Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Oct 17.
Published in final edited form as: Patient Educ Couns. 2021 Oct 8;105(7):1703–1713. doi: 10.1016/j.pec.2021.10.008

A systematic search and review of the discrimination in health care measure and its adaptations

Sheryl Thorburn 1, Olivia J Lindly 2,*
PMCID: PMC9576001  NIHMSID: NIHMS1840471  PMID: 34688522

Abstract

Background:

Discrimination occurs in health care settings contributing to health inequities. Yet guidance on how best to measure discrimination in health care is still limited.

Objectives:

We sought to (1) identify and describe the characteristics of published studies that used the Discrimination in Health Care Measure, a scale first published in 2001; (2) review how the measure has been used or adapted and summarize the measure’s published psychometric properties and its variations; and (3) summarize associations between the measure and health-related variables.

Methods:

We performed a systematic search and review of the measure by searching PsychINFO, PubMed, Sociological Abstracts, and Web of Science from January 1, 2001 through January 31, 2017. We screened 260 unique articles, identified 22 eligible articles, and completed a narrative synthesis.

Results:

Most studies measured race or ethnicity-based discrimination. All studies made minor revisions to the measure, and most reported high reliabilities. Discrimination in health care, using this measure, was associated with adverse health outcomes.

Discussion and Practice Implications:

Study results indicate that the measure is easy to use and adapt. Researchers should consider using the Discrimination in Health Care Measure when designing studies that will examine individuals’ discriminatory experiences when receiving health care.

Keywords: discrimination, health care, systematic review, systematic search, measurement

1. Introduction

Research on the health effects of discrimination has evolved during the past 20 years. In the early years, studies in the U.S. focused primarily on racial discrimination, did not assess the specific context in which the discrimination occurred, examined acute and/or chronic experiences of discrimination, and generally relied on biopsychosocial models (Clark et al., 1999; King & Williams, 1995; Williams, 1996; Williams et al., 1997) or ecosocial theory (Krieger, 1994, 2001). Over time, the literature expanded to include diverse study populations, a broadening of the types of discrimination examined, and a better understanding of the contexts in which discrimination occurs. Measures of discrimination have been compared and evaluated in systematic reviews (Bastos et al., 2010; Dolezsar et al., 2014; Paradies, 2006; Paradies et al., 2015; Pascoe & Richman, 2009; Priest et al., 2013). Still, gaps in our understanding and measurement of discrimination and its health effects remain.

Health care settings are a unique context in which discrimination occurs. The significance of this context is emphasized by the 2030 United Nations’ Agenda for Sustainable Development in which ending discrimination in health care settings is a central tenant (World Health Organization, 2017). Individuals seeking care may experience discrimination when interacting with health care providers, other health professionals, staff, or other patients. Negative experiences in these settings could impact the care received, patient-provider relationship quality, patient satisfaction and preferences, health care utilization, acceptance of provider recommendations, and health outcomes. For researchers focused on these outcomes, discrimination in health care settings may be a proximate cause of negative outcomes than more general discrimination. Additionally, context-specific discrimination measures can help identify targets for intervention. Yet, research on discrimination in health care is still relatively limited. In the early 2000s, few published studies had as their primary purpose improving understanding of discrimination in health care or the association of such experiences with health behaviors, health care utilization, or health outcomes (LaVeist et al., 2000; Lillie-Blanton et al., 2000). Furthermore, measures of personal discrimination in health care frequently consisted of single items and assessed discrimination in medical settings or by providers in general (Lillie-Blanton et al., 2000). Items about health care discrimination were sometimes included in measures that examined discrimination occurring in multiple contexts (Klonoff & Landrine, 1999; Krieger & Sidney, 1996; Ren et al., 1999). In addition, some measures asked about discrimination toward a group in general, rather than about personal experiences of discrimination (LaVeist et al., 2000). Although these measures served a purpose, understanding experiences with discrimination in multiple and specific health care contexts required further measurement development.

