Gender differences in attitudes toward mental illness: a cross-sectional study among the Saudi community in Jizan, Saudi Arabia

Abstract

Background

Despite recent improvements in attitudes toward mental illness, persistent stigma in some communities continues to negatively impact those affected. This study aimed to assess gender differences in attitudes toward mental illness among the Saudi community in Jizan.

Methods

A cross-sectional study was conducted between January and July 2024, enrolling 586 participants (267 males and 319 females) recruited using a non-probability convenience sampling from public places. Attitudes toward mental illness were assessed using the 40-item Community Attitudes toward Mental Illness (CAMI) scale, with negatively worded items reverse-coded. The 25th, 50th and 75th percentiles of the scale score were considered as cutoff points to describe low, medium, and high scores, respectively. Gender differences in the CAMI total score and its predefined subscales (Authoritarianism, Benevolence, Social Restrictiveness, and Community Mental Health Ideology) were examined using independent sample t-tests. Associations between demographic factors and attitudes toward mental illness were evaluated using multivariate linear regression (MLR), with regression models incorporating interaction terms to examine gender differences.

Results

The mean CAMI score for the participants was 107.16 ± 9.38, with males scoring significantly higher than females (108.64 ± 9.52 vs. 105.92 ± 9.10, p < 0.001 CI:1.20, 4.23), with a small effect size (Cohen’s d = 0.29, p < 0.001). Based on descriptive percentile-based cutoffs (≤ 101, 102–113, ≥ 114), more than half of the participants were classified as having moderate stigma (50.68%), with smaller proportions categorized as low (27.13%) and high (22.18%). Female participants showed a significantly more benevolent attitude toward individuals with mental illness than males (39.57 ± 4.61 vs. 38.72 ± 4.96, p = 0.033, CI: ‑1.62, -0.07), while males exhibited stronger social restrictiveness than females (25.74 ± 5.88 vs. 23.88 ± 5.24, p < 0.001, CI: 0.94, 2.72). No significant gender differences were observed in authoritarianism and community mental health ideology attitudes. Older and higher-educated males showed more negative attitudes, whereas being unmarried, unemployed, and having a low monthly income were significantly associated with the stigmatizing attitudes among female participants.

Conclusions

This exploratory study revealed gender disparities in attitudes toward mental illness, underscoring the necessity of incorporating gender considerations in mental health awareness initiatives. Future research should focus on selected mental health conditions and targeted demographics to maximize intervention effectiveness. Future confirmatory studies with representative samples that focus on specific psychiatric disorders across populations with diverse characteristics are needed to enhance the effectiveness of anti-stigma interventions.

Introduction

Mental illness represents a major global public health challenge, accounting for one in every six years lived with disability (15.6%) and contributing approximately 5.1% of the overall global disease burden [1]. In the Saudi National Mental Health Survey, 20.2% of the 4,004 respondents aged 15 to 65 reported experiencing a mental health condition at some point within the 12 months preceding the survey [2]. Despite the high prevalence of mental disorders in the Arab world, the proportion of individuals receiving mental health services remains below the global average. This discrepancy is partly due to the stigmatizing religious perception that mentally ill individuals lack faith, which often results in the somatization of psychological symptoms and a preference for seeking informal support from religious leaders [3]. Both global and local data indicate that many patients with mental illness identify negative attitudes and stigma as primary barriers to mental health-seeking behavior [1, 4].

Mental health stigma refers to the unfavorable stereotypes, beliefs, and attitudes that are held towards individuals diagnosed with mental health conditions [5]. The labeling of mental illness represents a substantial public health concern, with far-reaching implications for the affected patients, their caregivers, and the healthcare system professionals and institutions [6]. Stigmatization of mental illness occurs on three levels: intrapersonal, interpersonal, and structural. At the intrapersonal level, self-stigma leads to reduced self-esteem and reluctance to seek treatment. Interpersonally, public stigma leads to discrimination and rejection, impacting social integration. Finally, at the structural level, systemic discrimination obstructs access to mental health services and treatment for individuals affected by mental illness [5–8].

Identifying public attitudes toward mental illness is essential for advancing the WHO action plan-integrated mental health and social care in community-based settings, as well as for achieving the objectives of the Saudi Vision 2030, which aims to ensure a long, healthy, and productive life for everyone in the Kingdom [9]. Although community attitudes toward mental illness have improved substantially in recent years, research indicates that stigmatizing perceptions persist among many individuals, and the demographic factors influencing such ideologies remain unclear [10, 11]. Stigma related to mental illness often manifests in a gender-specific manner. Stereotypes surrounding individuals with mental illness frequently emphasize violence and dependency, with males stereotyped as violent and females as dependent [12]. In cultural contexts where gender roles are explicitly delineated, mentally ill individuals encounter intensified stigma when they violate gender-specific role expectations. For instance, schizophrenia disorder may be associated with stigma related to employment and economic productivity for men and marriage, motherhood, and household responsibilities for women [13].

Population-based studies have consistently documented significant gender differences in attitudes toward mental illness based on the Community Attitudes toward the Mentally Ill (CAMI) scale and its four subscales (Authoritarianism, Benevolence, Social Restrictiveness, and Community Mental Health Ideology), yet theoretically informed interpretations of these disparities remain poorly reported [14–19]. According to the Social Role Theory, gender differences in attitudes arise from culturally prescribed roles. Women have traditionally been socialized into caregiving roles that emphasize empathy, nurturance, and concern for others. On the other hand, the Social Dominance Theory suggests that men, as members of traditionally dominant social groups, are expected to have more hierarchical and controlling-oriented attitudes, expressed through increased authoritarianism and social restrictiveness toward socially marginalized groups [20].

