Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Dec 1.
Published in final edited form as: Glob Public Health. 2022 Aug 7;17(12):3583–3595. doi: 10.1080/17441692.2022.2105377

Validating a Measure of Anticipated Sex Work-Related Stigma among Male and Female Sex Workers in Kenya

Emmanuel Oga a, Melissa A Stockton b,c,*, Christin Stewart d, John Kraemer d,e, Joshua Kimani f, David Mbote g, Stella Njuguna h, Laura Nyblade d
PMCID: PMC10242293  NIHMSID: NIHMS1826739  PMID: 35938397

Abstract

Sex workers face different types of sex work-related stigma, which may include anticipated, perceived, experienced, or internalized stigma. Sex work stigma can discourage health care seeking and hamper STI and HIV prevention and treatment efforts. There is a paucity of validated sex work-related stigma measures, and this limits the ability to study the stigma associated with sex work. A cross-sectional survey was conducted that measured anticipated sex work-related stigma among male and female sex workers in Kenya (N=729). We examined the construct validity and reliability of the anticipated stigma items to establish a conceptually and statistically valid scale. Our analysis supported a 15-item scale measuring five anticipated sex work stigma domains: gossip and verbal abuse from family; gossip and verbal abuse from healthcare workers; gossip and verbal abuse from friends and community; physical abuse; and exclusion. The scale demonstrated good face, content, and construct validity. Reliability was good for all subscales and the overall scale. The scale demonstrated good model fit statistics and good standardized factor loadings. The availability of valid and reliable stigma measures will enhance efforts to characterize and address stigma among sex workers and ultimately support the protection, health and well-being of this vulnerable population.

Keywords: Sex work-related stigma, anticipated stigma, validation, sex workers, sub-Saharan Africa

Introduction

Stigma is a complex social process that involves marking and discrediting a person or group based on a real or perceived attribute such as identity, occupation, behavior, or membership to a group and is linked to negative health outcomes for stigmatized individuals (Hatzenbuehler, Phelan, & Link, 2013; Link & Phelan, 2001; Weiss, Ramakrishna, & Somma, 2006). Much of stigma research has largely focused on HIV-related stigma, with less attention to other types of stigma affecting specific populations, including sex workers, who often bear a larger burden of HIV risk (Logie, James, Tharao, & Loutfy, 2011). Research has linked stigma to negative health outcomes such as depression, anxiety, chronic pain, and morbidity related to lower levels of medication adherence (Fitzgerald-Husek et al., 2017; Lancaster, Cernigliaro, Zulliger, & Fleming, 2016; Nyblade, Mingkwan, & Stockton, 2021; Stockton et al., 2020). Stigma is also linked to decreased uptake of health services and is a significant barrier to HIV prevention, treatment, and care (Alvarez-Uria, Pakam, Midde, & Naik, 2013; Corrigan, Larson, & Ruesch, 2009; Govindasamy, Ford, & Kranzer, 2012; Katz et al., 2013; Mahajan et al., 2008; Nyblade et al., 2017). The link between stigma and uptake of health services is particularly important for sex workers. Female sex workers (FSW), globally, are 13 times more likely than the general population to acquire HIV (UNAIDS, 2018). Similarly, studies have shown a higher burden of HIV among male sex workers (MSW) than other men who have sex with men (MSM), who are 26 times more likely than the general population to acquire HIV (S. D. Baral et al., 2015; UNAIDS, 2020).

Among sex workers, different forms of stigma have been identified including experienced, internalized, perceived and anticipated (Fitzgerald-Husek et al., 2017; Grosso et al., 2019). Experienced stigma refers to a sex worker’s experience of manifested stigma in the form of discrimination by others (Fitzgerald-Husek et al., 2017). Internalized (self) stigma is when a sex worker accepts negative judgments or attitudes applied to them as true and feels devalued as a result (UNAIDS, 2018). The terms perceived and anticipated stigma are sometimes used synonymously in research, but the concepts are different (Moore, Stuewig, & Tangney, 2013). Perceived stigma refers to a sex worker’s awareness of stigma towards sex workers, whereas anticipated stigma – or fear of stigma and/or discrimination – refers to a sex worker’s expectation or fear that they may personally experience discrimination (Moore et al., 2013). Finally, sex workers may be subjected to intersectional stigma, in particular based on sex work and HIV, given the assumption of association between sex work and HIV transmission in many places (Fitzgerald-Husek et al., 2017; Logie et al., 2011; Nyblade et al., 2021; Tsang et al., 2019; J. M. Turan et al., 2019).

With growing recognition of the importance of stigma and the associated outcomes, there is a need for valid and reliable measures of sex work stigma to assess its prevalence and sources, inform stigma mitigation interventions, and ultimately evaluate the impact of such interventions; however, the existing literature is limited (Liu et al., 2011). A systematic review of studies measuring stigma towards sex workers and men who have sex with men found very few studies that measured sex work stigma and even fewer that used validated stigma metrics (Fitzgerald-Husek et al., 2017). Studies measuring stigma towards men who have sex with men and sex workers were commonly conducted in high-income countries with far fewer in low- and middle-income countries, or in the sub-Saharan context (Fitzgerald-Husek et al., 2017). Further, studies that used validated scales often measured perceived stigma; however, theory suggests that anticipated stigma may be more impactful than perceived stigma because it suggests that a person expects to personally experience negative consequences of stigma (Fitzgerald-Husek et al., 2017; Golub & Gamarel, 2013; Moore et al., 2013; B. Turan et al., 2017). Therefore, the present study developed and validated an instrument to measure anticipated sex work stigma among male and female sex workers in Kenya.

