Prevalence and determinants of chewing khat among women in Ethiopia: data from Ethiopian demographic and health survey 2016

Background In Ethiopia and other countries in eastern Africa, khat abuse is an increasing public health problem. Levels of use appear to be increasing in women, who are more vulnerable to khat-related problems. However, population-based data are lacking as studies have been small and related to specific settings. This study aimed to contribute to current knowledge on the prevalence of chewing khat and associated factors among women in Ethiopia, using data from the 2016 Ethiopian national demographic and health survey. Methods The 2016 EDHS used a two-stage stratified sampling design to select households. A total of 645 enumeration areas (202 urban and 443 rural) were selected, based on the 2007 Ethiopia Population and Housing Census. In these, 18,008 households were considered, from which 15,683 women were included from individual households. The women were interviewed by trained lay interviewers. Data were tabulated and logistic regression was used to examine mutually adjusted associations, expressed as adjusted odds ratios. Results The lifetime prevalence of chewing khat among women was 9.9%. Current khat use was 8.4%, with a mean of 14.2 days of use in the last month. Khat use increased with increasing age, remaining constant after age 35 years, having one child, lower educational level, being Muslim by religion and not pertaining to the lowest wealth index category. Not being in a marital relationship with the most recent sex partner and Protestant religion were protective factors. Conclusion Lifetime prevalence of chewing khat among women in Ethiopia is substantial and associated with specific sociodemographic risks. These can be used in targeted public health efforts to control the use of khat and reduce the associated health and economic burden.


Background
Khat (Catha edulis) is a natural stimulant widely cultivated and available in East Africa and the Middle East [1]. The biochemically active ingredients of khat are cathinone and cathine, which are alkaloid chemicals bearing resemblance to the psychoactive substance of amphetamine, both structurally and functionally [2]. Cathinone is the main psychoactive component of khat leaves and a potent stimulator of the sympathetic nervous system as well as the central nervous system, similar to the pharmacological effects of amphetamine [2]. Chewing fresh leaves of the khat tree is the most common mode of intake [3].
Khat abuse is an increasing public health problem and strongly associated with adverse mental and other outcomes and such as psychological distress, poorer quality of life and increased road traffic accident in low-income countries [4]. Across the world, an estimated 20 million people chew khat leaves daily [4]. Khat is generally used for its perceived ability to facilitate interpersonal communication in social settings, to induce euphoric and performance enhancing effects and to fend off fatigue, as well as for its perceived medicinal value for the treatment of headaches and common cold [5].
Khat use has been associated with numerous health problems. Basic research studies have found evidence of altered stress response [6], cognitive deficits [7,8], increased levels of depressive symptoms and distress [9] and insomnia [10] in habitual khat users. Research has also shown that prolonged exposure to khat may lead to a long-lasting sensitization to the effects of other drugs [11]. In addition to physical and mental harm, much time and household income is spent on obtaining and chewing khat [12], which severely affects users' social life and family [13]. The habit of chewing khat may be postulated as one of the risky behaviours that could fuel the spread of sexually transmitted infections (STIs) due to associated risky sexual behaviours like having casual sex, unprotected sex and early initiation of sexual activity reported among chewers [14]. Studies reveal that khat consumption may impact fetal health, for example affecting fetal growth by inhibiting utero-placental blood flow [15,16].
Research shows that Khat use is a common practice in Ethiopia [17]. Before 2000, khat chewing by women was generally not considered acceptable [18]. Currently, however, use of khat among women is increasing very rapidly [19]. Research suggests that husbands may be in part responsible for these changes as they encourage their wives to share in the chewing of khat [18].
People who use khat are more vulnerable to khatrelated problems like depressive symptoms, posttraumatic stress disorder and common mental disorder [20,21], the issue to date has not received much attention from local authorities.
Few studies have been conducted to establish prevalence of use and associated factors, and these tended to be small and focusing on specific settings [22].
The current investigation, therefore, aimed to contribute to current knowledge on the use of khat and associated factors among women in Ethiopia, using data from the uniquely representative and population-based 2016 Ethiopian national demo-graphic and health survey (EDHS).