In 2001, Bird and Bogart published a study which measured race-based and socioeconomic (SES)-based discrimination in health care (Appendix). To measure race-based discrimination, study participants were asked “When getting health care, have you ever had any of the following things happen to you because of your race or color?” To measure SES-based discrimination, participants were asked “When getting health care, have you ever had any of the following things happen to you because of your socioeconomic status or social class?” For each question, the following binary (yes/no) items were listed: 1) been treated with less courtesy than other people, 2) been treated with less respect than other people, 3) received poorer service than others, 4) had a doctor or nurse act as if he or she thinks you are not smart, 5) had a doctor or nurse act as if he or she is afraid of you, 6) had a doctor or nurse act as if he or she is better than you, and 7) felt like a doctor or nurse was not listening to what you were saying. The seventh item was created specifically for the measure. The first six items were adapted from Williams’ nine-item everyday discrimination scale, which assesses the frequency of “chronic, routine, and relatively minor experiences of unfair treatment…in the day-to-day lives of respondents” (Williams et al., 1997). Subsequently, the Discrimination in Health Care Measure was modified to measure discrimination in Human Immunodeficiency Virus (HIV) treatment (Bird et al., 2004). Question wording was revised to say, “When receiving treatment for HIV, how often do the following things happen to you because of your race or color?” and “…because of your socioeconomic status, position, or social class?” The same seven items were used with minor wording changes; the response options were always, most of the time, sometimes, rarely, and never. Another modification of the measure was used in two studies to assess race-based discrimination when receiving family planning or contraceptive services (Bird & Bogart, 2003; Thorburn & Bogart, 2005). In both studies, the first four items were taken from the earlier measure and combined with four or five new items that captured behaviors that were potentially more likely to occur in the context of family planning and contraceptive services (Bird & Bogart, 2003; Thorburn & Bogart, 2005). Except for the original 2001 article, these studies assessed the internal consistency of the scale and found good to excellent reliability (Bird & Bogart, 2001). Other researchers have adapted the Discrimination in Health Care Measure in studies with diverse populations. Furthermore, the multi-item Discrimination in Health Care Measure was significantly associated with patients’ reported problems with their health care, while a single-item measure of perceived (personal) discrimination in health care was not (Hausmann et al., 2010).

No previous studies have performed a comprehensive evaluation of the Discrimination in Health Care Measure. Given its frequent use and strengths, our overarching goals were to provide summary information about the measure for researchers studying and/or intervening to reduce discrimination in the health care context and increase the measure’s visibility. The specific purpose of this systematic search and review was threefold: 1) to identify and describe the characteristics of published studies that have used the original or a modified version of the Discrimination in Health Care Measure, 2) to review how the measure has been used or adapted and summarize the measure’s published psychometric properties and its variations, and 3) to summarize the associations between the measure and health-related variables.

2. Methods

We searched the PsychINFO (American Psychological Association, n.d.), PubMed (US National Library of Medicine, n.d.), Scopus (Elsevier, n.d.), Sociological Abstracts (ProQuest, n.d.), and Web of Science (Thomson, n.d.) databases for articles published from January 1, 2001 to January 31, 2017 citing any of the four original articles using the Discrimination in Health Care Measure (Bird & Bogart, 2001, 2003; Bird et al., 2004; Thorburn & Bogart, 2005). The beginning of the timeframe was chosen to coincide with the first publication of the measure (Bird & Bogart, 2001). Each database was searched using the full title of each of the primary articles. English language was the only other inclusion criterion; however, this limit was applied after the initial search and showed that no articles were excluded as a result. For the “cited reference search” conducted in the Web of Science database, we additionally specified the document type as article. Publications shown as citing any of the four primary articles in any of the five databases searched were identified. Citations for these records were then imported into Covidence, a Web-based platform used for systematic reviews.

Across the databases searched, we identified 472 records. Of these, 212 were duplicates that were removed in Covidence, leaving 260 unique records. Titles and abstracts for these records were screened to determine eligibility using the following inclusion criteria: 1) the article had to be original research and 2) the study was not a qualitative study due to the quantitative nature of the Discrimination in Health Care Measure. Eighty-four records were excluded including 81 articles that did not meet these criteria and the three original articles published after 2001. The full text articles for the remaining 176 records were then reviewed to determine if the research reported used the Discrimination in Health Care Measure based on the description and attribution of the measure in each article. Once ineligible articles were removed, 22 articles remained (Figure 1).

Figure 1.

Figure 1.

Open in a new tab

PRISMA Search Strategy

Each author independently reviewed these articles. Then, each author extracted and synthesized information on each study’s characteristics, the use and properties of the Discrimination in Health Care Measure, and associations of the measure with health-related variables. To describe the findings with the least amount of bias due to confounding, we present only results from multivariable analyses, except where noted. Discrepancies between the authors were resolved by means of discussion and consensus. The authors contacted the corresponding and/or first authors of all articles for which certain information (e.g., how the discrimination measure was adapted) was missing to request this information.

