Mental health literacy among primary healthcare workers in South Africa and Zambia

Abstract Background In developing countries, mental health literacy (MHL) still needs to be improved due to the high prevalence of mental disorders. It is widely recognized that MHL can improve health outcomes for both individuals and populations. Healthcare professionals’ development in MHL is crucial to the prevention of mental disorders. The aim of this study was to assess MHL of primary healthcare (PHC) workers in South Africa (SA) and Zambia and determinants thereof. Limited evidence is available on the levels of MHL among PHC workers in the sub‐Saharan Africa region, which faces a large burden of mental disorders. Methods The study population for this cross‐sectional survey comprised PHC workers (n = 250) in five provinces of SA and Zambia. MHL was measured with the Mental Health Literacy Scale (MHLS). We conducted a multivariate analysis to explore determinants of MHL. Results Results showed moderate MHL among PHC professionals, but with a wide range from low to high MHL. Knowledge‐related items had a greater dispersion than other attributes of MHL. PHC workers with more education showed a greater ability to recognize mental health‐related disorders. Those who had experience in the use of mental health‐related assessment scales or screening tools reported a higher total MHL. The results confirmed strong internal consistency for the MHLS. Conclusion The results highlighted varying mental health perceptions and knowledge in PHC. Implementation of specifically developed formal training programs and interventions to improve MHL in PHC workers to strengthen their competence may help bridge the treatment gap.

Although a great deal of work on mental health-related knowledge has been completed regarding stigma, attitudes, and skills of PHC professionals delivering mental health services (Ayano et al., 2017;Collins et al., 2011;Davies & Lund, 2017;Kakuma et al., 2010;Petersen et al., 2019;Sweetland et al., 2014), some knowledge gaps remain to be filled (Egbe et al., 2014;Elyamani et al., 2021;Linden & Kavanagh, 2012;Saxena & Belkin, 2017;Öztaş & Aydogan, 2021). Although PHC workers could be the key players in the promotion of public mental health literacy (MHL), as they work closely with communities (Atilola, 2016;WHO, 2018), relatively little research has been conducted in the South African and Zambian health sectors on the MHL of PHC workers. PHC workers require specific professional training on the identification and management of mental health problems (Hanlon et al., 2014;Kapungwe et al., 2011;Marais & Peterson, 2015;Mwape et al., 2010). For the promotion and effective implementation of mental health services in SSA, understanding the multiple and intersecting cultural factors that converge on mental health problems is pivotal, as these factors are linked to MHL among PHC workers and the communities in which they work and live (Atilola, 2016;Kutcher et al., 2016;.
The concept of MHL originally consisted of beliefs about mental disorders that aid in their recognition, management, or prevention (Jorm et al., 1997). Early development of the concept focused on the adult population, specifically healthcare providers for raising awareness of mental health concerns (Jorm, 2012). Over time, the need to respond more broadly to the mental health problems of communities has been recognized, and the concept can also be seen as a broad set of activities that support and benefits positive mental health (Jorm 2019;Kutcher et al., 2016). MHL is closely related to health literacy, which has been recognized as a clear determinant of socioeconomic status and prosperity in life WHO, 2013). Therefore, lack of health literacy is one factor that underlies the growing burden of (mental) health services (Sørensen et al., 2015). Although proper health literacy is a marker of economic prosperity (WHO, 2013), poor MHL among PHC professionals may contribute to the disease burden and be a barrier to adequate treatment of those in need (Ganasen et al., 2008;Loo et al., 2012;Mwape et al., 2012, Vistorte et al., 2018. Promoting MHL among PHC professionals is a core component of successful mental healthcare integration to reach the Sustainable Development Goals for global mental health in SSA (Target 3.4; United Nations, 2020). However, MHL is a complex concept and more than just a synonym for knowledge of disorders, and it needs to be considered within specific cultural contexts (Atilola, 2016;Ganasen et al., 2008). Multicultural attitudes and beliefs affect help-seeking behaviors in communities (Kutcher et al., 2019;Rathod et al., 2017). As SSA is a geographically and multiculturally diverse region, individuals' beliefs vary, and these beliefs may be unique to specific cultures. These strong cultural beliefs and traditional perspectives also influence individuals' views regarding mental health disorders (Al-Yateem et al., 2018;Atilola, 2016;Munakampe, 2020;Mwape et al., 2012) subsequently impacting their mental health literacy.
Although the common expectation is that PHC workers should be more health literate than laypeople, previous findings showed that African PHC workers have negative attitudes toward mentally ill patients (Mosaku & Wallymahmed, 2017 Maconick et al., 2018). In addition, stigma and strong negative attitudes among PHC professionals have been found to challenge mental health support in the region (Kapungwe et al., 2011;Mosaku & Wallymahdmed, 2017). In Zambia, for instance, large numbers of PHC providers show negative stereotypes toward mentally sick people (Kapungwe et al., 2011). For that reason, an understanding of the factors that reflect perceptions of mental health issues is pivotal (Atilola, 2016).
In this study, we sought to assess the MHL of PHC workers in South Africa (SA) and Zambia including background factors, namely, demographics and educational and professional training needs. Research questions posed in the study were as follows: (1)   . We conducted this study using a revised MHLS (Korhonen et al., 2019O'Connor & Casey, 2015) that was validated and psychometrically tested, especially for low-and middle-income countries (LMICs).
The revised MHLS is psychometrically sound with good construct validity and internal consistency O'Connor & Casey, 2015;Wei et al., 2016). The revised MHLS elaborates on new information about the factors that may affect PHC workers' provision of quality mental health care in the sub-Saharan region.

