer for these comments and revised the paragraphs as advised.

Methods:

-pg. 6: study design. What does ‘treatment’ for depression entail? Do you have data on men treated with meds vs therapy vs both? This could speak to severity of symptoms.

Authors' reply

• We did not ask our participants after the type of treatment they received. However, we assessed the severity of depressive symptoms by means of the PHQ-9 cut-off scores which are provided in Table 1.

• We added the information in the method section, that we included patients treated with medications or with psychotherapy or both.

-pg 7: there is a distinction drawn between psychiatric and psychosomatic hospitals. Are these different or the same thing?

Authors' reply

• We tried to improve our explanation of the differences between psychiatric and psychosomatic hospitals by the following formulation at pg 7.

-pg. 9: what does the acronym ‘AVEM’ stand for?

Authors' reply

• We added the German meaning of the acronym at page 9.

-pg. 10: good clarification of duration of untreated illness from prior draft

Authors' reply

Thanks.

-pg. 11, I worry that the analyses are overcontrolling by controlling for both household income and blue-collar work. Similarly, by controlling for recruitment setting, I wonder if this controlling for severity of symptoms as well. Were there major differences controlling and not controlling for those factors in the models?

Authors' reply

• We thank the reviewer for this important comment. We checked for multicollinearity by estimating the variance inflation factor (VIF) for each model variable and the average VIF for each regression model, revealing no indication for multicollinearity. In addition, we compared the adjusted results with the unadjusted results, indicating that the regression coefficients and the estimated means are quite similar and that the interpretation of the results remains the same after adjustment. Nevertheless, we think that adjustment for the socioeconomic variables is reasonable because it provides the effects of masculinity and job related attitudes independent of the confounding effects of economic conditions, education and the work environment. With regard to the inconclusive results we found in the literature, this seems important.

-pg 12: anxiety (PHQ-7)= anxiety (GAD-7), not PHQ-7.

Authors' reply

• We apologize for these mistakes and corrected them.

-pg. 12: similar worry about overadjustment with unemployment and income and blue-collar work all included in the model. Duration of mental illness should be ‘reported duration’

Authors' reply

• Regarding overadjustment, see above.

• We changed DUI into “reported DUI”.

Results:

-pg. 13: of the individuals who responded via press invitation, where did they receive their treatment for depression?

Authors' reply

• We provided this information at pg 14.

-there are inconsistencies in rounding throughout the paper. Sometimes ‘59%’, other times 37.2%. Rounding should be consistent throughout the paper based on journal guidelines.

Authors' reply

• We corrected the inconsistencies.

-pg. 13: I am not sure what ‘Abitur’ is. Please explain for an international audience.

Authors' reply

• We explained the term in the method section at p. 11.

-can psychometric properties of the study instruments in the current sample be moved to an appendix to shorten the paper which currently feels too long?

Authors' reply

• We moved this part of the results into the appendix.

-I wonder if the class differences (pg 16) could be summarized in a table for ease of interpretation. A simple 3 x 3 table of the 3 classes, and ‘status’ ‘toughness’ and ‘anti-femininity.’

Authors' reply

• Adding a table with the class differences would lead to a doubble display of the information already presented in figure 1. However, we followed this recommendation and added a new table S2, providing the means, the standard deviations and the results of the ANOVAs for the dimensions of the masculinity and the job-related attitudes to the online supplement.

-pg. 18 and on, the RR and p-values are already listed in the table, so I would remove them in the text for clarity.

Authors' reply

• We followed this recommendation and removed the RRs from the text.

-pg. 19: ‘class 3 have higher symptoms of depression…” Should this be ‘lower’?

-there are many mistakes in the text compared to the tables. It points to a lack of attention to detail.

Authors' reply

• We corrected these mistakes.

-there is a lot of repetition between the tables and the text. Can cut out one or the other to shorten the manuscript.

-figure 3 could be included as a supplement, rather than in the main body of the manuscript.

Authors' reply

• We followed the advice and moved Table 4 and figure 3 into the supplement.

-can summarize the path analyses, starting in pg 21, as the findings were very similar to those presented previously.

-I would structure the findings similarly, so that if you are comparing class 1 to class 3, you then compare class 2 to class 3 (as opposed to class 3 to class 2). Similar structure will help in the interpretation of so many results.

Authors' reply

• We followed the advice and changed the order of the comparison.

-last 3 paragraphs of the results are repetitive and can be removed or moved to a supplement

Authors' reply

• The last three paragraphs provide information about indirect effects and about the model fit of the path model. This information is needed for the interpretation of the path analysis. Therefore we want to leave this information in the main text. However, we combined the last two paragraphs into one.

Discussion:

-the summaries included in the discussion are good, but better belong in the results than in the discussion.

Authors' reply

• We followed the advice and moved the summary of the LPA results into the results section.

-the paragraph that begins with ‘Nevertheless, our findings’, pg. 23, should be the first paragraph of the discussion and should begin with ‘Our findings.’ The prior paragraph includes important information that seems to be more appropriate in a ‘limitations’ section. You should begin with a discussion of findings, putting them in context, then include the limitations of the current study.

Authors' reply

• We followed the advices and restructured the discussion.

-some of the discussion points are assuming causation, rather than correlation of the study findings. For example, pg. 24, 1st paragraph, ‘our data suggest that this relief may come with the price of losing crucial emotional rewards.’ Can’t draw this conclusion from the current study design, can’t assume cause and effect.

Authors' reply

• We thank the reviewer for this important advice and reformulated this conclusion into a hypothesis to be investigated in further studies.

Limitations:

-memory bias, may be better termed as ‘recall bias’ which is widely used in the literature.

Authors' reply

• We gladly follow this advice.

13 May 2020

Masculinity norms and occupational role orientations in men treated for depression.

PONE-D-19-27414R2

Dear Dr. Kilian,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Stephan Doering, M.D.

Academic Editor

PLOS ONE

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #3: The criticism of the previous reviewers has been addressed comprehensively and sufficiently. The manuscript can be accepted as it is.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

15 May 2020

PONE-D-19-27414R2

Masculinity norms and occupational role orientations in men treated for depression.

Dear Dr. Kilian:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Stephan Doering

Academic Editor

PLOS ONE