A Mixed Methods Study: Sex Differences in Experiences of Stigma Associated with Alcoholism and Alcohol Use Disorders Among Injury Patients in Tanzania

Abstract

Background

Each year, alcohol use causes 3.3 million deaths globally and accounts for nearly 30% of injuries treated at Kilimanjaro Christian Medical Center (KCMC) in Moshi, Tanzania. Prior research found significant stigma towards patients reporting alcohol use in general and among healthcare providers for this population.

Methods

This mixed-methods study aimed to identify sex-based perspectives of stigma among injury patients, family members, and local community advisory board members. Injury patients from the emergency room at KCMC were asked to complete surveys capturing consumption of alcohol, perceived stigma, and consequences of drinking. Patients who completed the survey, their family members, and members of a community advisory board were also recruited to take part in focus groups led by a trained bilingual research nurse. Data were analyzed using multiple linear regression and Wilcoxon rank sum tests with alpha level set at 0.05.

Results

Results showed that sex was a significant predictor of perceived discrimination (p = 0.037, SE =1.71(0.81)) but not for perceived devaluation (p = 0.667, SE = −0.38(0.89)). Focus groups revealed there were global negative perceptions of the amount of alcohol consumed as well as negative perceptions towards disclosure of alcohol use to healthcare providers. Sex differences in stigma emerged when participants were specifically asked about women and their alcohol consumption.

Conclusions

The findings of this study suggest there is an underlying sex difference, further stigmatizing women for alcohol use among the injury patient population at KCMC. Tanzanian women suffer from unequal access to healthcare and the stigmatization of alcohol-use likely increases this disparity.

Introduction

Excessive alcohol use is a global public health problem resulting in 3.3 million deaths each year and accounting for 5.1% of the global burden of disease (WHO, 2017a). Not only is excessive alcohol linked to an increased risk of developing communicable diseases and development or worsening of non-communicable diseases, it has a significant impact on mental health (WHO, 2018). Excessive alcohol use is a leading, avoidable risk factor for death and disability in sub-Saharan Africa (Lim et al., 2012). According to the World Health Organization (WHO), the sub-Saharan region suffered the greatest burden of disease and injury due to alcohol use, even when compared to countries with higher alcohol consumption (WHO, 2018). In Tanzania specifically, the WHO reports that 9.4 liters of alcohol are consumed per capita each year, which is higher than the overall 6.3 liters of alcohol in the total African region (WHO, 2018). The 2016 prevalence of heavy episodic drinking, categorized by consuming 5 or more drinks in one sitting, was noted to be 20.3% for the total population (males 33.4% and females 7.7%) in Tanzania (WHO, 2018). Recent data suggest that 30% of injured patients presenting to the emergency department (ED) at Kilimanjaro Christian Medical Center (KCMC) tested positive for alcohol on arrival (Staton et al., 2018b).

In developed countries, individuals with alcohol dependency are considered to be more heavily stigmatized than others with mental health issues (Schomerus et al., 2011). Stigma is defined as negatively labeling someone because of a culture or community’s ideas or beliefs (Link and Phelan, 2001). Stigma can ultimately affect a person’s identity and self-esteem (Link and Phelan, 2001). Alcohol-related stigma affects people both on a personal level and a community level through rejection, discrimination, and outside stereotypes (Schomerus et al., 2011). In addition, stigma related to alcohol use has a profound impact on people with alcohol use disorders (AUDs) and alcohol dependence, their families, and their care-seeking behavior (Hunter et al., 2017). There are different types of stigma: public stigma, perceived stigma, experienced (felt) stigma, and self stigma (Glass et al., 2013b). With regard to alcohol use, public stigma refers to how the community may view someone who excessively consumes alcohol (Hunter et al., 2017). For example, the community may assume that someone who drinks excessive alcohol is not a worthy contributor to society. Perceived stigma is the awareness of discrimination and devaluation in already stigmatized individuals. Discrimination and devaluation can affect an individual with AUD negatively since the awareness of stigma may cause lower rates of treatment (Glass et al., 2013b). Experienced or felt stigma is being exposed to discrimination or being devalued because of alcohol use (Glass et al., 2013b, Glass et al., 2013a). Lastly, self stigma is the internalization of discrimination and viewing onself as damaged because of alcohol use (Glass et al., 2013b).