The study setting and population
The study population comprised PHC workers (n = 306), recruited to participate in the study from five MEGA project regions and 45 clinics in SA (specifically in the Free State, Gauteng, and Western Cape Provinces) and Zambia (Lusaka and Central Provinces). We included healthcare workers in Zambia from rural, peri-urban, and urban health- All data collected from participants were anonymous.
We conducted the study during clinics' operating hours, and partic- D12, Work experience. Finland, as the project coordinator, managed the data collection process, data analysis, and storage of data (paper and electronic).

Data analysis
We analyzed data using SPSS version 26 software. We analyzed the sample's characteristics using descriptive statistics (

Demographic characteristics of participants included in the final study
The MEGA project researchers proposed a convenience sample of PHC workers (N = 505), of which 306 agreed to participate. We explored Cronbach's alphas for internal consistency of the revised MHLS with the total number of recruited participants (n = 306). Of the scale's six attributes, three showed preferred alpha values (≥.70). The results also showed strong internal consistency (α = .80) for the scale's 35 items.
Three knowledge-related attributes fell below the cutoff criteria of internal consistency.
We were able to calculate the total MHLS score with complete data for 250 participants. The majority of these PHC workers were South

Results for mental health literacy of PHC workers
We explored the distribution of individual items, sum of scores, and six attributes of the revised MHLS for MHL among PHC workers (n = 250).
The highest total score of the 35 items in the revised MHLS was  We examined the distribution of results in more detail in individual responses in two groups, which employed 4-and 5-point scales. In the group of 1-to 4-point items measuring recognition and knowledge, the mean scores ranged from 2.13 to 3.57. A total of 12 of 15 items, including all questions regarding "Ability to recognize disorders" (Q1-Q8), had mean scores above 3.0 (minimum = 1, maximum = 4). We found the lowest mean scores, receiving less than 3 of 4, in three knowledgerelated questions (Q10, Q12, and Q15) with reversed scoring. We also found the greatest SD and CV of a single 4-point question in the question related to the theme of knowledge as "a condition that would allow a mental health professional to break confidentiality" (Q15; SD = 1.09, CV = 44%). Table 3 shows the mean scores, SD, and CV of items of the 4-point scale in the revised MHLS.
The mean scores of 5-point items measuring knowledge and attitude ranged from 2.58 to 4.5. Most of the 20 items consisted of questions related to "Attitudes that promote recognition or appropriate help-seeking behavior (stigma)" (Q20-Q35), with eight of them receiving scores ≥4.0. We found that two items that measured attitudes (Q33, Q34) had scores below 3.0. These questions were about willingness to have someone with a mental illness marry into the respondent's family (Q33; SD = 1.18, CV = 40%) and willingness to vote for a politician if they had suffered a mental illness (Q34; SD = 1.3, CV = 51%). We found that sixteen 5-point items had an SD ≥ 1.000 and eight 5-point items had a CV ≥ 30%. All items had a generally large level of dispersion around the means. Table 4 shows the mean scores, SD, and CV of items of the 5-point scale.