Methods

Participants

We conducted a cross-sectional study of 729 sex workers (497 FSW and 232 MSW) recruited from four Kenyan cities—Nairobi, Homabay, Busia, and Kitui—using snowball sampling. These sites were chosen to include a variety of settings including rural, urban, and transit corridors. Sex worker partner organizations at these sites identified initial participants who were given coupons to recruit other potential participants. Eligible participants had to be age 18 or older, acquire most of their income from sex work and be a resident in the study site location for at least six months preceding the study. Participants were given 500 Kenyan Shillings (slightly more than $5 USD) to reimburse their time and travel for participation in the study. More detail on the methodology for this study was previously published.(Nyblade et al., 2015; Nyblade et al., 2017). Informed consent was obtained from each participant and we adhered to all ethical guidelines as required for conducting human research. The study was approved by the Institutional Review Boards of the Kenya Medical Research Institute (KEMRI) (Protocol Number 465) and Health Media Lab.

Scale Development

Due to the lack of published sex worker stigma scales available at the time of this study (2013), the team developed the 23 anticipated sex work items for this study through consultations with researchers working on stigma measurement with sex workers in other African countries, as well as on published HIV stigma scales. While recognition of the importance of measuring anticipated stigma was emerging, a validated HIV anticipated stigma was not available for adaptation for this study. Hence the team developed the anticipated sex work stigma measures informed by several validated HIV perceived, experienced and vicarious stigma scales developed and tested largely in African countries (Holzemer et al., 2007; Kalichman et al., 2009; Kalichman et al., 2005; Nyblade et al., 2005; Steward, Bharat, Ramakrishna, Heylen, & Ekstrand, 2013; Visser, Kershaw, Makin, & Forsyth, 2008). With an eye to interventions with specific target groups, items asked not only about how frequently the participant feared that they would experience certain forms of stigma (e.g., gossip, verbal abuse, physical abuse, and exclusion), but also from what sources (i.e., family, friends, neighbors/community, healthcare workers, police, and city/town council askaris) they anticipated (feared) that form of stigma because they sell sex, regardless of whether or not these things ever actually happened to them. Participants scored each item using a 5-point scale, from 0–4, corresponding with the frequency with which participants reported fear of stigma. Face and content validity of the anticipated stigma items were determined through review of the items to determine whether as a set of items they appeared to appropriately measure anticipated sex work stigma (face validity) and were representative of the entire domain of anticipated sex worker stigma (content validity) (Lebet, Asaro, Zuppa, & Curley, 2018; Mosier, 1947; Nevo, 1985; Salkind, 2010). Both face and content validity were assessed by the research team, which included international and Kenyan stigma experts, leaders of female and male sex worker partner organizations with lived experience of sex work stigma, and Kenyan university and government researchers with a long history of research with, and service delivery to, sex workers. We then solicited additional feedback from members of local partner sex worker organizations in each of the study sites to further check face and content validity and tested the items with a small sample of participants during interviewer training, with a focus on comprehension of question intent, further tailoring of items for cultural saliency and to refine the Swahili translation for clarity. Based on these exercises, minor changes were made before finalization.

Procedure

Third-party trained interviewers administered the survey in January 2015 through face-to-face interviews in either English or Kiswahili in private offices after participants provided informed consent. We chose to hire third-party trained Kenyan interviewers through a competitive process, rather than peers, to minimize response bias and further protect participants’ confidentiality. However, this did not preclude sex workers applying and the interview team included several sex workers who only interviewed respondents of the opposite sex in locations away from their usual place of residence and work. These interviewers attended a five-day long training on survey administration, ethical considerations, stigma, and stigma reduction.

Construct Validity

Confirmatory Factor Analysis

First, we used exploratory factor analysis (EFA) to make an initial decision about the number of factors. Then we used confirmatory factor analysis (CFA) using structural equation modelling (SEM) to examine the scale. Item responses were assigned numerical equivalents and treated as interval data (Items scored: never=0; not in last 12 months but have before=1; once in the last 12 months=2; a few times in the last 12 months=3; and often in the last 12 months=4). Factor loadings (standardized coefficients) were assessed and a value of ≥ 0.60 was the cut-off for item retention.