Study design and sampling
The 2016 EDHS used a two-stage stratified sampling design to select households. In the first stage, there were 645 enumeration areas (202 urban and 443 rural) based on the 2007 Ethiopia Population and Housing Census (PHC). A total of 18,008 households were considered, of which 16,650 households and 15,683 women were eligible. The women were interviewed by trained lay interviewers. All women at reproductive age (15 to 49 years) who were either permanent residents of the selected households or visitors who stayed in the selected household the night before the survey, were eligible for the study. A total of 15,683 women aged 15-49 years were interviewed with a response rate of 95%. The EDHS was designed to provide sociodemographic and health indicators at national (urban and rural) and regional levels. The survey follows an international methodological approach and is conducted at five-year intervals. Detailed methodology is found elsewhere [23].

Measurements
The dependent variable was chewing khat in the last 30 days. Women who chewed khat in the last month were considered as current khat user. Sociodemographic, socioeconomic and sex-related variables associated with risk were used as independent variables. The independent variables included in this analysis were age, educational status, and marital status.
Household wealth index is based on household ownership of consumer items such as a fan, television, and car. In addition, dwelling characteristics, such as the flooring material; type of drinking water source; and toilet facilities, are among other characteristics related to wealth.
It has five categories: poorest, poorer, middle, richer, and richest), residence, religion, residence, sex of household head (person in the house with main responsibility for making decisions and providing income; can be female regardless of presence of males in the household), number of children, number of sexual partners (if participants were not willing to disclose the information, the variable was coded '9' as a category for 'not known'), sexual transmitted infection (STI) in the last 12 months, relationship to household head, ever reads newspaper or magazine (yes/no), ever listens to radio (yes/no), ever watches TV (yes/no), ever use of internet (yes/no), most recent sex partner (if participants indicated they had no sexual partner, this was coded as a separate category; if the participant was not willing to disclose the information, the variable was coded '9' as a separate category for 'not known').

Statistical data analysis
The data were extracted, edited, and analysed using SPSS version 23 for Windows. Frequency tables were used to summarize sociodemographic characteristics and prevalence of khat use. Bivariate logistic regression was performed separately for each independent variable. Independent variables with a p-value< 0.05 were entered into the final model for multivariable analysis. Variables in the mutually adjusted multivariable model with a two-sided p-value< 0.05 were considered statistically significant. In order to check for multicollinearity among variables in the multivariable model, variance inflation factors were calculated for all independent variables.

Results
A total of 15,683 participants were included. The mean (SD) age of women was 27.9+ (9.2) years. The majority were Orthodox Christian by religion (40.9%), had no education (44.8%) and were from rural settings (65.9%) ( Table 1). The highest proportion were from the age group 15-19 years (22.3%) followed by the 20-24 years group (18.5%) and the 25-29 years group (18.1%). A total of 24.8% of participants were identified as 'poorest' and 13% as 'poorer' according to the wealth index. The mean number of children ever born was 2.6.
The lifetime prevalence of khat chewing among women was 9.9%. The prevalence of current khat chewing was 8.4%, with a mean number of 14.2 days of use in the last month. Highest prevalence of current khat chewing was seen in the age group 40-44 years old (12.9%), 30-34 years old (12.4%), 35-39 years old (10.9%) and 45-49 years old (10.1%).
Prevalence of current khat chewing was lowest among participants from the lowest wealth index (5.9%), compared to participants from other wealth index categories. Higher prevalence of current khat chewing was also seen among individuals without education (10.3%) or only primary education (8.3%), compared to those with secondary education (4.9%) and higher education (4.6%). Among current khat chewers, the prevalence was highest among those of Muslim religion (18.2%) as compared to other religious groups (0.0-2.8%). The highest prevalence of khat chewing was seen when the relationship to the household head was wife (10.7%), and mother (10.7%), followed by head (9.5%). Prevalence of current khat use was also higher among respondents who had a single sexual partner (8.6%), two partners (8.4%) and three or more partners (8.2%), compared to those with no sexual partner (5.4%). Prevalence of khat use was similar regardless of STI in the last 12 months. The prevalence of current khat use was 8.8% in urban areas versus 8.2% in rural areas, and rates were slightly higher among those not reading newspaper/magazines (8.9% versus 5.9% of those reading these), those not listening to radio (8.7% versus 7.8% of those listening to radio), those not watching TV (8.5% versus 8.3% of those watching), and those without access to internet (8.6% versus 6.0% of those with access).