Global research using the CAMI scale across diverse cultural contexts has observed similar gender patterns, with women consistently showing fewer stigmatizing attitudes than men. Specifically, survey-based studies from Saudi Arabia, Lebanon, Korea, Spain, and Serbia found that males showed significantly more authoritarian and socially restrictive attitudes toward mentally ill individuals compared to their female counterparts [14–18]. Females generally demonstrate lower levels of mental illness stigma than males, as evidenced by their greater benevolence, more accepting attitudes, and higher levels of support for individuals with mental illness [21, 22]. Nevertheless, gender differences in mental illness stigma are not unidirectional. While females often exhibit more sympathy and support for those with mental illnesses, they may simultaneously experience higher levels of fear and anxiety when interacting with them, and they more strongly endorse beliefs about social restriction and isolation, indicating that stigma varies across multiple attitudinal dimensions [12, 15, 22].

Alleviating stigma surrounding mental illness requires transformation across multiple dimensions, with research serving as a critical initial focus. There is a notable scarcity of published mental health studies in the Arab world, particularly within Saudi Arabia. This study aims to investigate gender differences in attitudes toward mental illness as measured by the CAMI scale. Based on patterns reported in prior research and within the context of established theories, this study hypothesizes that gender differences would be observed across CAMI dimensions (benevolence, authoritarianism, social restrictiveness, and community mental health ideology). To our knowledge, no prior study in Saudi Arabia has specifically addressed this topic. The findings of this study are expected to deepen understanding of mental health stigma in the Saudi context and to identify key areas for targeted anti-stigma initiatives.

Methods

Study design and participants

An analytical community-based cross-sectional study was conducted between January and July 2024, targeting the Saudi community in Jizan, a city and the capital of Jazan Province in the southwest of Saudi Arabia. Individuals under 18 years of age, those who were illiterate, unable to respond, or had a history of psychological conditions were excluded from the study.

Sample size and sampling procedure

The required sample size for comparing two independent means was calculated using the Epi-Info version 7. The calculation assumed a two-sided confidence level of 95%, a statistical power of 80%, and an equal allocation ratio of 1:1 between the two groups. The calculation was based on gender-specific mean scores of the CAMI subscale (SR) reported in a previous study conducted among an Arab population [14]. The predictable mean difference was − 1.31 (33.74 vs. 35.05), with corresponding standard deviations of 5.54 and 5.41 for males and females, respectively. Accordingly, the minimum required sample size was 550 participants (275 per gender group). Participants were recruited through a non-probability convenience sampling at various public venues in Jizan, including shopping centers, Jazan University, walkways, parks, and coffee shops.

Data collection tools

The data were collected using an Arabic self-administered questionnaire that required approximately 8 to 10 min to complete. The questionnaire included a total of 47 items distributed into two sections. The first section included seven items designed to capture participants’ socio-economic and demographic characteristics, including age, gender, education level, occupation, monthly income, marital status, and whether they knew anyone with mental illness.

The second section assessed participants’ attitudes towards mental illness using the full 40-item version of the Community Attitudes Toward the Mentally Ill (CAMI) developed by Taylor and Dear in 1981 [23]. The CAMI scale consists of four subscales, each containing 10 items that were rated on a five-point Likert scale (1 = strongly agree to 5 = strongly disagree). Negatively worded items were reverse-coded before analysis and are marked by a superscript R in the item-level results in Table 2. The subscales are: the Authoritarianism (AU) subscale which refers to the belief that a mentally ill person is inferior and requires supervision and coercion; the Benevolence (BE) subscale, which corresponds to ‘a humanistic and sympathetic view of mentally ill persons; the Social restrictiveness (SR) subscale, and it points to the belief that mentally ill patients are a threat to society and should be avoided, and the Community Mental Health Ideology (CMHI) subscale, which measures the acceptance of mental health services and the integration of mentally ill patients in the community. The CAMI scale is a globally validated tool for assessing social stigma towards mentally ill individuals [19]. The Arabic version of the CAMI scale used in this study was adopted from a published Lebanese study (permission was obtained from the author), in which the tool underwent content validation and reliability testing. The Arabic CAMI demonstrated satisfactory internal consistency reliability, with reported Cronbach’s alpha coefficients of 0.876 for the total scale and 0.555 for (AU), 0.637 for (BE), 0.690 for (SR), and 0.804 for (CMHI) [14]. The same content-validated Arabic version has been utilized in a study conducted in Saudi Arabia, supporting its applicability in the Saudi community context, with a reported reliability coefficient of 0.75 [24]. The total score stigma towards mental illness was computed by summing up the subscales, as greater stigma was characterized by higher AU and SR scores alongside lower BE and CMHI scores. It demonstrated acceptable internal consistency with Cronbach’s alpha estimated at 0.84 for the total CAMI scale score (40 items), 0.637 for BE, 0.69 for SR, and 0.60 for the CMHI subscale. The (AU) subscale exhibited modest internal consistency (α = 0.56), consistent with an estimate reported in a previous study [14]. Future research may further evaluate the psychometric properties of the CAMI dimensions in diverse Arab populations. The total score of the CAMI scale was categorized into three levels: low, medium, and high based on the cutoff points of the 25th, 50th, and 75th percentiles. These categories were used for descriptive purposes and do not represent validated thresholds, consistent with approaches used in prior CAMI-based research [14]. The adopted Arabic version of the scale was pilot-tested with a sample of 30 people to verify its clarity before conducting the data collection.

Table 2.