To assess model goodness of fit, we computed fit indices including chi-square, root mean square error of approximation (RMSEA), comparative fit index (CFI), and standardized root mean square residual (SRMR) goodness-of-fit statistics (Kline, 2015). Satorra-Bentler adjustments were used for the fit indices (Kline, 2015). The chi-square goodness-of-fit statistic represents the difference between the expected and observed covariance matrices, and a value close to zero indicates little difference, with a probability value greater than 0.05 (Hu & Bentler, 1999). RMSEA is a measure of the average of the residual variance and covariance. RMSEA values at or less than 0.06 were considered desirable (Hu & Bentler, 1999). CFI is an index that fall between 0 and 1 and values greater than 0.90 were considered to be indicators of good fitting models (Hu & Bentler, 1999). Finally, SRMR is the square-root of the value obtained from subtracting the residuals of the sample covariance matrix from the hypothesized model and SRMR ≤ 0.05 were considered acceptable (Hu & Bentler, 1999).

Convergent, Discriminant, and Known-Group Validity

To further assess construct validity, we assessed convergent and discriminant validity. Convergent validity measures the degree to which two constructs that we hypothesize are theoretically related, are in fact related. Whereas discriminant validity measures the degree to which items which theoretically should not related are, in fact, unrelated. We measured convergent and discriminant validity by measuring average variance extracted (AVE) and squared correlations (SC). AVE values above 0.5 were considered acceptable and providing evidence of convergent validity; while for discriminant validity, we required the AVE of any two constructs be greater than the SC between the two constructs. We also assessed known-group validity (KGV) using by sex, depressive symptoms, education and healthcare avoidance or delay in the last year to see how anticipated stigma measured using our scale compared to the literature. Depressive symptoms were measured using the Patient Health Questionnaire-9 (PHQ-9), which consists of nine questions on the frequency of depressive symptoms experienced over the past two weeks. Responses to each of the nine questions were totaled; the total score could range from 0 to 27. PHQ-9 scores of 0–4, 5–9, 10–14, 15–19, and 20–27 were considered indicative of minimal, mild, moderate, moderately severe, and severe depressive symptoms, respectively (Kroenke, Spitzer, & Williams, 2001). Healthcare avoidance or delay over the past year was measured in the questionnaire by asking respondents if they (or their children) needed health services in the prior 12 month and avoided or delayed healthcare utilization.

Reliability

We computed Cronbach’s alphas to assess the internal consistency of items.

Results

Participant Characteristics

The study participants had a mean (SD) age of 27.9 (6.3). Around half of the participants (46%) had only a primary education or less, were single/never married (53%) and were residents of Nairobi (45%). Participants were mostly female (68%) and HIV negative (66%). On average, participants had been a sex worker for about five years, mean (SD) = 5.4 (4.6), engaged in sex work about four days per week, mean (SD) = 4.4 (1.8), and had a median income of 1000 Kenyan Shillings (about $11 USD at the time) per week. (Table 1)

Table 1.

Sample Characteristics (n=729)

Age in years, mean (SD) 27.9 (6.3)
Sex, n (%)
Female 497 (68.2)
Male 232 (31.8)
Education, n (%)
Primary or less 334 (46.0)
Secondary 315 (43.4)
Tertiary 77 (10.6)
Income (weekly), median (Q1-Q3) 1000 Kenyan Shillings (800–2000)
Marital status, n (%)
Single/Never Married 384 (53.0)
Married 35 (4.8)
Divorced 219 (30.2)
Widowed 42 (5.8)
Partner 45 (6.21)
Years in sex work, mean (SD) 5.4 (4.6)
Residence, n (%)
Nairobi 331 (45.4)
Busia 155 (21.3)
Kitui 76 (10.4)
Homabay 167 (22.9)
Days of sex work per week, mean (SD) 4.4 (1.8)
HIV status (self-reported), n (%)
Positive 172 (23.6)
Negative 484 (66.4)
No response 73 (10.0)
Open in a new tab

Prevalence of Anticipated Stigma

Most participants, 91.8% (669/729) reported ever anticipating at least one kind of anticipated sex work stigma. The most prevalent forms of stigma ever anticipated were gossip and verbal abuse from family, 83.1% (611/729); exclusion, 82.9% (604/729); and gossip and verbal abuse from friends and community, 82.7% (603/729) (Table 2).

Table 2.

Self-reported Anticipated Sex Work Stigma (Ever/Never) Among Sex Workers (n=729)

N(%) Ever* Never
Anticipated Stigma (overall) 669 (91.8%) 60 (8.2%)
Gossip and Verbal Abuse from Family 611 (83.1%) 118 (16.2%)
Gossip and Verbal Abuse from Healthcare Workers 436 (59.8%) 293 (40.2%)
Gossip and Verbal Abuse from Friends and Community 603 (82.7%) 126 (17.3%)
Physical Abuse 497 (68.2%) 232 (31.8%)
Exclusion 604 (82.9%) 125 (17.1%)
Open in a new tab
*

Ever includes response options: once, a few times, often, and not in the last 12 months, but have before.

Reliability

The final scale included 15 items, with excellent internal consistency (Cronbach’s α=0.93) and five subscales—gossip and verbal abuse from family (two items, α=0.81); gossip and verbal abuse from healthcare workers (two items, α=0.80); gossip and verbal abuse from friends and community (four items, α=0.89); physical abuse (three items, α=0.84); and exclusion (four items, α=0.81).