3. Results

3.1. Key Characteristics of Included Studies that Used or Adapted the Discrimination in Health Care Measure

In relation to the first study objective, Table 1 displays key characteristics of the 22 studies included in the review. Of note, many studies were published after 2009, were cross-sectional, and used non-probability sampling procedures. In general, sample sizes were small. The majority of study populations included African American/Black participants (5 studies exclusively included this racial group). The study populations were frequently defined by their health status.

Table 1.

Key Characteristics of 22 Empirical Studies that Used the Discrimination in Health Care Measure

Study characteristic n % of total studiesa
Publication year
  2001-2004 1 5
  2005-2009 1 5
  2010-2013 10 45
  2014-2017 10 45
Study design
  Cross-sectional 19 86
  Prospective cohort 2 9
  Randomized controlled trial 1 5
Sampling procedure
  Nonprobability 21 95
  Probability 1 5
Sample size
  n < 100 5 23
  100 < n < 300 11 50
  300 < n < 900 6 27
Region of the U.S.
  Midwest 6 27
  Northeast 7 32
  South 1 5
  West 2 9
  More than one region 6 22
Data collection mode
  Self-administered survey 9 41
  In-person interview 4 18
  Computer assisted interview 3 14
  Telephone interview 4 18
  Mixed-mode (in-person interview or self-administered survey) 2 9
Study population characteristics
  Racial/ethnic group(s)b
 African American/Black 17 77
 American Indian 2 9
 Arab American 2 9
 Asian American 2 9
 Latino/Hispanic 2 9
 White 11 50
  Health status
 Breast cancer 5 23
 Diabetes 4 18
 Hypertension 3 14
 Kidney disease 1 5
 Osteoarthritis 1 5
 Spinal cord injury 2 9
 Not specified (for sampling) 6 27
  Sex
 Female only 12 55
 Male only 0 0
 Male & Female 10 45
Basis of discrimination in health care c
  Race or ethnicityd 19 86
  Socioeconomic status or class 1 5
  Religion 2 9
  Sexual orientation 1 5
Open in a new tab
a

Percentages may not sum to 100 due to rounding.

b

Studies may have included individuals of one or more racial/ethnic groups. If studies included more than one racial/ethnic group, it is indicated here (e.g., a study that included both Black and White individuals is indicated twice; once in the African American/Black row and once in the White row). One study (Jabson, Donatelle, & Bowen, 2011) reported that their sample was “largely characterized as being….of non-Hispanic Caucasian ethnicity,” which was coded as White.

c

One study separately examined discrimination in health care based on race or color and based on socioeconomic status or class.

d

The Discrimination in Health Care Measure may have asked about experiences of discrimination related to race, ancestry, national origin, or color generally or related to a specific race (e.g., Native/American Indian).

3.2. Discrimination in Health Care Measure Use, Adaptations, and Psychometric Properties

In relation to the second study objective, 19 studies measured discrimination in health care based on race, ancestry, national origin, color, or ethnicity (hereafter, “race or ethnicity”). One study measured both race-based and SES-based discrimination in health care (Hausmann et al., 2011). Discrimination based on religion (the study population was Muslim) was measured in two articles (Padela et al., 2014; Vu et al., 2016). Another study measured discrimination in health care based on sexual orientation (Jabson et al., 2011; Table 1). The vast majority of measures (Table 2 and Appendix) asked about discrimination in the general context of receiving health care rather than about specific health care contexts. One study measured discrimination in a specific health care setting (Greer et al., 2014). Another study measured discrimination during breast cancer treatments and meeting with doctors and nurses (Jabson et al., 2011).

Table 2.