Gender differences in responses to the CAMI scale (N = 586)

Item		Strongly agree/agree n (%)	Neutral n (%)	Disagree/strongly disagree n (%)	χ 2	P-value	q (BH-FDR)
Authoritarianism (AU)
1r. As soon as a person shows signs of mental disturbance, he should be hospitalized.	Male	127 (47.6)	47 (17.6)	93 (34.8)	4.67	0.097	0.300
	Female	130 (40.8)	50 (15.7)	139 (43.5)
2. Mental illness is an illness like any other.	Male	180 (67.4)	23 (8.6)	64 (24.0)	3.49	0.174	0.435
	Female	206 (64.6)	43 (13.5)	70 (21.9)
3r. There is something about the mentally ill that makes it easy to tell them from normal people.	Male	120 (44.9)	63 (23.6)	84 (31.5)	0.54	0.761	0.828
	Female	153 (48.0)	70 (21.9)	96 (30.1)
4. Less emphasis should be placed on protecting the public from the mentally ill.	Male	129 (48.3)	64 (24.0)	74 (27.7)	1.42	0.49	0.700
	Female	168 (52.6)	65 (20.4)	86 (27.0)
5r. Mental patients need the same kind of control and discipline as a young child.	Male	197 (73.8)	31 (11.6)	39 (14.6)	2.39	0.302	0.522
	Female	245 (76.8)	25 (7.8)	49 (15.4)
6. The mentally ill should not be treated as outcasts of society.	Male	231 (86.5)	15 (5.6)	21 (7.9)	0.70	0.705	0.805
	Female	276 (86.5)	14 (4.4)	29 (9.1)
7r. The best way to handle the mentally ill is to keep them behind locked doors.	Male	18 (6.7)	19 (7.1)	230 (86.1)	3.02	0.22	0.463
	Female	18 (5.6)	13 (4.1)	288 (90.3)
8. Mental hospitals are an out-dated means of treating the mentally ill.	Male	101 (37.8)	74 (27.7)	92 (34.5)	1.64	0.439	0.651
	Female	113 (35.4)	104 (32.6)	102 (32.0)
9r. One of the main causes of mental illness is a lack of self-discipline and will power.	Male	177 (66.3)	63 (23.6)	27 (10.1)	2.87	0.238	0.475
	Female	198 (62.1)	74 (23.2)	47 (14.7)
10. Virtually anyone can become mentally ill.	Male	211 (79.0)	32 (12.0)	24 (9.0)	1.67	0.432	0.651
	Female	263 (82.4)	28 (8.8)	28 (8.8)
Benevolence (BE)
11. More money should be spent on the care and treatment of the mentally ill.	Male	230 (86.1)	16 (6.0)	21 (7.9)	1.04	0.592	0.756
	Female	273 (85.6)	25 (7.8)	21 (6.6)
12r. The mentally ill are a burden on society.	Male	31 (11.6)	45 (16.9)	191 (71.5)	1.02	0.599	0.756
	Female	32 (10.0)	47 (14.7)	240 (75.3)
13. The mentally ill have for too long been the subject of ridicule.	Male	176 (65.9)	55 (20.6)	36 (13.5)	5.5	0.062	0.300
	Female	238 (74.6)	46 (14.4)	35 (11.0)
14r. Increased spending on mental health services is a waste of tax dollars.	Male	34 (12.7)	23 (8.6)	210 (78.7)	3.87	0.143	0.383
	Female	32 (10.0)	17 (5.3)	270 (84.6)
15. We need to adopt a far more tolerant attitude toward the mentally ill in our society.	Male	245 (91.7)	13 (4.9)	9 (3.4)	0.43	0.804	0.846
	Female	294 (92.2)	17 (5.3)	8 (2.5)
16r. There are sufficient existing services for the mentally ill.	Male	76 (28.5)	67 (25.1)	124 (46.4)	0.19	0.909	0.909
	Female	86 (27.0)	80 (25.0)	153 (48.0)
17. Our mental hospitals seem more like prisons than like places where the mentally ill can be cared for.	Male	108 (40.5)	73 (27.3)	86 (32.2)	1.09	0.577	0.756
	Female	142 (44.5)	78 (24.5)	99 (31.0)
18r. The mentally ill do not deserve our sympathy.	Male	20 (7.5)	17 (6.4)	230 (86.1)	2.65	0.265	0.505
	Female	14 (4.4)	19 (6.0)	286 (89.6)
19. We have the responsibility to provide the best possible care for the mentally ill.	Male	223 (83.5)	33 (12.4)	11 (4.1)	2.54	0.280	0.509
	Female	273 (85.6)	28 (8.8)	18 (5.6)
20r. It is best to avoid anyone who has mental problems.	Male	51 (19.1)	61 (22.8)	155 (58.1)	4.94	0.084	0.300
	Female	45 (14.1)	61 (19.1)	213 (66.8)
Social Restrictiveness (SR)
21r. The mentally ill should be isolated from the rest of the community.	Male	29 (10.9)	47 (17.6)	191 (71.5)	3.04	0.218	0.463
	Female	31 (9.7)	41 (12.9)	247 (77.4)
22. The mentally ill are far less of a danger than most people suppose.	Male	152 (56.9)	61 (22.8)	54 (20.3)	0.32	0.850	0.872
	Female	184 (57.7)	67 (21.0)	68 (21.3)
23r. A woman would be foolish to marry a man who has suffered from mental illness, even though he seems fully recovered.	Male	59 (22.1)	53 (19.9)	155 (58.0)	0.53	0.766	0.828
	Female	68 (21.3)	57 (17.9)	194 (60.8)
24. No one has the right to exclude the mentally ill from their neighbourhood.	Male	169 (63.3)	64 (24.0)	34 (12.7)	6.99	0.030*	0.242
	Female	231 (72.4)	50 (15.7)	38 (11.9)
25r. I would not want to live next door to someone who has been mentally ill.	Male	74 (27.7)	70 (26.2)	123 (46.1)	3.99	0.135	0.383
	Female	67 (21.0)	84 (26.3)	168 (52.7)
26. Mental patients should be encouraged to assume the responsibilities of normal life.	Male	212 (79.4)	39 (14.6)	16 (6.0)	4.70	0.095	0.300
	Female	267 (83.7)	44 (13.8)	8 (2.5)
27r. Anyone with a history of mental problems should be excluded from taking public office.	Male	146 (54.7)	61 (22.8)	60 (22.5)	17.84	0.0001*	0.005 a
	Female	120 (37.6)	90 (28.2)	109 (34.2)
28. The mentally ill should not be denied their individual rights.	Male	218 (81.6)	16 (6.0)	33 (12.4)	3.08	0.214	0.463
	Female	276 (86.5)	17 (5.3)	26 (8.2)
29r. The mentally ill should not be given any responsibility.	Male	111 (41.6)	69 (25.8)	87 (32.6)	9.87	0.007*	0.095
	Female	98 (30.7)	80 (25.1)	141 (44.2)
30. Most women who were once patients in a mental hospital can be trusted as babysitters.	Male	81 (30.3)	107 (40.1)	79 (29.6)	11.77	0.002*	0.055
	Female	140 (43.9)	97 (30.4)	82 (25.7)
Community Mental Health Ideology (CMHI)
31. The best therapy for many mental patients is to be part of a normal community.	Male	188 (70.4)	54 (20.2)	25 (9.4)	4.96	0.083	0.300
	Female	247 (77.4)	55 (17.2)	17 (5.4)
32r. Locating mental health facilities in a residential area downgrades the neighbourhood.	Male	82 (30.7)	45 (16.9)	140 (52.4)	4.65	0.097	0.300
	Female	77 (24.1)	47 (14.7)	195 (61.2)
33. As far as possible mental health services should be provided through community-based facilities.	Male	225 (84.3)	27 (10.1)	15 (5.6)	2.26	0.322	0.522
	Female	278 (87.1)	31 (9.7)	10 (3.2)
34r. Having mental patients living within residential neighbourhoods might be good therapy, but the risks to residents are too great.	Male	146 (54.7)	73 (27.3)	48 (18.0)	0.81	0.667	0.790
	Female	167 (52.4)	98 (30.7)	54 (16.9)
35. Residents should accept the location of mental health facilities in their neighbourhood to serve the needs of the local community.	Male	221 (82.8)	28 (10.5)	18 (6.7)	7.08	0.028 *	0.242
	Female	286 (89.7)	24 (7.5)	9 (2.8)
36r. Local residents have good reason to resist the location of mental health services in their neighbourhood.	Male	58 (21.7)	61 (22.8)	148 (55.5)	6.03	0.049 *	0.300
	Female	90 (28.2)	84 (26.3)	145 (45.5)
37. Locating mental health services in residential neighbourhoods does not endanger local residents.	Male	186 (69.7)	57 (21.3)	24 (9.0)	0.79	0.671	0.790
	Female	222 (69.6)	74 (23.2)	23 (7.2)
38r. Mental health facilities should be kept out of residential neighbourhoods.	Male	71 (26.6)	73 (27.3)	123 (46.1)	1.00	0.605	0.756
	Female	80 (25.1)	79 (24.8)	160 (50.1)
39. Residents have nothing to fear from people coming into their neighbourhood to obtain mental health services.	Male	191 (71.5)	59 (22.1)	17 (6.4)	5.32	0.069	0.300
	Female	250 (78.4)	47 (14.7)	22 (6.9)
40r. It is frightening to think of people with mental problems living in residential neighbourhoods.	Male	61 (22.8)	60 (22.5)	146 (54.7)	2.23	0.326	0.522
	Female	58 (18.2)	82 (25.7)	179 (56.1)