Results for the determinants of mental health literacy
We conducted variance of analysis on the scored responses (n = 200).
We explored the analysis in comparison to background questions as determinants by calculating the total sum of the 35 items of the MHLS and three main groups of items based on attributes of the MHL: the ability to recognize disorders (Q1-Q8), knowledge of how to seek information (Q16-Q19), and attitudes that promote recognition or appropriate help-seeking behavior (stigma; Q20-Q35). We omitted three attributes in relation to knowledge-related questions from the analysis due to their low alpha levels ( Table 2). This finding is consistent with our previous research findings on psychometric testing .
We found a statistically significant dependence in determinants relating to the level of education and the use of mental health-related assessment scales or screening tools in work. Those who had more education were better able to recognize mental health-related disorders (F = 2.869; σ = .038; partial η 2 = .046). In pair-wise comparison, PHC workers with a degree-level education had a higher mean than workers with certificates (p = .05).
The use of mental health-related assessment scales or screening tools in work had a statistically significant effect on expressed attitudes and the whole MHLS instrument. In terms of mean scores, those who reported using scales or screening tools in their work showed more attitudes that promote recognition or appropriate help-seeking behavior (stigma; F = 4.523; σ = .035; partial η 2 = .025). Moreover, they performed better in the MHL survey overall (F = 5.285; σ = .023; partial η 2 = .029). We found no statistical significance between determinants and a single attribute for knowledge of how to seek information.
Neither continuous professional training/activity nor working area, as country or province, was found to be statistically significant. Table 5 shows the main results of the multifactor analysis of variance.

DISCUSSION
In this study, we explored the levels and determinants of MHL among PHC workers in SA and Zambia. Based on the authors' knowledge, this is the first study that explored the level of MHL among PHC workers in SA and Zambia using the MHLS that was revised and validated (Korhonen et al., 2019.
In our findings, the scores among PHC workers ranged from 76 to 150, with the mean scores being less than a group of professionals Furthermore, the results indicated strong internal consistency for the entire revised MHLS, which is also in line with our previous findings .
Mental health-related disorders were more readily recognized by PHC workers with higher education. This is in line with Jorm (2000), who asserted that sufficient education and high MHL levels among healthcare workers make it possible for them to recognize mental disorders and promote help-seeking behaviors. Moreover, Adu et al. (2021) found the level of education positively associated with higher MHL among Ghanaians, while it was agreed by Öztas and Aydogan (2021) that healthcare professionals had higher MHL as a result of their education level. Linden and Kavanagh (2012) found that PHC nurses with less training as well as little exposure and experience in mental health had negative, prejudiced, and fearful attitudes toward and perceptions of people with mental health problems.
The results of this study showed that 59% of participants considered a diploma to be a sufficient indicator of professional education. The WHO (2007) and the International Council of Nurses already reported a scarcity of mental health education and training opportunities for nurses worldwide. Taking into account the quality and standardization of education, it is also key to improving the mental health gap, but also labor mobility. According to Bitta et al. (2017), in resource-constrained economies, younger professionals are advantageous because training can be increased and sustained for a longer period of time.