In an international WHO study examining stigma towards substance use disorders across 16 countries, alcohol addiction was the fourth most stigmatized illness (Room et al., 2001). Although alcohol related stigma persists globally, there is minimal literature examining alcohol related stigma in low- and middle-income countries (LMICs). In the sub-Saharan region, such studies originate primarily from South Africa where the general population appears to stigmatize alcohol disorders over other substance abuse and mental disorders (Sorsdahl et al., 2012, Sorsdahl and Stein, 2010). Little to no data are available for stigma related to alcohol use in the Tanzanian culture. However, recent data suggest that there is a strong stigma towards complications stemming from excessive alcohol use, and healthcare providers at KCMC exhibit stigma toward people who report excessive alcohol use (Meier et al., 2020, Staton et al., 2018b). Gaining insight into what kind of stigma exists could help us understand how it impacts reporting of alcohol use as well as potential barriers to using alcohol-related interventions.

Sex is an important social determinant of health. Inequalities can ultimately damage the health of both men and women, especially in poverty-stricken areas (WHO, 2017b). Robust studies from high-income countries (HICs) have revealed differences in perceived alcohol-related stigma when comparing men and women with AUDs (Keyes et al., 2010). However, research in LMICs regarding stigma and its association with AUDs is still lacking (Semrau et al., 2015). Studies from sub-Saharan Africa and India have revealed a high prevalence of internalized stigma among individuals with AUDs, and an association between internalized stigma and reduced treatment seeking behavior (Zewdu et al., 2019, Rathod et al., 2015, Nalwadda et al., 2018). Yet, these studies do not evaluate differences in AUD-related stigma between sexes. The cultures and environments surrounding alcohol vary more widely across LMICs than across HICs (Walls et al., 2020) Consequently, elucidating stigma patterns among individuals with AUDs in specific low-income regions is vital to reducing alcohol-related harm in LMICs. Given recent data about an abundance of stigma about alcohol use in the Tanzanian society (Staton et al., 2018b, Meier et al., 2020), our goal is to analyze differences in the perception of alcohol-related stigma between women and men suffering an injury.

Method

Ethics

This study was approved by the Ethics Committee at the University of New England, Duke University Medical Center Institutional Review Board (IRB), the Ethics Committee of the Kilimanjaro Christian Medical Center, and the National Institute for Medical Research in Tanzania. All participants provided written informed consent to participate in the study and it was approved by Duke University Medical Center IRB and the Ethics Committee of KCMC.

Setting

This study was conducted at KCMC in Moshi, Tanzania. Located in the foothills of Mount Kilimanjaro, KCMC is a referral hospital for Northern Tanzania and serves nearly 15 million urban and rural people (approximately 70–100 patients each day) (KCMC). Of note, the ED sees nearly 2,000 injury patients per year, nearly 30% of whom are positive for alcohol (Staton et al., 2018a). Moshi is home to nearly 180,000 people and is comprised of Chagga, Pare, and Maasai ethnic groups. The primary spoken language is Swahili (TNBS, 2018, TDHSP). The unemployment rate for this area is 19%, and the majority of residents hold a primary education (TDHSP).

Participants

Participants for the survey were selected from the Developing a Brief Intervention (BI) For Alcohol in Tanzania study that took place from 2015 to 2019 at KCMC (ClinicalTrials.gov). People presenting to the KCMC ED for care of acute injuries were prospectively enrolled. Inclusion criteria were: ≥ 18 years of age; suffered an acute injury; medically stable; and able to speak Swahili. Patients were excluded if they were unable to complete the surveys or focus group because they were clinically unstable due to the severity of their injury.

The target population for focus groups included injury patients included in the quantitative survey, family members of these patients, and members of a community advisory board. Injury patients and their family members were identified in the ED and approached for participation and informed consent after being treated or considered stable. Community advisory board member focus groups were held just before their monthly meeting for all interested.

Data collection procedures

The Perceived Alcohol Stigma Scale (PAS) has not been validated in Tanzanian Swahili, so it was translated and back-translated to Swahili by trained research assistants fluent in both Tanzanian Swahili and English. The Drinker Inventory of Consequences (DrInC) and the Alcohol Use Disorders Identification Test (AUDIT) are both validated surveys for Tanzanian Swahili (Vissoci et al., 2018, Zhao et al., 2018). All surveys were administered in a quiet room near the ED at KCMC verbally in Swahili, by a trained, bilingual research nurse. Surveys were administered verbally due to the variability in literacy rates among patients. Data for surveys were collected by hand and then entered manually into a REDCap database. The principal investigator (CAS) reviewed data for accuracy.