_r = Negatively worded item reverse-scored prior to analysis, p-values are from Pearson’s chi-square tests. An asterisk (*) indicates statistical significance at p < 0.05. q-values (superscript a) represent Benjamini–Hochberg false discovery rate (BH-FDR) adjusted p-values; q < 0.05 was considered statistically significant.

Statistical analysis

Data were analyzed using the Statistical Package for Social Sciences (SPSS, version 26). Descriptive statistics were conducted and presented to summarize the participants’ socio-demographic data. Pearson’s chi-square tests were employed to examine gender differences across individual CAMI items (n = 40). These item-level analyses were conducted on an exploratory basis. To control for multiple testing, p-values were adjusted using the Benjamini–Hochberg false discovery rate (FDR) method, and adjusted p-values (q-values) were reported. Gender differences in the CAMI total score and its predefined subscales (AU, BE, SR, and CMHI) were examined using independent-sample t-tests. This study was aimed to explore gender-specific patterns in attitudes toward mental illness. Hence, multiple regression analysis was applied as a single modelling strategy per outcome. Besides the CAMI total score, each of the four CAMI predefined subscales (AU, BE, SR, and CMHI) was analyzed separately as a dependent variable. In light of the study’s focus, parallel MLR models were conducted separately for males and females, using the same set of socio-demographic factors in each model to ensure conceptual clarity and allow for direct comparison of the associations between the demographic factors and the attitude outcomes across gender groups. A p-value of < 0.05 was considered statistically significant.

Results

A total of 586 eligible individuals were approached and included in the final analysis, exceeding the pre-determined minimum sample size for this study. Table 1 presents the socio-demographic profiles of the male and female groups. Participants’ ages ranged from 18 to 68 years, with a mean of 31.71 ± 10.89 years. The highest percentage of males were married (59.9%), whereas the highest percentage of females were single (54.5%). The majority of respondents in both groups were university graduates (61.8% of males and 65.2% of females) and predominantly resided in urban areas (61.8% of males and 57.1% of females). Regarding occupation, most males were employed (67.4%), while most females were unemployed (60.8%). More than half of the females (55.5%) reported a monthly income of less than 5,000 SR, compared to nearly one-third of males (36.3%) in the same category. Conversely, a higher proportion of males than females reported incomes above 10,000 SR (34.5% vs. 24.8%, respectively). More than half of both male and female respondents reported knowing someone with a mental illness (58.1% and 52%, respectively).

Table 1.