Construct Validity

Exploratory Factor Analysis

For the exploratory factor analysis, we first assessed the factorability of the initial 23 items; we report a KMO statistic of 0.91; thus suggesting a factor structure is likely to underlie the data. Bartlett’s test of sphericity for assessing the intercorrelatedness of the items was significant, indicating the existence of large correlations amongst the variables, (253) = 11036.866 (p =0.000). Factor analysis of the 23 items, using varimax (orthogonal) rotations, and retaining items with factor loadings of at 0.4 on their primary factor, with no cross loading of greater than 0.3 on any other factors, we found that eigen values indicated that 5 factors explained 100% of the variance cumulatively. We thus retained a 5-factor structure, encompassing 15 items, which was then tested further with confirmatory factor analysis. Cronbach’s alpha of the 15 items retained was 0.93, indicating excellent scale reliability.

Confirmatory Factor Analysis

Our final model had good fit indices as indicated in Figure 1 (chi-square=530.9; chi-square p value <0.001; RMSEA=0.088; SRMR=0.047; and CFI=0.92). Table 3 shows descriptions of the scale and subscales; individual survey items are also listed with standardized factor loadings for each construct.

Figure 1.

Figure 1.

Open in a new tab

CFA Anticipated Sex Work Stigma Scale

Table 3:

Final Anticipated Stigma Sex Work Stigma Scale with Factor Loadings and Descriptions of Items (N=729)

Items Question Standardized Factor Loadings
Have you ever been fearful of…
Subscale Gossip and Verbal Abuse from Family (Range of Possible Scores=0–8, alpha=0.81)
X1 Being gossiped about by family 0.79
X2 Being verbally insulted, harassed or threatened by family 0.87
Subscale Gossip and Verbal Abuse from Healthcare Workers (Range of Possible Scores=0–8, alpha=0.80)
X3 Being gossiped about by healthcare workers 0.89
X4 Being verbally insulted, harassed or threatened by healthcare workers 0.75
Subscale Gossip and Verbal Abuse from Friends and Community (Range of Possible Scores=0–16, alpha=0.89)
X5 Being gossiped about by friends 0.77
X6 Being gossiped about by neighbors and general community 0.82
X7 Being verbally insulted, harassed or threatened by friends 0.84
X8 Being verbally insulted, harassed or threatened by neighbors and general community 0.86
Subscale Physical Abuse (Range of Possible Scores=0–12, alpha=0.84)
X9 Being physically hurt (pushed, shoved, slapped, hit, kicked, choked, or otherwise physically hurt you) by family 0.74
X10 Being physically hurt (pushed, shoved, slapped, hit, kicked, choked, or otherwise physically hurt you) by friends 0.84
X11 Being physically hurt (pushed, shoved, slapped, hit, kicked, choked, or otherwise physically hurt you) by neighbors and general community 0.82
Subscale Exclusion (Range of Possible Scores=0–8, alpha=0.81)
X12 Being excluded from family gatherings 0.80
X13 Being excluded from community events, such as weddings, parties or funerals 0.79
X14 Being rejected by your friends 0.69
X15 Being forced to change your place of residence or being unable to rent accommodation 0.63
Open in a new tab

Scale items treated as continuous (0 = Never; 1 = Not in last 12 months but have before; 2= Once in the last 12 months; 3 = A few times in the last 12 months; and 4 = Often in the last 12 months).

Total scale alpha= 0.93

List of initial 23 items available on request.

Convergent and Discriminant Validity

For convergent validity, 15/15 items (100.0%) had a correlation coefficient with the score of their own dimension greater than 0.5. For discriminant validity, 14/15 items (93.3%) had a correlation coefficient with the score of their own dimension greater than those computed with other scores (Table 4). The AVE estimates for the constructs (also Table 4) - Gossip and Verbal Abuse from Family (0.70), Gossip and Verbal Abuse from Healthcare Workers (0.66), Gossip and Verbal Abuse from Friends and Community (0.68), Physical Abuse (0.65) and Exclusion (0.55) – demonstrated acceptable convergent validity of the scale, while the pairwise AVE-SC comparisons demonstrated acceptable discriminant validity, although ideally, all AVE values for all factors should be greater than the AVE-SC comparisons for the factors. (Table 5)

Table 4.