Discussion
To the best of our knowledge, this was the first reported study that investigated the prevalence of khat chewing and associated factors among women in the Ethiopian population using the latest nationally representative data obtained from EDHS 2016. The findings reveal that in women, the life time prevalence of khat chewing is 9.9% versus 8.4% of current use. Multivariable analysis revealed that current use was more prevalent with higher age, having one child, lower educational level, being Muslim by religion and not pertaining to the lowest wealth index category. Protective factors were not being in marital relationship with sex partner and Protestant religion.
The prevalence of current khat use among women in this study, at 8.4%, was lower as the rate reported in Yemen (29.6%) [18]. Cultural factors underlying use may differ between the two countries, but differences may also be due to variation in survey design, and other methodological issues.
The multivariable model revealed that khat use increased with age until age 35 years, after which it remained approximately constant. This is consistent with a study conducted in eastern Ethiopia [24]. The possible reason may be that as age increases, women are more likely to have alterations in life circumstances such as bereavement, social isolation, lack of social support and financial difficulties, all of which have been found to increase the risk of substance use [25,26]. It has also been observed that depressive symptoms may increase during the transitional phase from peri-menopause to post-menopause [27,28], possibly driving selfmedication strategies. Another reason may be that younger women tend to be more under family control, which likely reduces risk of exposure. Not pertaining to the lowest wealth index category was associated with higher khat use rates. One obvious reason is that buying khat requires a certain level of income [29]. Furthermore, chewing khat is often developed as part of doing business in higher income groups [13]. Also, khat chewing practices affect the economic status of society due to the fact that collectively, it contributes to the loss of thousands of acres of arable land and billions of hours of work as well as utilization of scarce resources to buy khat rather than acquiring nutritious foods and care for household members [30].
Those pertaining to the religion of Islam were at higher odds to chew khat as compared to those of Orthodox religion. This is consistent with earlier work in Ethiopia [31,32]. Chewing khat is a common practice and traditionally accepted in Islamic communities, associated with perceived benefits in concentration during work and prayer time [33,34].
Women with lower educational status had more risk of current khat chewing compared to women with higher educational status. The current study was in agreement with a study conducted in the Jazan region, in Saudi Arabia, which showed that illiteracy was associated with higher odds of chewing khat [35]. A possible reason is that a lower level of education is associated with lack of knowledge about the negative consequences of khat use [35]. Furthermore, women who had one child had higher odds of khat chewing as compared to women without children. A possible reason is joint use under the influence of a stable partner using khat. However, this would then only apply to the period of the first child, as women with more children again had reduced risk for chewing khat. One explanation is that with an increasing number of children, less resources would be available for the khat habit.  Given that khat use among women is prevalent, and with a mean number of 14.2 days of use in the last month, information campaigns are required in order to inform the public about harmful effect of khat use [36]. For those who are affected by khat use-related disorder and comorbid conditions, treatment services should be provided. Addiction services can be offered using the existing health system. For example, primary health care may introduce routine screening and treatment for substance use [36,37].
However, given that the majority of people do not have access to good quality of care; it is a timely need for the sustainable programs [38]. Therefore, it is a timely need for pertinent stakeholders of the Ethiopian public health care system to introduce novel approaches to generate financially sustainable programs for the early diagnosis, treatment and control of addiction through a group of well-trained health care providers.

Limitations
The results should be interpreted in the light of several limitation. First, this was a secondary data analysis which overlooked key variables that are relevant for khat use such as availability, affordability, family substance use and awareness of health consequences of khat use. Second, we cannot exclude social desirability-related underreporting of khat use behaviour, given that female khat use is still associated with a degree of stigma in Ethiopian society.

Conclusion
In conclusion, the lifetime rate of chewing khat among women in Ethiopia is substantial and associated with specific sociodemographic risks. These can be used in targeted public health efforts to control increasing levels of khat use among women and reduce the associated health and economic burden.