Socio-demographic characteristics of male and female participants in Jizan, Saudi Arabia, 2024 (N = 586)

Variable		Male n = 267(%)	Female n = 319 (%)	Total n = 586 (%)
Mean age (± SD)		33.88 (11.50)	29.89 (10.02)	31.71 (10.89)
Age (years)	18–34	156 (58.40)	216 (67.70)	372 (63.50)
	35–45	61 (22.80)	81 (25.40)	142 (24.20)
	Above 45	50 (18.70)	22 (6.90)	72 (12.30)
Marital Status	Single	103 (38.60)	174 (54.50)	277 (47.30)
	Married	160 (59.90)	125 (39.20)	285 (48.60)
	Divorced & widowed	4 (1.50)	20 (6.30)	24 (4.10)
Educational Level	School	91 (34.10)	107 (33.50)	198 (33.80)
	University Graduate	165 (61.80)	208 (65.20)	373 (63.70)
	Postgraduate	11 (4.10)	4 (1.30)	15 (2.60)
Employment Status	Employed	180 (67.40)	125 (39.20)	305 (52.0)
	Unemployed	87 (32.60)	194 (60.80)	278 (47.40)
Monthly Income	Less than 5000 SR	97 (36.30)	177 (55.5)	274 (46.70)
	5000 to 10,000 SR	78 (29.20)	63 (19.70)	141 (24.10)
	More than 10,000 SR	92 (34.50)	79 (24.80)	171 (29.20)
Place of Residence	Urban Area	166 (61.80)	182 (57.10)	348 (59.20)
	Rural Area	101 (38.20)	137 (42.90)	238 (41.80)
Know Person with Mental Illness	No	112 (41.90)	153 (48.0)	265 (45.20)
	Yes	155 (58.10)	166 (52.0)	321 (54.80)

Table 2 illustrates item-level gender differences in attitudes toward mental illness across the 40 items of the CAMI scale using Pearson chi-square tests, with responses collapsed as Agree/Strongly agree (1–2), Neutral (3), and Disagree/Strongly disagree (4–5). To control for multiple testing across 40 items, p-values were adjusted using the Benjamini–Hochberg false discovery rate (BH-FDR) procedure. The findings revealed statistically significant differences in six out of forty items (p < 0.05), indicating that while most attitudes were comparable between male and female participants, certain aspects differed. However, after BH-FDR correction, one item [27] remained significant.

Regarding the Social Restrictiveness (SR) subscale, a higher proportion of female participants (72.4%) than males (63.3%) agreed that no one has the right to exclude individuals with mental illness from their neighborhood (p = 0.030). Similarly, more females (43.9%) than males (30.3%) believed that most women who had been patients in a mental hospital could be trusted as babysitters (p = 0.003). Conversely, a greater percentage of males than females endorsed that mentally ill individuals should not be given any responsibility (41.6% vs. 30.7%, p = 0.007) and supported excluding individuals with a history of mental illness from taking public office (54.7% vs. 37.6%, p < 0.001). After BH-FDR correction, only item 27 showed a statistically significant association between gender and response category, χ² (2, N = 586) = 17.84, p = 0.00013, q = 0.00534, with a small-to-moderate effect size (Cramér’s V = 0.175). Within the Community Mental Health Ideology (CMHI) subscale, females were significantly more likely than males to agree that residents should accept the location of mental health facilities in their neighborhoods (89.7% vs. 82.8%, p = 0.029). However, females also more frequently agreed that local residents have valid reasons to resist the establishment of such facilities (28.2% vs. 21.7%, p = 0.049).

No significant gender differences were observed in the Authoritarianism (AU) subscale. Nevertheless, the overall responses reflected a complex mixture of supportive and stigmatizing attitudes. A majority of both male and female participants (88.4%) rejected the view that the best way to handle the mentally ill is to keep them behind locked doors.

However,75.4% believed that mental patients need the same kind of control and discipline as a young child. While 80.9% agreed that anyone can become mentally ill, 64% attributed mental illness primarily to a lack of self-discipline and willpower. Responses on the Benevolence (BE) subscale were generally more positive, with over 80% of participants supporting that more money should be spent on the care and treatment of the mentally ill individuals, the need to adopt a more tolerant attitude toward the mentally ill individuals in the society, and having the responsibility to provide the best possible care for the mentally ill.

Based on the percentile-based categorization of the total CAMI score as low (≤ 101), moderate (102–113), and high (≥ 114), over half of the participants were classified as having moderate stigma towards mental illness (n = 297, 50.68%), followed by low stigma (n = 159, 27.13%) and high stigma (n = 130, 22.18%), as shown in Fig. 1.

Fig. 1.

Levels of stigma toward mental illness among the participants (N = 586)

Gender differences in the total CAMI score and its subscales are presented in Table 3. The findings revealed a statistically significant difference between the gender mean total CAMI score, male participants exhibiting higher levels of stigma toward mental illness compared to females (M = 108.64, SD = 9.52 vs. M = 105.92, SD = 9.10, respectively) with a small effect size (Cohen’s d = 0.29, p < 0.001). Gender-based comparisons of CAMI scores revealed modest differences. The final sample (n = 586) provided statistical power exceeding 90% (α = 0.05, two-sided) to detect a small effect (d = 0.29).

Table 3.

Gender differences in mean CAMI total and subscales’ scores

CAMI Scale	Male (n = 267)	Female (n = 319)	Mean Difference	95% CI	t (df)	P-value	Hedges g
	Mean (+ SD)	Mean (+ SD)
AU (10 items)	27.26 (4.35)	26.66 (4.18)	0.61	-0.08, 1.30	1.71 (584)	0.088	0.142
BE (10 items)	38.72 (4.96)	39.57 (4.61)	‑0.84	‑1.62, -0.07	-2.13 (584)	0.033*	-0.151
SR (10 items)	25.74 (5.88)	23.88 (5.24)	1.83	0.94, 2.72	4.05 (584)	< 0.001*	0.336
CMHI (10 items)	36.08 (5.21)	36.82 (4.60)	-0.71	-1.49, 0.08	-1.82 (584)	0.079	-0.146
CAMI scale (40 items)	108.64 (9.52)	105.92 (9.10)	2.72	1.20, 4.23	3.52 (584)	< 0.001*	0.290

*Significant (p < 0.05)

Significant gender differences were also observed in the BE and SR subscales. The females demonstrated a more benevolent attitude toward individuals with mental illness compared to males (39.57 ± 4.61 vs. 38.72 ± 4.96, t = -2.13, p = 0.033). On the other hand, males endorsed stronger social restrictiveness attitudes compared to females (25.74 ± 5.88 vs. 23.88 ± 5.24, t = 4.05, p < 0.001). No significant gender differences were found in the AU or CMHI subscales (p > 0.05).