Convergent Validity Assessment

Gossip and Verbal Abuse from: Physical Abuse Exclusion
Family Healthcare workers Friends and Community
Gossip and Verbal Abuse from Family
Being gossiped about by family 0.686 0.424 0.665 0.456 0.515
Being verbally insulted, harassed or threatened by family 0.686 0.429 0.674 0.525 0.626
Gossip and Verbal Abuse from Healthcare Workers
Being gossiped about by healthcare workers 0.451 0.664 0.565 0.407 0.445
Being verbally insulted, harassed or threatened by healthcare workers 0.396 0.664 0.45 0.374 0.425
Gossip and Verbal Abuse from Friends and Community
Being gossiped about by friends 0.616 0.491 0.725 0.441 0.516
Being gossiped about by neighbors and general community 0.63 0.473 0.778 0.486 0.517
Being verbally insulted, harassed or threatened by friends 0.645 0.501 0.768 0.501 0.608
Being verbally insulted, harassed or threatened by neighbors and general community 0.646 0.473 0.784 0.532 0.593
Physical Abuse
Being physically hurt (pushed, shoved, slapped, hit, kicked, choked, or otherwise physically hurt you) by family 0.547 0.374 0.462 0.637 0.601
Being physically hurt (pushed, shoved, slapped, hit, kicked, choked, or otherwise physically hurt you) by friends 0.412 0.395 0.497 0.752 0.59
Being physically hurt (pushed, shoved, slapped, hit, kicked, choked, or otherwise physically hurt you) by neighbors and general community 0.436 0.349 0.511 0.719 0.601
Exclusion
Being excluded from family gatherings 0.608 0.378 0.508 0.56 0.691
Being excluded from community events, such as weddings, parties or funerals 0.489 0.383 0.5 0.58 0.703
Being rejected by your friends 0.451 0.42 0.581 0.521 0.605
Being forced to change your place of residence or being unable to rent accommodation 0.447 0.348 0.469 0.54 0.537
Open in a new tab
Table 5.

Discriminant Validity Assessment

Squared correlations (SC) among latent variables (factors) AVE
Gossip and Verbal Abuse from Physical Abuse Exclusion
Family Healthcare Workers Friends and Community
Gossip and Verbal Abuse from Family 1.000 0.696
Gossip and Verbal Abuse from Healthcare Workers 0.349 1.000 0.658
Gossip and Verbal Abuse from Friends and Community 0.726 0.451 1.000 0.679
Physical Abuse 0.384 0.257 0.411 1.000 0.650
Exclusion 0.587 0.323 0.526 0.620 1.000 0.548
Open in a new tab

AVE=Average Variance Extracted

Known-Group Validity

In the known-group analysis, mean scores (SD) for all anticipated sex work stigma measures were higher among MSW compared to FSW except for gossip and verbal abuse from healthcare workers. All stigma measures were higher among sex workers who reported severe or moderately severe depression compared to those who reported minimal, mild, or moderate depression. Higher stigma measures were reported among those who completed tertiary education compared to those who completed secondary, primary, or less. Finally, all stigma measures were higher among sex workers who reported healthcare avoidance or delay in the past year compared to those who did not (Table 6).

Table 6.

Known-group Analysis of Measures of Anticipated Sex Work Stigma by Sex, Depression, Education and Healthcare Utilization, Mean (SD)

Gossip and Verbal Abuse from: Physical Abuse Exclusionx
Family Healthcare workers Friends and Community
Range 0–8 0–8 0–16 0–12 0–16
Sex
 Female 4.8 (3.1) 3.08 (3.16) 8.2 (6.1) 4.5 (4.5) 7.5 (5.6)
 Male 5.7 (2.8) 2.70 (2.85) 10.7 (5.7) 6.0 (4.8) 8.5 (5.5)
 p-value <0.001 0.12 <0.001 <0.001 0.03
Depression (PHQ-9 Scores)
 Minimal (0–4) 4.2 (3.3) 2.3 (2.8) 7.7 (6.3) 3.5 (4.4) 5.7 (5.7)
 Mild (5–9) 5.3 (3.0) 2.9 (3.1) 9.4 (6.0) 5.4 (4.7) 8.3 (5.5)
 Moderate (10–14) 5.3 (3.0) 3.2 (3.1) 9.1 (5.9) 5.7 (4.4) 8.8 (5.1)
 Moderately Severe (15–19) 6.2 (2.2) 4.2 (3.1) 10.9 (5.3) 6.4 (4.5) 10.0 (4.9)
 Severe (20–27) 7.3 (1.0) 5.1 (3.4) 12.5 (3.8) 6.2 (4.4) 10.9 (5.7)
 p-value <0.001 <0.001 <0.001 <0.001 <0.001
Education
 Primary or less 5.0 (3.1) 3.2 (3.2) 8.7 (6.1) 4.6 (4.5) 7.5 (5.7)
 Secondary 5.0 (3.1) 2.6 (3.0) 8.8 (6.1) 5.1 (4.7) 7.9 (5.5)
 Tertiary 5.7 (2.4) 3.6 (2.9) 11.1 (5.3) 6.1 (4.5) 8.9 (5.2)
 p-value 0.16 0.01 <0.01 0.03 0.14
Healthcare Avoidance or Delayed Utilization in the Past Year
 Yes 5.3 (3.0) 3.1 (3.1) 9.6 (5.9) 5.3 (4.7) 8.3 (5.5)
 No 4.2 (3.3) 2.3 (2.9) 7.1 (6.2) 3.9 (4.5) 6.2 (5.8)
 p-value <0.001 <0.01 <0.001 <0.001 <0.001
Open in a new tab

Discussion

We measured anticipated sex work related stigma among a cross-sectional sample of male and female sex workers. Our analysis supported a 5-factor, 15-item anticipated sex work stigma scale measuring five anticipated sex work stigma domains: gossip and verbal abuse from family; gossip and verbal abuse from healthcare worker; gossip and verbal abuse from friends and community; physical abuse; and exclusion. The scale demonstrated good face, content, and construct validity. Reliability (internal consistency) was good for all subscales and the overall scale. The scale demonstrated good model fit statistics and good standardized factor loadings.

