Is workplace violence against health care workers in Mozambique gender related?

Abstract

This report revisits data used to describe the typology and the perceived impact of violence against health care workers (VHCW) at the health services of the City of Lichinga in Mozambique, based on an observational, descriptive, cross‐sectional study, carried out from March to May 2019. In this report we attempt to understand if our reanalysis of VHCW in Niassa can explain it as an example of gender‐based violence. Our findings—particularly that women more than men reported not knowing if the health services had any policies or procedures to deal with VHCW, felt that they were not encouraged to report acts of VHCW and were more frequently threatened/violented by different sex aggressors—although not conclusive, support the need to consider gender as a dimension when conducting research on VHCW. If we do not do so, gender will continue to be an invisible and ignored dimension of intervention strategies to prevent and address VHCW.

Highlights

• Our findings support the need to consider gender as a dimension when conducting research on violence against health care workers (VHCW).

• If we do not do so, gender will continue to be an invisible and ignored dimension of intervention strategies to prevent and address VHCW.

• This is particularly important in health and care contexts where there is a growing feminisation of the workforce, as well as gender differences in engagement with the profession.

1. BACKGROUND

Violence in specific contexts, such as domestic environments ¹ and the workplace, ² , ³ is a major problem, reflecting ‘gender stereotypes, multiple and intersecting forms of discrimination, and unequal gender‐based power relations’ (https://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:C190) with significant impact on health and well‐being. It is acknowledged that violence against health care workers (VHCW) tends to remain invisible in comparison to other forms of violence. ⁴

In sub‐Saharan Africa, studies on VHCW have been reported from several countries, acknowledging the endemic dimension of the phenomenon, the negative impact on the well‐being of health care workers (HCW) and their in‐service performance, a high level of tolerance to non‐physical violence and the absence of policies to deal with workplace violence, contributing to the underreporting of the problem and its neglect in health workforce planning. ⁵ , ⁶ , ⁷ Several studies have shown that all forms of VHCW are recognised as a significant issue both by healthcare leaders and workers in Mozambique. ⁶ , ⁷

However, gender aspects of VHCW have not been properly addressed, remaining the question as to what extent it should be interpreted as gender‐based violence (GBV), a global public health problem with a significant prevalence in developing countries, ⁸ , ⁹ , ¹⁰ including SSA ¹¹ and Mozambique. ¹²

This report revisits data used for two articles that describe the typology and the perceived impact of VHCW at the Provincial Hospital and at the Health Centre of the City of Lichinga in Mozambique during 2019. ⁶ , ⁷ In this report we attempt to understand if our results on VHCW in Niassa can be considered as an example of GBV.

This was driven by the 2019 Centenary Conference of the International Labour Organization which adopted a Convention, accompanying Recommendations and a Declaration to address, among other issues, violence and harassment in the workplace. These documents acknowledged ‘that GBV and harassment disproportionately affects women and girls’, and recognise ‘that an inclusive, integrated and gender‐responsive approach, which tackles underlying causes and risk factors … is essential to ending violence and harassment in the world of work’ (https://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:C190).

2. POPULATION AND METHODS

This was an observational, descriptive, cross‐sectional study, carried out from March to May 2019 in all departments of the Lichinga Provincial Hospital and at the Lichinga City Health Centre in the Niassa Province of Mozambique.

The methodological details are described in other articles. ⁵ , ⁶ This questionnaire survey collected self‐reported information on VHCW in the selected health care units in the 12 months preceding the survey date.

The study participants were recruited following a simple random sampling strategy. The study excluded HCW: who were on vacation or taking another type of leave; working at the study health unit for <12 months; who did not report any history of having suffered violence in the workplace in the past 12 months. Two hundred and sixty HCW were eligible to participate based on inclusion criteria.

The study was approved by the Institutional Committee on Bioethics in Health, Faculty of Medicine, Eduardo Mondlane University and Central Hospital of Maputo (registration number CIBS FM & HCM 097/2018).

Data were entered into SPSS 20.0. The analysis cross tabulated sex of the HCW with other categorical variables testing statistical significance with either Fisher Exact test, Pearson chi‐square or likelihood ratio chi‐square, as appropriate. Means were compared using ANOVA.

3. RESULTS

Female victims of VHCW in Niassa are usually younger (63% of females were <35 years of age, compared to 45% of males, p = 0.020) and more inexperienced than their male colleagues (men [11.3 + 7.5 years] had a longer working presence in the health sector than women [8.0 + 4.9] [Anova p = 0.003]).