Trained bilingual nurses conducted the focus groups in Swahili using an interview guide. Once 5–10 patients or family members were recruited, focus groups were formed and scheduled. Focus groups took place in a quiet room near the ED and lasted 45–60 minutes. Family member focus groups were separate from patient focus groups. The focus groups with the community advisory board were conducted in the same manner as the patient and family member groups. Participants were recruited before the monthly community advisory board meeting. All focus groups were audiotaped. The audio recording was transcribed and translated into English.

Instruments

Perceived Alcohol Stigma Scale

The PAS scale is an adaptation of the Perceived Devaluation-Discrimination Scale which is used to assess perceived stigma around mental disorders (Link, 1987). The PAS scale measures perceived stigma around AUDs and assesses two constructs: discrimination (how the public stigmatizes alcohol use) and devaluation (external devaluation). The scale is 12 questions in a 6-point Likert-type scale format (Glass et al., 2013a). Seven questions assess discrimination (i.e., “Most people would willingly accept a former alcoholic as a close friend.”) (Glass et al., 2013a). The other questions assess devaluation (i.e., “Most people feel that entering alcohol treatment is a sign of personal failure.”) (Glass et al., 2013a). Six questions were listed with reverse wording to ensure participants understood the questions to reduce response bias (Glass et al., 2013a). For reporting, scores on the scale range from 1 to 6, with higher scores indicating a higher PAS.

DrInC

The DrInC Scale utilizes 50 questions to assess for negative consequences of alcohol use. The DrInC measures 5 subscales that summarize the following consequences: physical, intrapersonal, social responsibility, interpersonal, and impulse control (Miller et al., 1995). The physical consequences subscale has 8 questions that address the physical signs of excessive drinking. These include items such as having a hangover, eating habits, and appearance (Miller et al., 1995). The intrapersonal subscale also has 8 questions and captures the subjects’ perceptions of excessive alcohol use, including interest and activities, and poor moral and spiritual perceptions. The 7 questions in the social responsibility subscale capture others’ perceptions, such as problems at work or school, getting in trouble, and failing to meet the expectations of others. Ten questions make up the interpersonal consequences subcategory, and they are associated with relationships, such as loss of friendships, harmful family relationships, and loss of love. Any remaining questions that did not fit in the first 4 categories were grouped into the impulse control subcategory. This subcategory scales consists of use of other drugs, overeating, getting into physical fights, and causing injury to others. Lastly, 5 questions were considered to be control questions and were unassociated with alcohol use problems (Miller et al., 1995).

AUDIT

The AUDIT was developed by the WHO as a screening tool for both harmful and hazardous alcohol use (Babor et al., 2001). This scale has 10 questions that capture the frequency of alcohol consumption, risky alcohol consumption, and harmful behaviors associated with alcohol consumption(Babor et al., 2001). Each answer on the AUDIT is scored from 0 to 4 on a 5-point Likert-type scale; a high score indicates the likelihood of problematic drinking and dependence (Vissoci et al., 2018, Babor et al., 2001). There are 4 suggested risk zones for the AUDIT score. These zones indicate the level of harm drinking has on health and are as follows: low risk (score of 0–7), risky (score of 8–15), harmful (score of 16–19), and severe (score of 20+) (Babor et al., 2001). For the purposes of this study, scores of 0 classify participants as abstainers, < 8 classifies them as low-risk drinkers, and ≥ 8 classifies as them as risky drinkers.

Focus Groups

Injury patients, their family members, and community advisory board (CAB) members took part in semi-structured focus groups to assess stigma related to drinking alcohol. Questions targeted stigma related to people who drive while intoxicated; differences in locations where men and women drink; differences in the amount of alcohol women drink compared to men; acceptability of alcohol use, especially differentially by sex; the choice to disclose alcohol use to medical professionals; and how drunkenness is perceived. Qualitative questions were guided by quantitative results as possible. Injury patients were clinically sober and medically stable, ≥ 18 years of age, and were able to communicate in either Swahili or English at enrollment. Family members and CAB members were enrolled to gain a community view into perceptions of alcohol use.