This study documented the high prevalence of different forms and sources of anticipated sex work-related stigma among female and male sex workers in Kenya. While other studies have found a high burden of stigma among sex workers, few studies have use validated measures to parse out the types, forms and sources of the stigma sex works face (S. Baral et al., 2014; Fitzgerald-Husek et al., 2017; Tun, De Mello, Pinho, Chinaglia, & Diaz, 2008). In fact, a recent review of studies measuring sex work stigma did not find any validated anticipated stigma scales (UNAIDS, 2018). Of note, a study among MSM who engage in transactional sex in the US measured anticipated stigma around pre-exposure prophylaxis (PrEP) use from primary and casual partners, specifically the fear their partner(s) would discover they engaged in transactional sex if they told disclosed their PreP use (Biello et al., 2017). Another recent study conducted among female sex workers in the Dominican Republic and Tanzania has developed a scale with 4 sub-domains: shame (internalized), dignity (resisted), silence (anticipated) and treatment (enacted) (Kerrigan et al., 2021). However, this anticipated sub-domain captures concerns around managing disclosure, with items such as “You have tried to make sure that no one knows that you do sex work” (Kerrigan et al., 2021). Further, a recent study did validate a measure of sex work related stigma among male and female sex workers in Togo and Burkina Faso that includes two anticipated stigma items related to health care seeking due to worry someone may learn that you sell sex/have sex with men (Grosso et al., 2019). As such, this is the first scale dedicated to measuring anticipated stigma that has been developed among both male and female sex workers in sub-Saharan Africa.

Anticipated stigma is a unique and insidious type of stigma that may have far reaching consequences on disclosure, risk perception and health care seeking and engagement. Borrowing from Earnshaw et al.’s HIV Stigma Framework (Earnshaw & Chaudoir, 2009; Earnshaw, Smith, Chaudoir, Amico, & Copenhaver, 2013), anticipated stigma from family, friends, the community and health care workers is theorized to directly undermine health care seeking and interpersonal outcomes; anticipated stigma from friends and family can disrupt social support and anticipated stigma from health-care workers can discourage clinic visit attendance and undermine trust in providers (B. Turan et al., 2017). In a recent effort to understand how HIV stigma impacts people living with HIV, identifying the source of anticipated stigma was particularly useful to understanding its impact on different health outcomes (B. Turan et al., 2017). A study of MSM and Transwomen in New York City found that anticipated HIV stigma was both a barrier to HIV testing and associated with lower HIV risk perception (Golub & Gamarel, 2013). The authors argue that as individuals strive to distance themselves from potentially stigmatized conditions, they underestimated their risk for HIV. A study among MSM who engage in transactional sex found that anticipated stigma from primary and casual partners was a barrier to PreP use, undermining HIV prevention efforts (Biello et al., 2017). Further research is warranted into the mechanisms through which sex work-related stigma – especially anticipated stigma – undermines a range of health care outcomes, from sexual and reproductive health, including STI and HIV prevention and treatment, to mental health and chronic health conditions.

Limitations

This study used a snowball sample of Kenyan sex workers, as is common among hidden or difficult populations, and the final sample did include a diverse group of sex workers from a variety of environments. However, the generalizability of the study results may be limited. The survey tool could benefit from further validation in other cultural contexts. Of note, the population of MSW was relatively small, and thus we did not have adequate power to assess construct validity by gender.

Conclusion

This analysis yielded a short, 15-item anticipated sex work stigma scale that measured anticipated gossip and verbal abuse from family, healthcare workers, friends and community, physical abuse and exclusion. As efforts to mitigate stigma faced by key populations such as sex workers in sub-Saharan Africa continue, valid and reliable measures of sex work stigma will be increasingly necessary. Our validated anticipated stigma tool can allow researchers and program managers to characterize the forms and sources of anticipated sex work-related stigma among sex workers. Further, such valid measurement tools will ultimately support rigorous evaluation of stigma-reduction efforts that address this insidious type of stigma as well as barriers to healthcare seeking or HIV and STI prevention and treatment efforts and achieving the UNAIDS 95–95-95 targets.

Acknowledgments

We would like to express our gratitude to the partner organizations, research assistants and survey participants. The study was funded by the U.S. Agency for International Development (USAID) and the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR) through funding to the Health Policy Project [agreement number AID-OAA-A-10-00067]. The content of this manuscript is the sole responsibility of the authors and does not necessarily reflect the views or policies of U.S. Agency for International Development or the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR), and does not imply endorsement by the U.S. Government.

Funding:

MAS was supported by the National Institute of Mental Health (T32MH096724).

Footnotes

Declarations

Disclosure: The authors declare that they have no conflict of interest.