TA B L E 5
We did not find work experience as a statistically significant determinant of MHL, but education played a bigger role in improved MHL.
In line with Marangu et al. (2021), most of the study participants were young adults and women. Almost half of the participants (49%) in the present study reported working experience of less than 10 years. In a study by Al-Yateem et al. (2018), younger nurses with less clinical experience were more knowledgeable about mental health treatment.
In contrast, a systematic review found that factors associated with a lower recognition of mental illness across all demographics included young adults, illiteracy, and females (Elyamani et al., 2021).
Having an inadequate understanding of mental illness can influence mental health utilization, treatment compliance, and help-seeking behaviors. This review by Tonsing (2018) also revealed an association between attitudes toward mental illness and help-seeking behavior.
Early detection and intervention of mental disorders are dependent on the individual's mental health literacy (Loo et al., 2012). Moreover, healthcare professionals working in primary health providing care to people with mental disorders and who are stigmatizing toward them are more likely to believe that people with mental illness will not comply with treatment (Vistorte et al., 2018).
We also found that those who had experience in the use of mental health-related assessment scales or screening tools reported fewer stigmatizing attitudes and a higher overall MHL. However, we did not find statistical significance between the background and knowledge of how to seek information, and the rest of the three knowledgerelated questions that showed great relative dispersion fell below the cutoff criteria (α ≥.70), which may be caused by reversed scoring. In our study, more than half of the participants reported using a mental health-related screening tool in their work. Marangu et al. (2021) revealed significantly low MHL results among PHC workers, evidenced by low diagnostic accuracy for serious mental health disorders. Thus, according to studies, it seems that the use of health-related professional assessment scales and screening disorders promotes healthcare professionals' MHL. However, the development of tools for mental health assessment alone is not useful, but training in their use would be important to better address the health challenges in the region.
Standardizing PHC workers' mental health-related competence proved to be a challenge in the future, as the results show a large variance in MHL levels for the entire MHLS and its attributes. All items had generally a great level of relative dispersion around the means in terms of CV. In comparison to the distribution of MHLS results among PHCs by attribute, knowledge-related questions (of how to seek information) showed great variation (CV = 23%), with the total score having the smallest CV (10%). Generally, items of "Ability to recognize disorders" scored well on the 4-point scale, but also had relatively large variance.
Knowledge-related questions were found to be more challenging for the respondents when examining individual items. Considering these findings, our results show heterogeneous MHLs among PHC workers, and again, raise a need for standardized mental health education.
Although significant progress has been made in improving mental health-related knowledge, stigma, attitudes, and skills of PHC workers, knowledge gaps are still evident. Nurses are expected to be cognizant of the needs of people with mental illnesses without prejudice or discrimination (Shahif et al., 2019). Therefore, promoting MHL among PHC professionals is critical for successful integration of mental health care into mainstream healthcare services. This study's findings present factors related to MHL among PHC workers in SA and Zambia. Despite inadequate knowledge and competence of healthcare professionals to screen and care for people with mental health problems, as well as poor attitudes toward mental health problems, we found gaps in their MHL.
By identifying these gaps, we provide an opportunity for and guide potential future research in mental health services in PHC.

CONCLUSION
Our findings reflected MHL among PHC in SA and Zambia as moderate, but it varies according to educational background. One of the prominent findings of this study was the significance of education and guidelines for screening mental health disorders that promotes MHL.
Education may foster PHC workers' ability to recognize mental health disorders, while professionals' routine assessment of mental health can reduce stigma and strengthen MHL overall. Therefore, it is important to increase PHC workers' mental health knowledge and strengthen their skills through training. Standardizing and enhancing their mental health competence is critical for successful integration of mental health care into PHC services and will improve the response to mental health problems in SA and Zambia.

LIMITATIONS
This study has limitations that should be considered when interpreting the results. First, although the study included a small (n = 250) but statistically relevant sample, it consisted of only nurses and clinical officers. Extending the participation to other healthcare disciplines may provide more insights into the levels and determinants of MHL among PHC professionals in SSA. Second, based on our findings, we suggest using the revised MHLS to measure MHL using the scale's total score, as three of the knowledge-related attributes that mainly had reversed scoring system, and as an obvious reason, fell below the appropri-