Table 4 presents the multivariate linear regression (MLR) results for the CAMI scale by gender (males, females), examining associations between socio-demographic variables and five dependent variables (total stigma, AU, BE, SR, and CMHI scores). Initially, multicollinearity was examined using variance inflation factors (VIF) and tolerance values. Residual diagnostics (standardized residual plots and normal P–P plots) were also inspected to assess normality, linearity, and homoscedasticity. These inspections did not indicate violations of model assumptions. The regression model examining factors associated with the total stigma score was statistically significant (F (7,311) = 2.75, p = 0.009), explaining 5.8% of the variance (adjusted R² = 0.037). Results from the male model revealed age as a significant factor associated with stigma (β = 0.23, p < 0.01), with older age associated with higher stigma among males. In contrast, higher monthly income was significantly negatively associated with stigma among females (β = −0.14, 95% CI − 5.73 to − 0.14, p = 0.040).

Table 4.

Multivariate linear regression models of socio-demographic factors associated with attitudes towards mental illness

Variables				Male (n = 267)									Female (n = 319)
		Estimate		SE		β	t-value	P-value		95% CI		Estimate		SE	β	t-value	P-value		95% CI
MLR taking the total CAMI scale score as dependent variable
Constant		105.1		2.35			44.69					103.64		2.04		50.80
Age		0.23		0.07			2.76	0.006 *				0.11		0.06	0.11	1.59	0.112		-0.02, 0.21
Marital Status (Ref= Single)
Married		-2.28		1.72		-0.11	-1.32	0.188		-5.68, 1.12		1.91		1.11	0.1	1.72	0.086		-0.27, 4.09
Educational Level (Ref = Secondary School)
University graduate		-1.66		1.21		-0.08	-1.36	0.174		-4.05, 0.73		-1.84		1.08	-0.09	-1.69	0.091		-3.98, 0.29
Employment status (Ref = Unemployed)
Employed		0.76		1.59		0.04	0.48	0.63		-2.37, 3.91		1.53		1.31	0.08	1.16	0.244		-1.05, 4.11
Monthly Income (Ref: Less than 5,000 SR)
5,000 to 10,000 SR		-1.08		1.78		-0.05	-0.61	0.544		-4.61, 2.43		-1.83		1.39	-0.08	-1.31	0.188		-4.58, 0.91
More than 10,000 SR		-2.09		1.88		-0.11	-1.11	0.267		-5.81, 1.62		-2.96		1.43	-0.14	-2.06	0.04 *		-5.78, -0.14
Place of Residence (Ref: Rural area)
Urban area		-0.34		1.19		-0.02	-0.28	0.777		-2.70, 2.02		0.52		1.02	0.02	0.51	0.61		-1.49, 2.55
Know Person with Mental Illness (Ref = No)
Yes		0.29		1.19		0.01	0.24	0.807		-2.05, 2.63		0.15		1.01	0.01	0.15	0.876		-1.83, 2.15
MLR taking the AU subscale score as the dependent variable
Age		0.08		0.03		0.23	2.85	0.005*		0.03, 0.15		0.02		0.02	0.06	1.03	0.302		-0.03, 0.08
Educational Level (Ref = Secondary School)
University graduate		-1.53		0.54		-0.17	-2.80	0.005 *		-2.61, -0.45		-1.02		0.49	-0.11	-2.06	0.039 *		-2.01, -0.05
Monthly Income (Ref: Less than 5,000 SR)
5,000 to 10,000 SR		0.12		0.80		0.01	0.15	0.88		-1.46, 1.71		-0.85		0.63	-0.08	-1.34	0.18		-2.11, 0.39
More than 10,000 SR		-0.99		0.84		-0.11	-1.16	0.244		-2.66, 0.68		-1.94		0.65	-0.2	-2.97	0.003 *		-3.23, − 0.065
MLR taking the BE subscale score as the dependent variable
Educational Level (Ref = Secondary School)
University graduate		0.25		0.58		0.02	0.43	0.666		-0.90, 1.41		1.29		0.55	0.13	2.33	0.02 *		0.20, 2.37
Employment status (Ref = Unemployed)
Employed		0.21		0.77		0.02	0.27	0.785		-1.30, 1.73		-1.4		0.66	-0.14	-2.1	0.036 *		-2.71, -0.08
Monthly Income (Ref: Less than 5,000 SR)
5,000 to 10,000 SR		1.02		0.91		0.1	1.11	0.265		-2.50, 0.90		0.66		0.71	0.05	0.93	0.352		-0.73, 2.05
More than 10,000 SR		-0.8		0.86		-0.07	-0.92	0.356		-0.77, 2.81		1.6		0.72	0.15	2.2	0.028 *		0.17, 3.04
MLR taking the SR subscale score as the dependent variable
Age		0.09		0.04		0.18	2.18	0.03 *		0.01, 0.17		0.03		0.03	0.06	1.01	0.313		-0.03, 0.10
Marital Status (Ref = Single)
Married		-1.17		1.03		-0.1	-1.13	0.256		-3.20, 0.85		1.24		0.63	0.11	1.97	0.049 *		0.01, 2.48
Monthly Income (Ref: Less than 5,000 SR)
5,000 to 10,000 SR		1.51		1.07		0.12	1.41	0.161		-0.60, 3.61		-0.67		0.79	-0.05	-0.85	0.396		-2.22, 0.88
More than 10,000 SR		-0.77		1.12		-0.06	-0.68	0.493		-2.99, 1.44		-2.36		0.81	-0.19	-2.9	0.004 *		-3.96, -0.75
MLR taking the CMHI subscale score as the dependent variable
Monthly Income (Ref: Less than 5,000 SR)
5,000 to 10,000 SR		-1.72		0.94		-0.15	-1.82	0.069		-3.59, 0.13		0.03		0.71	0.003	0.051	0.959		-1.36, 1.43
More than 10,000 SR		0.87		0.99		0.08	0.87	0.382		-1.09, 2.83		1.84		0.73	0.17	2.52	0.012 *		-0.06, 1.97