References

  1. Alvarez-Uria G, Pakam R, Midde M, & Naik PK (2013). Entry, retention, and virological suppression in an HIV cohort study in India: description of the cascade of care and implications for reducing HIV-related mortality in low-and middle-income countries. Interdisciplinary perspectives on infectious diseases, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baral S, Ketende S, Green JL, Chen P-A, Grosso A, Sithole B, . . . Kennedy CE (2014). Reconceptualizing the HIV epidemiology and prevention needs of Female Sex Workers (FSW) in Swaziland. PLoS One, 9(12), e115465 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baral SD, Friedman MR, Geibel S, Rebe K, Bozhinov B, Diouf D, . . . Cáceres CF (2015). Male sex workers: practices, contexts, and vulnerabilities for HIV acquisition and transmission. The Lancet, 385(9964), 260–273 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Biello KB, Oldenburg CE, Mitty JA, Closson EF, Mayer KH, Safren SA, & Mimiaga MJ (2017). The “safe sex” conundrum: anticipated stigma from sexual partners as a barrier to PrEP use among substance using MSM engaging in transactional sex. AIDS and Behavior, 21(1), 300–306 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Corrigan PW, Larson JE, & Ruesch N. (2009). Self-stigma and the “why try” effect: impact on life goals and evidence-based practices. World psychiatry, 8(2), 75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Earnshaw VA, & Chaudoir SR (2009). From conceptualizing to measuring HIV stigma: a review of HIV stigma mechanism measures. AIDS and Behavior, 13(6), 1160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Earnshaw VA, Smith LR, Chaudoir SR, Amico KR, & Copenhaver MM (2013). HIV stigma mechanisms and well-being among PLWH: a test of the HIV stigma framework. AIDS and Behavior, 17(5), 1785–1795 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fitzgerald-Husek A, Van Wert MJ, Ewing WF, Grosso AL, Holland CE, Katterl R, . . . Baral SD (2017). Measuring stigma affecting sex workers (SW) and men who have sex with men (MSM): a systematic review. PLoS One, 12(11), e0188393 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Golub SA, & Gamarel KE (2013). The impact of anticipated HIV stigma on delays in HIV testing behaviors: findings from a community-based sample of men who have sex with men and transgender women in New York City. AIDS patient care and STDs, 27(11), 621–627 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Govindasamy D, Ford N, & Kranzer K. (2012). Risk factors, barriers and facilitators for linkage to antiretroviral therapy care: a systematic review. Aids, 26(16), 2059–2067 [DOI] [PubMed] [Google Scholar]
  11. Grosso AL, Ketende SC, Stahlman S, Ky-Zerbo O, Ouedraogo HG, Kouanda S, . . . Anato S. (2019). Development and reliability of metrics to characterize types and sources of stigma among men who have sex with men and female sex workers in Togo and Burkina Faso. BMC infectious diseases, 19(1), 208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hatzenbuehler ML, Phelan JC, & Link BG (2013). Stigma as a fundamental cause of population health inequalities. American journal of public health, 103(5), 813–821 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Holzemer WL, Uys LR, Chirwa ML, Greeff M, Makoae LN, Kohi TW, . . . Phetlhu RD (2007). Validation of the HIV/AIDS Stigma Instrument—PLWA (HASI-P). AIDS care, 19(8), 1002–1012 [DOI] [PubMed] [Google Scholar]
  14. Hu L. t., & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural equation modeling: a multidisciplinary journal, 6(1), 1–55 [Google Scholar]
  15. Kalichman SC, Simbayi LC, Cloete A, Mthembu PP, Mkhonta RN, & Ginindza T. (2009). Measuring AIDS stigmas in people living with HIV/AIDS: the Internalized AIDS-Related Stigma Scale. AIDS care, 21(1), 87–93 [DOI] [PubMed] [Google Scholar]
  16. Kalichman SC, Simbayi LC, Jooste S, Toefy Y, Cain D, Cherry C, & Kagee A. (2005). Development of a brief scale to measure AIDS-related stigma in South Africa. AIDS and Behavior, 9(2), 135–143 [DOI] [PubMed] [Google Scholar]
  17. Katz IT, Ryu AE, Onuegbu AG, Psaros C, Weiser SD, Bangsberg DR, & Tsai AC (2013). Impact of HIV-related stigma on treatment adherence: systematic review and meta-synthesis. Journal of the International AIDS Society, 16, 18640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kerrigan D, Karver TS, Barrington C, Davis W, Donastorg Y, Perez M, . . . Shembilu C. (2021). Development of the experiences of sex work stigma scale using item response theory: implications for research on the social determinants of HIV. AIDS and Behavior, 1–14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kline RB (2015). Principles and practice of structural equation modeling: Guilford publications. [Google Scholar]
  20. Kroenke K, Spitzer R, & Williams J. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606–613. doi: 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lancaster KE, Cernigliaro D, Zulliger R, & Fleming PF (2016). HIV care and treatment experiences among female sex workers living with HIV in sub-Saharan Africa: A systematic review. African Journal of AIDS Research, 15(4), 377–386 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lebet RM, Asaro LA, Zuppa AF, & Curley MA (2018). Face and content validity of variables associated with the difficult-to-sedate child in the paediatric intensive care unit: A survey of paediatric critical care clinicians. Australian Critical Care, 31(3), 167–173 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Link BG, & Phelan JC (2001). Conceptualizing stigma. Annual review of Sociology, 27(1), 363–385 [Google Scholar]