Analysis

Descriptive statistics were used to report sex (frequencies and percentages), age categories, alcohol intake 6 hours prior to injury, and total AUDIT score (mean and standard deviation). Multiple linear regression analysis was used to understand whether or not perceived devaluation and perceived discrimination can be predicted by sex and adjusting for age, total AUDIT score, abstainers of alcohol consumption, total DrInC score, and reported consumption of alcohol within 6 hours of injury. Wilcoxon ranked sum test, which does not require the assumption of normal distribution, was used for the individual PAS survey questions to test differences in score by sex. Significant values for both multiple linear regression and Wilcoxon ranked sum tests were set at 0.05. Missing value imputation was not conducted. Analyses were conducted using Stata/SE 15.1 (Statcorp, 2017).

Focus groups were analyzed using a thematic content analysis approach, where themes emerged as stigma categories, and representative quotations for each of the identified themes were selected. Focus groups were transcribed into a spreadsheet organized by questions, participant narratives, content, and stigma themes. All focus groups were coded independently by 3 coders (SMG, MCCM, AZ). Discrepancies were resolved through discussion. Emerging themes were validated by a Tanzanian collaborator (FK) who reviewed all thematic codes and narratives. Some Swahili words do not translate directly to English; therefore, quotes that have improved English have been put within brackets.

Results

Participants

Three hundred forty injury patients enrolled and completed the quantitative study. The majority of participants were males (n = 277, 81.47%) between the ages of 26 to 35 years (n = 86, 31.05%). Only 18 (5%) participants tested positive for alcohol at the time of enrollment, although 51 (15%) participants reported using alcohol within 6 hours of injury. The mean AUDIT score was 7.17 (SD 8.11), indicating that on average, for this study population, participants are classified as low-risk drinkers.

Perceived Alcohol Stigma

Table 2 lists the mean and standard deviation (SD) of itemized responses. The mean and SD perceived alcohol stigma scale score was 51.90 (7.76) for the study population. There was no significant association with sex for the overall stigma scale. The mean total PAS score for men was 4.32 (SD= 0.67) and for women 4.34 (SD=0.57); the mean total PAS score was 4.32 (SD=0.65). When stratifying the PAS, perceived devaluation total score was 4.21 (SD=0.98) for men and 4.15 (SD=0.95) for women, indicating men had an overall higher score (greater than 3) for perceived devaluation. With regards to perceived discrimination, the average score was 4.39 (SD=0.74) for men and 4.48 (SD=0.64) for women, indicating women had an overall higher score. Table 2 depicts the mean score for each question by sex. A significant difference was observed in men and women for several individual PAS questions. Women are more likely to have a higher devaluation or discrimination score when responding to the following questions: “Most people believe that a former alcoholic is just as trustworthy as the average person,” “Most people think less of a person who has been in alcohol treatment;” and “Most employers will pass over the application of a former alcoholic in favor of another applicant.”

Table 2.

PAS by Sex

		Total Mean (SD)	Male Mean (SD)	Female Mean (SD)	/Mann Whit WilCox sig
PAS TOTAL		51.90 (7.76)	51.81 (7.94)	52.11 (6.88)	0.9681
PAS Average		4.32 (0.65)	4.32 (0.67)	4.34 (0.57)	0.9681
Devaluation		4.20 (0.98)	4.21 (0.98)	4.15 (0.95)	0.4555
Discrimination		4.41 (0.72)	4.39 (0.74)	4.48 (0.64)	0.5703
Q1	Most people believe that a person who has had alcohol treatment is just as intelligent as the average person	3.86 (2.41)	3.91 (2.39)	3.59 (2.47)	0.3810
Q2	Most people believe that a former alcoholic is just as trustworthy as the average person	4.76 (2.01)	4.86 (1.96)	4.28 (2.17)	0.0131**
Q3	Most people feel that entering alcohol treatment is a sign of personal failure	2.83 (2.31)	2.83 (2.30)	2.76 (2.34)	0.704
Q4	Most people think less of a person who has been in alcohol treatment	4.71 (2.08)	4.58 (2.14)	5.2 3(1.70)	0.0180**
Q5	Once they know a person was in alcohol treatment, most people will take his or her opinion less seriously	4.82 (1.95)	4.70 (1.96)	4.89 (1.93)	0.7052
Q6	Most people would willingly accept a former alcoholic as a close friend	5.45 (1.44)	5.48 (1.40)	5.29 (1.61)	0.2056
Q7	Most people would accept a fully recovered former alcoholic as a teacher of young children in a public school	5.17 (1.75)	5.18 (1.76)	5.11 (1.75)	0.3256
Q8	Most people would not hire a former alcoholic to take care of their children, even if he or she had been sober for some time	3.81 (2.22)	3.72 (2.22)	4.27 (2.20)	0.0758
Q9	Most employers will hire a former alcoholic if he or she is qualified for the job	5.47 (1.42)	5.53 (1.35)	5.21 (1.67)	0.0502
Q10	Most employers will pass over the application of a former alcoholic in favor of another applicant	4.69 (2.10)	4.51 (2.18)	5.43 (1.50)	0.0015**
Q11	Most people in my community would treat a former alcoholic just as they would treat anyone else	5.57 (1.35)	4.51 (2.18)	5.43 (1.50)	0.0762
Q12	Most young women would be reluctant to date a man who has been hospitalized for alcoholism	4.54 (2.09)	4.56 (2.09)	4.44 (2.12)	0.5401