Our results indicate that the health centre is a higher risk environment than the hospital and the percentage of female victims of VHCW working at the health centre (n = 14/65, 22%) is significantly higher than the percentage of men (n = 6/75, 8%) (Fisher exact test p = 0.020). The relative percentage of women (n = 39/65, 60%) doing night shifts (working from 18:00 h to 07:00 h) was higher than men (n = 33/75, 44%) (Fisher exact test p = 0.043).

Women (n = 58/65, 89%) more than men (n = 53/75, 71%) reported not knowing if the health services had any policies or procedures to deal with VHCW (likelihood ratio p = 0.040). Females (n = 42/64, 66%) more frequently than males (n = 34/74, 46%) felt that they were not encouraged to report acts of VHCW (Fisher exact test p = 0.016).

Women were more frequently threatened/violented by different sex aggressors than men (Table 1).

TABLE 1.

Sex of the aggressors cross‐tabulated with sex of victims of workplace violence

			Sex of the victim of violence
			Male	Female
Sex of the verbal aggressor a	Female	n	6/34	11/28
		%	18%	39%
	Male	n	28/34	17/28
		%	82%	61%
Sex of the physical aggressor b	Female	n	0/4	2/4
		%	0%	50%
	Male	n	4/4	2/4
		%	100%	50%
Sex of the sexual harasser b	Female	n	3/5	1/3
		%	60%	33%
	Male	n	3/5	2/3
		%	60%	67%
Sex of the person discriminating b	Female	n	3/12	2/4
		%	25%	50%
	Male	n	10/12	2/4
		%	83%	50%
Sex of the person bullying b	Female	n	5/28	6/26
		%	18%	23%
	Male	n	24/28	2026
		%	86%	77%

^a Fisher exact test p = 0.05.

^b Fisher exact test not significant.

4. DISCUSSION AND CONCLUSIONS

If we seek to eliminate VHCW then quality research and theoretical work must underpin such developments, and our findings, although not conclusive, support the need to consider gender as a dimension when conducting research on VHCW. If we do not do so, gender will continue to be an invisible and ignored dimension of intervention strategies to prevent and address VHCW. ¹³

This is particularly important in a context where there is a growing feminisation of the health workforce in lower‐middle‐ and upper‐middle‐income countries, ¹⁴ as well as differences between male and female physicians' engagement with the profession, ¹⁵ potentiating exposure to violence in the workplace.

If future research supports the hypothesis that VHCW can be construed as GBV, then the observed prevalence of VHCW might be due to the prevalence of gender inequality in society that underlies such violence. This will indicate the need to frame policies and strategies against VHCW within a broader framework to tackle the social, cultural and political contexts that sanction this type of violence. ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵

CONFLICT OF INTEREST

The authors declare that they have no competing interests. The study did not involve any type of financial gains or interests.

ETHICS STATEMENT

The study was approved by the Institutional Committee on Bioethics in Health, Faculty of Medicine, Eduardo Mondlane University and Central Hospital of Maputo (registration number CIBS FM & HCM 097/2018).

Biographies

Paulo Ferrinho: Portuguese and South African medical doctor, PhD in Community Health. Professor and Director of Department of Global Public Health at Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa (Nova‐IHMT), Portugal. Guest Professor at Universidade de Cabo Verde. His interests include health workforce planning and development.

Sérgio Roques Patrício, anthropologist; MPH; researcher at the Niassa Operational Research Unit, Provincial Health Department, Lichinga, Niassa Province, Mozambique. Areas of interest: Health systems; violence; Gender‐based violence; Rituals; Traditional medicine, occupational health and sexual and reproductive health.

Isabel Craveiro, sociologist, PhD in International Health. Assistant Professor at Department of Global Public Health at Nova‐IHMT where she coordinates the Master Program in Public Health and Development. Her interests include health systems' evaluation, health workforce development, fair research partnerships, gender equity and mixed research methods.

Mohsin Sidat, Mozambican, MD, MSc (infectious diseases), PhD in medicine (public health). Associate Professor at Faculty of Medicine of the University Eduardo Mondlane with interests in public health, health systems research, epidemiology and prevention of infectious diseases and chronic non‐communicable diseases. Guest Associate professor in global public health a Nova‐IHMT.

Ferrinho P, Patrício SR, Craveiro I, Sidat M. Is workplace violence against health care workers in Mozambique gender related? Int J Health Plann Mgmt. 2023;38(1):265‐269. 10.1002/hpm.3603

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.