  24. Liu S-H, Srikrishnan A, Zelaya CE, Solomon S, Celentano DD, & Sherman SG (2011). Measuring perceived stigma in female sex workers in Chennai, India. AIDS care, 23(5), 619–627 [DOI] [PubMed] [Google Scholar]
  25. Logie CH, James L, Tharao W, & Loutfy MR (2011). HIV, gender, race, sexual orientation, and sex work: a qualitative study of intersectional stigma experienced by HIV-positive women in Ontario, Canada. PLoS medicine, 8(11), e1001124 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mahajan AP, Sayles JN, Patel VA, Remien RH, Ortiz D, Szekeres G, & Coates TJ (2008). Stigma in the HIV/AIDS epidemic: a review of the literature and recommendations for the way forward. AIDS (London, England), 22(Suppl 2), S67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Moore K, Stuewig J, & Tangney J. (2013). Jail inmates’ perceived and anticipated stigma: Implications for post-release functioning. Self and Identity, 12(5), 527–547 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mosier CI (1947). A critical examination of the concepts of face validity. Educational and Psychological Measurement, 7(2), 191–205 [DOI] [PubMed] [Google Scholar]
  29. Nevo B. (1985). Face validity revisited. Journal of Educational Measurement, 22(4), 287–293 [Google Scholar]
  30. Nyblade L, MacQuarrie K, Phillip F, Kwesigabo G, Mbwambo J, & Ndega J. (2005). Working report measuring HIV stigma: results of a field test in Tanzania. Washington, DC: United States Agency for International Development [Google Scholar]
  31. Nyblade L, Mbote DK, Barker C, Morla J, Mwai D, Oneko T, . . . Musyoki H. (2015). Impact of stigma on utilization of health services among sex Workers in Kenya: Washington, DC: Futures Group, Health Policy Project. [Google Scholar]
  32. Nyblade L, Mingkwan P, & Stockton MA (2021). Stigma reduction: an essential ingredient to ending AIDS by 2030. The Lancet HIV, 8(2), e106–e113 [DOI] [PubMed] [Google Scholar]
  33. Nyblade L, Reddy A, Mbote D, Kraemer J, Stockton M, Kemunto C, . . . Dutta A. (2017). The relationship between health worker stigma and uptake of HIV counseling and testing and utilization of non-HIV health services: the experience of male and female sex workers in Kenya. AIDS care, 29(11), 1364–1372 [DOI] [PubMed] [Google Scholar]
  34. Salkind NJ (2010). Content Validity Encyclopedia of research design (Vol. 1): Sage. [Google Scholar]
  35. Steward WT, Bharat S, Ramakrishna J, Heylen E, & Ekstrand ML (2013). Stigma is associated with delays in seeking care among HIV-infected people in India. Journal of the International Association of Providers of AIDS Care (JIAPAC), 12(2), 103–109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Stockton MA, Pence BW, Mbote D, Oga EA, Kraemer J, Kimani J, . . . Nyblade L. (2020). Associations among experienced and internalized stigma, social support, and depression among male and female sex workers in Kenya. International journal of public health, 65(6), 791–799 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Tsang E. Y. h., Qiao S, Wilkinson JS, Fung A. L. c., Lipeleke F, & Li X. (2019). Multilayered stigma and vulnerabilities for HIV infection and transmission: a qualitative study on male sex workers in Zimbabwe. American journal of men’s health, 13(1), 1557988318823883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tun W, De Mello M, Pinho A, Chinaglia M, & Diaz J. (2008). Sexual risk behaviours and HIV seroprevalence among male sex workers who have sex with men and non-sex workers in Campinas, Brazil. Sexually transmitted infections, 84(6), 455–457 [DOI] [PubMed] [Google Scholar]
  39. Turan B, Budhwani H, Fazeli PL, Browning WR, Raper JL, Mugavero MJ, & Turan JM (2017). How does stigma affect people living with HIV? The mediating roles of internalized and anticipated HIV stigma in the effects of perceived community stigma on health and psychosocial outcomes. AIDS and Behavior, 21(1), 283–291 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Turan JM, Elafros MA, Logie CH, Banik S, Turan B, Crockett KB, . . . Murray SM (2019). Challenges and opportunities in examining and addressing intersectional stigma and health. BMC medicine, 17(1), 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. UNAIDS. (2018). Global AIDS Update: Miles to go—closing gaps, breaking barriers, righting injustices. . Geneva, Switzerland: http://www.unaids.org/sites/default/files/media_asset/miles-to-go_en.pdf. [Google Scholar]
  42. UNAIDS. (2020). Fact Sheet. Geneva, Switzerland: https://www.unaids.org/sites/default/files/media_asset/UNAIDS_FactSheet_en.pdf. [Google Scholar]
  43. Visser MJ, Kershaw T, Makin JD, & Forsyth BW (2008). Development of parallel scales to measure HIV-related stigma. AIDS and Behavior, 12(5), 759–771 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Weiss MG, Ramakrishna J, & Somma D. (2006). Health-related stigma: rethinking concepts and interventions. Psychology, health & medicine, 11(3), 277–287 [DOI] [PubMed] [Google Scholar]

RESOURCES