^*P-values of Wilcoxon rank sum test reported

There were no associations of sex with perceived devaluation, even when adjusting for age, total AUDIT score, total DrInC score, and participant consumption of alcohol. However, there was a positive association for sex with perceived discrimination. Women were more likely to report higher discrimination scores than men when adjusting for age, total AUDIT score, total DrInC score, and participant reported consumption of alcohol.

Focus groups

In total, 101 participants, including 39 patients (26 males), 50 family members (29 males), and 12 CAB members (8 males), took part in 12 semi-structured focus groups (FGs). Each FG had 6–10 participants. As our sample has more men, our overall result is likely representative of a male perspective.

Focus groups identified 3 primary themes: (1) negative perceptions associated with amount of alcohol used, (2) negative perceptions of disclosure of alcohol use to healthcare providers, (3) sex-related stigma toward alcohol use. Within stigma, two sub-themes emerged: enacted stigma (discrimination) and felt stigma (devaluation).

Overall Negative Perceptions Associated with Amount of Alcohol Used

When participants were asked about how people who drank too much in public were viewed, enacted stigma was reported across all focus groups. Most participants reported that people who drink excessive alcohol are a loss to the community or even considered bad people. As one participant explains, people who drink excessively in public are shamed: “First of all, you become disrespected, on your street, people will not respect you and even children will not respect you. [If he is drunk on the road, people will not help him, they will just leave him, therefore it is a shame.]”(Patient 2, FG #3)

A family member expressed that even the average alcohol drinker would be stigmatized:

“Myself, I can stigmatize someone who drinks alcohol even if he/she only drinks half a glass.” (Family member 5, FG #2)

It was also common for participants to say that people who drink excessive alcohol are disrespected by the community. As stated by one participant: “If someone is drinking too much alcohol by my side, I take him like he is disrespecting himself…..even his children cannot respect him because of alcohol. Even his family is being humiliated due to his alcohol use…...people don’t take it serious if you are a drunkard.”(Patient 5, FG #3)

Negative Perceptions of Disclosure Of Alcohol Use To Healthcare Providers

When participants were asked about disclosing how much alcohol they had consumed, felt- stigma was commonly expressed, especially with regards to telling doctors and nurses. For example, one CAB member stated: “...people who are drinking alcohol, they feel that society underrates them, so he will not want to show that he is of low level in front of the doctor.” (CAB member 7, FG# 1)

In fact, when asked about reporting alcohol use to doctors and nurses, many participants replied they simply would not be honest about how much alcohol they had consumed. As described by one participant, people would not tell a doctor or nurse how much because they would humiliate themselves: “.....Cannot be open due to the environment he/she is. [Sometimes he/she will feel like they humiliate himself/herself by being open about the amount of alcohol used. Because if he/she is drunk, then they can do other things that are not right to the surrounding society. Therefore he/she is afraid to be ashamed.]” (Family member 5, FG #1)

It is important to note that the majority of participants stated they would not be honest to a nurse or doctor about the amount of alcohol they had consumed. Not only did participants fail to disclose alcohol amounts to doctors and nurses, they did not want to disclose this to their wives or others because it is a “secret”:“[Because it is his secret and he is the only one who knows how much alcohol and the type of alcohol he takes. Because there are some people who drink alcohol and when they go home and their wife, she will ask how much alcohol they have had to drink, they tell their wife that they have not drank any alcohol.] Now if he is not able to tell his wife if he uses alcohol, then he cannot publicly admit it if he is asked about his alcohol use.” (Patient 2, FG #1)