Detailed Information on Use of the Discrimination in Health Care Measure

Study Discrimination Basis Health Care Setting Number of Items & Response Scale Construction Psychometric Properties Statistically Significant Results from Multivariable Analyses Non-Significant Results from Multivariable Analyses
Original Studies
(Bird & Bogart 2001) Race- or color-based; or SES- or social class-based discrimination. General health care settings. 7 race-based items & 7 SES-based items rated on a binary scale (Yes/No). For each item set (i.e., race-based or SES-based), scores were summed then a measure of whether participants had ‘none’ versus ‘any’ of the experiences was created. Not reported. None reported. None reported.
(Bird & Bogart, 2003) Race- or color-based discrimination. Health care settings where family planning or birth control services accessed. 8 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always). Responses were recoded as binary (i.e., ‘never’ versus ‘ever’) for each item, then summed, with higher scores indicating greater discrimination. Item responses were recoded as binary (‘never’ versus ‘ever’)

Kuder-Richardson 20 = 0.88		 None reported. None reported.
(Bird et al., 2004) Race- or color-based; or SES, position, or social class-based discrimination. Health care settings where HIV treatment accessed. 7 race-based items & 7 SES-based items rated on a 5-point scale (Always, Most of the Time, Sometimes, Rarely, Never). Responses were recoded as binary (i.e., ‘never’ versus ‘ever’) for each item, then summed for all 14 items and for each item set. In addition, for each item set, mean score was computed, with higher scores indicating greater discrimination. Cronbach’s alpha for race-based discrimination items = 0.92

Cronbach’s alpha for SES-based discrimination items = 0.95		 None reported. None reported.
(Thorburn & Bogart, 2005) Race- or color-based discrimination. Health care settings where family planning or birth control services accessed. 9 items rated on a 5-point scale (Always, Most of the Time, Sometimes, Rarely, Never). Two subscales were created based on the exploratory factor analysis results. Subscale scores were recoded as “no discrimination” (i.e., scores of 0 across items) and “some discrimination” (i.e., scores of 1 or greater). Item responses were recoded as binary (i.e., ‘never’ versus ‘ever’)

Exploratory factor analysis results supported a two-factor solution. The first factor consisted of the first four items and captured more general experiences of health care discrimination (Kuder-Richardson 20 = 0.89).

The second factor consisted of four items and reflected stereotypes about African Americans (Kuder-Richardson 20 = 0.80).		 Stronger black identity, younger age, and lower income were associated with reports of race-based discrimination when accessing family planning or birth control services. Hispanic ethnicity, education level, religiosity, if the participant was currently working, number of people income supports, marital status, if the participant was currently living with a partner, number of live births, if the participant ever had an STD, if the participant had ever been testing for HIV, if the participant was currently using birth control, if the participant had vaginal sexual intercourse in the past 3 months, or number of vaginal sexual partners in the past 3 months were not significantly associated with race-based discrimination when accessing family planning or birth control services.
Found Articles
(Bogart, 2001) Race- or color-based discrimination. General health care settings. 7 items rated on a binary scale (Yes/No). The number of experiences (items) endorsed was summed. Not reported None reported. None reported.
(Burgess et al., 2014) Race- or color-based discrimination. General healthcare settings. 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always). Mean score across items was computed, with higher scores indicating greater discrimination. Cronbach’s alpha = 0.89 None reported. There were no statistically significant differences between the self-affirmation intervention and control groups on discrimination in health care. Because of randomization, only bivariate analyses were performed.
(Gonzales et al., 2013) Discrimination based on being Native American/American Indian.a General health care settings. 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always). A binary measure was created to categorize responses as ‘never’ (to all items) versus ‘any’ (rarely or more to at least one item). Mean score across items was also computed, with higher scores indicating greater discrimination. Kaiser criterion (for a one-factor solution) = 0.74