β = standardized regression coefficient. 95% confidence intervals are for unstandardized coefficients (B). Ordered categorical factors were dummy-coded. All VIF values were < 2, indicating no multicollinearity

*Significant (p < 0.05)

The second linear regression analysis, using the AU score as the dependent variable, indicated that the model for AU subscale was statistically significant (F(8,258) = 2.43, p = 0.015), explaining 7.0% of the variance (adjusted R² = 0.041). Among males, increasing age was associated with higher (AU) scores (β = 0.24, p = 0.005). Also, those with university education were more likely to exhibit authoritarian attitudes toward individuals with mental illness compared with secondary school education (β = −0.17, p = 0.005). Other socio-demographic factors were not significantly associated with Authoritarianism. Regarding females, the MLR model for the AU was also statistically significant (F (8,310) = 2.76, p = 0.006), explaining 6.7% of the variance (adjusted R² = 0.042). Females with university education showed significantly lower (AU) compared with those with lower education levels (β = −0.12, p = 0.039). Higher monthly income (more than 10,000 SR) was also associated with a lower authoritarian attitude (β = −0.20, p = 0.003).

The third MLR analysis, taking the BE score as the dependent variable, indicated that socio-demographic factors showed a significant association with benevolent attitudes only among females. The model for the BE was statistically significant (F(8,310) = 2.07, p = 0.039), explaining 5.1% of the variance (adjusted R² = 0.026). Educational level, employment status, and monthly income were significantly associated with benevolence in female participants. Specifically, those with university education (β = 0.13, p = 0.020), higher monthly income (> 1000 SR) (β = 0.15, p = 0.028), and those who were employed (β = 0.15, p < 0.05) exhibited a more sympathetic view of individuals with mental illness than their counterparts.

The MLR model for the SR was statistically significant (F (8,310) = 3.40, p = 0.001), explaining 8.1% of the variance (adjusted R² = 0.057). Among the sociodemographic factors, only age was significantly associated with SR attitude in the male group (β = 0.18, p = 0.030). Married females reported higher SR scores compared with unmarried females (β = 0.12, p = 0.049). Those with high monthly income exhibited less SR attitudes towards mental illness compared to those with lower incomes (β = −0.20, p = 0.004). Regarding the CMHI subscale, only monthly income was significantly associated with community mental health ideology among female participants (β = 0.18, p < 0.05).

Discussion

Mental illness remains one of the most dialectical social health issues, due to the prejudices and negative attitudes surrounding it, which directly impact those affected [25]. In this study, the mean score of the CAMI scale among participants was 107.16 ± 9.38, ranging from 77 to 141, indicating a moderate level of stigma toward mental illness. This result is in line with findings from other Arab countries, including Lebanon and Iraq, where similar levels of stigma were reported among study populations [3, 14, 26]. Conversely, a study conducted in Al-Ahsa, Saudi Arabia, in 2021 observed lower levels of stigma, using the same scale [24]. Several factors could explain these discrepancies, such as differences in demographic and sociocultural characteristics, as well as variations in timing and exposure to awareness programs [3]. In addition, the CAMI tool assesses attitudes towards general mental disorders, rather than specific conditions. Using tools that measure stigma related to specific mental health conditions may yield more nuanced stigma. For example, one study found more stigmatizing attitudes toward schizophrenia compared to depression, emphasizing a hierarchy of stigma within psychiatric diagnoses [12]. Understanding these nuances contributes to a more comprehensive interpretation of the data.

When analyzing stigma total scores by gender, this study found that males scored significantly higher than females. Numerous global studies have documented notable gender differences in attitudes toward mental illness. Specifically, females tend to demonstrate greater tolerance and more positive attitudes toward individuals with mental disorders [26–29]. However, some studies have reported contrasting findings, indicating that males may exhibit more favorable attitudes toward mental illness, while a considerable proportion of females express fear and hesitance in interacting with individuals experiencing mental health conditions [30, 31].

The current study identified gender differences in community attitudes toward mental illness in both the social restrictiveness and benevolence subscales, consistent with findings from previous research. Specifically, males exhibited more socially restrictive attitudes toward individuals with mental illness compared to females, while females demonstrated greater benevolence than their male counterparts. Social psychology theories may provide an interpretive framework for understanding how biological and sociocultural differences influence gender, mental health conditions, and attitudes toward affected individuals. The Social Role Theory posits that gender roles shape attitudes and behaviors, with men often expected to be emotionally reserved, which may reduce their openness to acknowledging or accepting mental illness in themselves or others. On the other hand, women are socially encouraged to be supportive, fostering empathy and reducing stigma. The Social Identity Theory further supports this perspective by highlighting that cultural values influence attitudes towards mental illness, encouraging compassion in women and emphasizing strength and discipline in men [20]. The responses of participants in this study may be interpreted in light of this perspective. For example, a higher percentage of males (54%) than females (37%) agreed with the statement that “anyone with a history of mental problems should be excluded from taking public office.”