Sex-Related Stigma Toward Alcohol Use

In the last two patient and family member focus groups, participants were specifically asked about women and their consumption of alcohol based on preliminary quantitative data showing potentially report higher discrimination scores from women. When asked whether it was more acceptable for a man or woman to drink alcohol, a clear sex based difference in stigma emerged; a family member stated:“[....a woman who is going to the bar will be viewed as a prostitute. [...] If she is at a bar that is far from her home, we say this woman is not good. [...] In this society she cannot be married.]” (Family member 8, FG #6)

When asked if women who drank alcohol were viewed differently than men who drank alcohol, enacted stigma emerged within the groups. Within this population, the common viewpoint is that a woman should stay home with her children, otherwise she is not respected. According to one patient:“[When a woman appears to be a drunk person in the community, it is not a good thing because she has more family responsibilities than the father. Therefore, it is a disappointment in the community for a woman to be seen a drunk like a man.]” (Patient 8, FG #5)

According to one family member, if a woman is seen drinking alcohol the community will talk badly about her: “[She seems as not behaving well in society and people must discuss her badly.]” (Family member 3, FG #5)

Mixed Methods Results Summary:

While in general, our quantitative results showed a strong overall perceived alcohol stigma in this population there was no statistically significant sex based difference. Yet, our adjusted model of the PAS descrimination scores found women were more likely to report higher discrimination scores than men. These results are mirrored in our qualitative data where sex appeared to be a significant factor, specifically through traditional sex roles, and displayed through public, community, perceived and expereinced stigma.

Discussion

This study set out to describe sex differences of perceived stigma within the injury population at KCMC. This is the first study, to our knowledge, to assess whether or not sex is a predictor in stigma in alcohol use–related injuries in Tanzania. Overall, this study demonstrated sex differences in stigma associated with alcohol use–related injuries in Tanzania; however, it is intertwined with sex roles.

The injury population at KCMC is primarily male because men are more likely to present for injuries sustained after alcohol consumption (Staton et al., 2018a). This is important as we interpret the results of this data as being driven from the male perspective. However, even though there were few injured women in our population, being female was associated with higher perceived discrimination. Individual items on the PAS demonstrated that women were more likely to report a mix of discrimination and devaluation, specifically relating to hiring former alcoholics, opinions on former alcoholics being as intelligent as the average person, and opinions on being trustworthy and a former alcoholic. Although we found differences between men and women with regard to specific PAS items, we did not find a difference with regard to overall PAS scores. However, this does not mean such differences do not exist. Language disparities as well as underlying sex roles inherent in this population may have obscured differences in perceived stigma between sexes. In addition, very few studies have implemented the PAS scale in the Tanzanian setting or other swahili speaking populations. Consequently, cultural validity of the PAS in our study setting has yet to be demonstrated. That being said, we did observe a sex difference in the perception of discrimination due to alcohol use, and given the overall qualitative results supported sex related stigma we believe that our quantitative results, albeit small, are likely real, but further study in this area with a more sensitive and culturally-validated discriminate tool is warranted.

This population held stigma and negative perceptions toward the amount of alcohol consumed. We can assume that this stigma was more related to the excessive amount of alcohol consumed by men. The response was similar when participants were asked about reporting the amount of alcohol consumed to a healthcare provider. Only when participants were asked specifically about women and alcohol consumption did sex roles emerge and women appeared to be more heavily stigmatized than the males. In our population, women were described as being prostitutes if they consume alcohol in a bar, and even seen as unable to be married. This is similar to how women with other illnesses are viewed in Tanzania. According to Miller et al., women are described as promiscuous if they test positive for tuberculosis (Miller et al., 2017). Similarly, in human immunodeficiency virus (HIV) stigma research, women report significantly higher rates of enacted and self stigma than their male counterparts in Tanzania (Parcesepe et al., 2019). Future research is required to better understand women’s roles in alcohol use–related stigma. Women and girls in Tanzania are considered to be some of the most marginalized people in sub-Saharan Africa (USAID, 2019). Women not only suffer discrimination with regard to employment, education, and decision-making, they also suffer from unequal access to healthcare (UNDP, 2019). In future research, it is important to find ways of including more women to better understand the depth of stigma through the eyes of women in this culture.

From a broader perspective, perceived stigma in our setting appears to be much higher than perceived stigma in settings from HICs. In two German population surveys measuring perceived alcohol-related stigma with the PAS scale, average total PAS scores were 41.3 (SD 9.4) among a sample of 1,022 people in the year 1990, and 37.2 (SD 8.1) among a sample of 967 people in the year 2011 (Schomerus et al., 2014). Similarly, a nationally representative survey of 34,653 individuals in the United States from the year 2005 revealed a total average PAS score of 38.8 (Smith et al., 2010). The total average PAS score among all respondents in our study was 51.9 (SD 7.8), suggesting a stronger presence of perceived alcohol-related stigma in Tanzania when compared to other regions of the world. Most research regarding alcohol-related stigma has been conducted in HICs. The large difference in perceived alcohol related stigma between our low-income setting and other high-income countries highlights the need for continued investigation in LMICs. Understanding regional stigma patterns surrounding alcohol use in LMICs is key to developing effective interventions that address AUDs.

Limitations

These findings should be interpreted with caution. Most of our respondents were men, therefore the views of women are underrepresented. This limits the number of patients who would have tested positive for alcohol prior to injury. The PAS scale has not been culturally validated in Tanzanian Swahili. The participants rated as mostly low-risk drinkers. Different views might be elicited by high-risk drinkers, those who are most likely to have personal experience with feeling stigmatized by their alcohol use. Lastly, given that stigma is feeling discriminated against, participants may not have accurately reported a true interpretation of this within the scales and the focus groups, so there may be social desirability biases.

Conclusions

In summary, sex is a predictor of perceived discrimination among injury patients at KCMC. However, the mixed results from the PAS data indicate that this scale might not be appropriate for this culture. The focus group data provide evidence that stigma exists for people who consume alcohol within this population. It is unsurprising that participants expressed negative perceptions of disclosure of alcohol to healthcare providers. Women are the most stigmatized with regard to consumption of alcohol, as this community would see them as failures, prostitutes, and disappointments. Further research is warranted to further describe perceptions of stigma specifically for women and the impact of this perception on seeking care and for alcohol use behaviors.

Figure 1.

Study Enrollment Processes

Table 1.

Characteristics of the Study Population

Variable		Total n (%)	Male n (%)	Female n (%)
Participants		340 (100)	277 (81.47)	63 (18.53)
Age Categories	18 to 25 years	89 (26.10)	73 (26.35)	16 (25.40)
	26 to 35 years	102 (29.91)	86 (31.05)	15 (23.81)
	36 to 45 years	68 (19.94)	58 (20.94)	10 (15.87)
	46 to 55 years	46 (13.49)	34 (12.27)	12 (19.05)
	56 years and older	36 (10.56)	26 (9.39)	10 (15.87)
AUDIT	Mean (SD)	7.17 (8.11)	7.28 (8.65)	4.43 (5.35)
Alcohol Pos	Mean (SD)	18 (5.29)	16 (5.80)	2 (3.17)
Reported alcohol use 6 hours prior to injury	Mean (SD)	51 (15)	44 (15.94)	7 (11.11)

Table 3.

Association between multiple linear regression of perceived devaluation by sex, adjusted for age AUDIT, DrInC, and consumption

	Coefficient (SE)	95% CI	P-value
PAS
Female	1.32(1.32)	−1.29,3.93	0.318
Age	0.08(0.04)	0.01,0.15	0.025**
AUDIT score	−0.00(0.09)	−0.19,0.18	0.996
Abstainers	−0.76(1.51)	−3.72,2.21	0.617
Alcohol-related consequence	0.04(0.029)	−0.14,0.09	0.143
Devaluation
Female	−0.38(0.89)	−2.13,1.36	0.667
Age	0.08(0.02)	0.04,0.13	0.001**
AUDIT score	−0.01(0.06)	−0.13,0.011	0.839
Abstainers	0.39(1.00)	−1.59,2.37	0.699
Alcohol-related consequence	0.01(0.19)	−0.02,0.05	0.482
Discrimination
Female	1.71(0.81)	0.11,3.30	0.037**
Age	−0.00(0.22)	−0.046,0.04	0.888
AUDIT score	0.01(0.06)	−0.10,0.12	0.879
Abstainers	−1.14(0.92)	−2.96,0.67	0.216
Alcohol-related consequence	0.03(0.18)	−0.01,0.06	0.105