According to the Social Dominance Theory, men tend to be more oriented toward authoritarianism, a finding supported by several global studies [14, 20, 32]. However, female participants in this study exhibited attitudes similar to their male counterparts regarding the perception of individuals with mental illness as inferior and in need of supervision. This finding can be explained by the Social Role Theory, the Social Interaction, and the Social Expectations Theory [20]. In many cultures, including Saudi Arabia, women are socialized to embrace values of nurturing, motherhood, and providing emotional support. Women’s tendency to view individuals with mental illness as requiring close supervision and care—similar to children—may also be reinforced by the stereotype that mental illness signifies weakness. This interpretation is further supported by the higher proportion of female participants, compared to males, who agreed with the statement, “Mental patients need the same kind of control and discipline as a young child.” The interaction between psychological factors and cultural norms underscores that gendered responses to mental illness are not merely differences in opinion but reflect broader, systematic societal standards.

This study revealed no statistically significant gender differences, with both males and females demonstrating positive attitudes toward the acceptance of mental health services and the integration of mentally ill patients in the community (CMHI). Saudi Arabia has made considerable progress in recognizing and addressing mental health disorders over the past two decades, with substantial improvements in service delivery across general health care and specialty hospitals, alongside an increase in mental health specialist staff [33]. The findings may contribute to the body of evidence supporting the agenda of improving mental health services. Acceptance of mental health services and the integration of individuals with mental illness into society may also be influenced by a shared humanity, especially considering that nearly 60% of participants reported prior interaction with someone diagnosed with a mental disorder, and 80% agreed that “Virtually anyone can become mentally ill,” indicating the commonality of these conditions in the population. Given that mental health can be affected by trauma, stress, and genetics, fostering acceptance and integration of individuals with mental illness is crucial for the well-being of society as a whole.

While the gender difference in attitudes was statistically significant in this study, the small effect size (d = 0.29) implies that gender differences represent a modest component of stigma and suggests that other factors likely contribute substantially to attitudes toward mental illness. The association between sociodemographic factors and the dependent variables revealed that AU and SR attitudes toward individuals with mental illness were significantly higher among older males. Notably, AU attitudes were significantly lower among participants with secondary education compared to those with higher education, and were more pronounced among females with higher monthly incomes. A national study from Lebanon reported contrasting findings, indicating that younger and more highly educated participants exhibited lower levels of stigma toward individuals with mental illness [14]. However, our results align with research from Singapore and Saudi Arabia, which found that older individuals tend to exhibit more unfavorable attitudes and reduced tolerance toward those with mental health problems [32, 34]. This discrepancy may be attributed to younger participants being more open-minded and exposed to fewer negative stereotypes, as well as greater mental health awareness through the media.

Unlike a study conducted in an Asian community, which found that being unmarried and unemployed was linked to greater tolerance and less social restrictiveness toward individuals with mental illness [32], our findings reveal a different pattern. Specifically, single and unemployed female participants expressed more avoidant attitudes and less compassion toward mental illness, which may be due to limited life experience and social support needed to enhance empathy and reduce their stress, fears, and concerns about social image. These comparisons underscore the variability and the complexity of stigma-associated factors across different contexts and point out the necessity of future gender-sensitive anti-stigma strategies.

Limitations of the study

This study has several limitations that should be considered when generalizing findings to broader populations. First, participants were recruited using non-probability convenience sampling at specific times from public venues, which may introduce selection bias in the demographic composition and attitudes. The overrepresentation of some groups, such as females, young people, high literacy participants, may bias estimates toward less stigmatizing attitudes compared to the general population. Second, the reliance on self-reported data introduces the possibility of response bias, as attitudes are subjective and may be influenced by social desirability. Third, the cross-sectional design inherently limits the ability to infer causal relationships between variables. Despite the regression models accounting for measured covariates, the possibility of residual confounding from unmeasured or insufficiently captured factors, such as dichotomously measured contact with individuals with mental illness, mental health literacy, cognitive factors, and other psychosocial factors, remains a potential threat to the validity of the study’s findings. Fourth, though the CAMI scale showed satisfactory internal consistency, potential measurement errors may exist due to response biases and the absence of systematic testing for measurement invariance across gender and other demographic cohorts, which may affect the comparison of subgroup results. Fifth, multiple linear regression analyses were performed on the CAMI total score and subscales without formal correction for multiple testing, which may increase the risk of type I error, although findings were consistent across domains. To enhance the robustness of findings, future research should use probability-based sampling and perform measurement invariance testing to further examine the comparability of the CAMI scale across diverse demographic groups. A mixed-methods design with a representative sample would also improve understanding of gender-based variations in attitudes toward mental illness and their predictors.

Conclusions

This study is among the first to systematically compare attitudes toward mental illness between Saudi males and females using the CAMI scale. Despite the methodological limitations, including the convenience sampling, this study found important baseline indications of gender-differentiated attitudes toward mental illness (benevolence and social restriction) in the Saudi context. Accordingly, confirmatory studies with probability sampling and validated measurement invariance testing are highly needed before these findings can inform the design and implementation of mental health anti-stigma programs. These results highlight the critical importance of incorporating gender considerations into future mental health research and policy development. Furthermore, future anti-stigma interventions should be designed to be adaptable and tailored to the specific demographic profiles, with findings from adequately powered, probability-based studies to maximize effectiveness in raising awareness and delivering psychological support.

Author’s contributions

The author was responsible for conceptualization, methodology, project administration, data curation, formal analysis, resources, software, validation, visualization, writing – original draft preparation, and writing – reviewing and editing.

Funding

This research received no external funding.

Data availability

The data used during the current study are available from the author.

Declarations

Ethics approval and consent to participate

The study was conducted according to the guidelines of the Declaration of Helsinki, and the ethical approval was obtained from the Standing Committee for Scientific Research at Jazan University (approval number REC-45/05/842) on November 16, 2023. Written informed consent was obtained from all subjects involved in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.