Standardized item factor loadings ranged from 0.77 to 0.90

Cronbach’s alpha = 0.94		 Discrimination in health care was associated with significantly higher adjusted odds of not being current for clinical breast examination and PAP test. Higher mean levels of discrimination in health care were also associated with a higher number of suboptimal health care seeking behaviors (i.e., put off or postponed health care, hesitant to get health care, did not come back for a follow-up appointment, did not follow the treatment plan or get a needed test, avoided the respective health care provider involved with the discrimination, and no longer used the respective health care facility where discrimination occurred). Discrimination in health care was not significantly associated with not being current for mammography screening.
(Gonzales et al., 2014) Discrimination based on being Native American/American Indian. General health care settings. 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always). A binary measure was created to categorize responses as ‘never’ (to all items) versus ‘any’ (rarely or more to at least one item). Cronbach’s alpha = 0.94a Discrimination in health care was associated with lower adjusted odds of dental exam; checks for blood pressure, creatinine, and total cholesterol; and pneumococcal vaccination. Any discrimination in health care was also associated with higher adjusted odds of having hemoglobin A1C values above target levels for diabetes control. Discrimination in health care was not significantly associated with blood pressure (>130/80 mm Hg) or total cholesterol (>200 mg/dl).
(Greer et. al, 2014) Discrimination based on being Blacka Specific health care settinga 7 items rated on a 5-point scale (Always, Most of the Time, Sometimes, Rarely, Never)a Items were reverse scored then summed, with higher scores indicating greater discrimination. Cronbach’s alpha = 0.93 Discrimination in health care was significantly and positively correlated with health care mistrust (e.g., “people in my ethnic group should be suspicious of modern medicine”). Discrimination in health care was also significantly associated with decreased adherence to hypertension treatment (e.g., adherence to a low sodium diet). There was additionally a significant interaction between discrimination in health care and systemic racism on treatment adherence, such that treatment adherence was lowest for individuals with high discrimination in health care and low or average (versus high) systemic racism. None reported.
(Hausmann et al., 2011) Race- or color-based, and SES- or class-based discrimination General health care settings 7 race-based items & 7 SES-based items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always)a Responses were recoded as binary (i.e., ‘never’ versus ‘ever’) for each item, then summed for each item set, with higher scores indicating greater discrimination. Cronbach’s alpha for race-based discrimination items = 0.93

Cronbach’s alpha for SES-based discrimination items = 0.90		 Among African American participants, high levels of race-based discrimination in health care were significantly associated with less positive nonverbal affect tone on both the part of participants and their providers (during their health care interaction), as well as with low participant reports of provider warmth/ respectfulness and ease of communication. Among Whites, race-based discrimination in health care was negatively associated with verbal psychosocial exchange and with participant ratings of visit informativeness and ease of communication. Results were similar for SES-based discrimination. Discrimination in health care based on race or color was not significantly associated with verbal measures of biomedical exchange, rapport building, or patient activation for African Americans and Whites. Among African Americans only, race or color based discrimination in health care was also not significantly associated with psychosocial exchange with providers. Informativeness of visit also was not significantly associated with race or color based or SES-based discrimination in health care for African Americans only. For Whites, provider warmth or respect were not significantly associated with race or color-based discrimination in health care. For both African Americans and Whites, SES-based discrimination in health care was not significantly associated with biomedical exchange, psychosocial exchange, rapport building, patient activation, positive patient affect, or positive provider affect. For Whites, discrimination in health care based on SES was not significantly associated with provider warmth or ease of communicating.
(Hausmann et al., 2010) Race- or color-based discrimination General health care settings 7 items on a 4-point scale (Never, Once, 2 or 3 Times, 4 Times or More)a Responses were recoded as binary (i.e., ‘never’ versus ‘ever’) for each item, then summed. If participants reported ‘never’ to all measures, they were categorized as experiencing ‘none’ versus ‘any’ discrimination. Cronbach’s alpha = 0.94 Discrimination in health care was associated with higher adjusted odds of experiencing problems with diabetes care. Discrimination in health care was not associated with receiving all 5 recommended screenings for diabetes complications in the past 2 years.
(Jabson et al., 2011) Sexual orientation-based discrimination Health care settings where breast cancer treatments received 7 items rated on a 5-point scale (Response options were not reported in the article.) Mean score across items were computed. Cronbach’s alpha = 0.75 Discrimination in health care was significantly associated with quality of life. None reported.
(Kressin et al., 2010) Race-, color-, or ethnicity-based discriminationa General health care settings 7 items rated on a binary scale (Yes/No) The number of experiences reported was summed, with higher scores indicating more discrimination. Cronbach’s alpha = 0.90 None reported. Discrimination in health care was not included in multivariable analysis of blood pressure control due to nonsignificant association found in bivariate analysis. None reported.
Discrimination in health care was not included in multivariable analysis of blood pressure control due to nonsignificant association found in bivariate analysis.
(Manze et al., 2010) Race-, color-, or ethnicity-based discriminationa General health care settings 7 items rated on a binary scale (Yes/No) A binary measure was created to categorize anyone who answered “yes” to any item as having experienced discrimination. Cronbach’s alpha = 0.90a None reported. Treatment intensification for hypertension was not significantly associated with discrimination in health care.
(Myaskovsk et al., 2011) Race- or ethnicity- discriminationa General health care settings 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always) a Mean score across items was computed. Cronbach’s alpha = 0.89 Discrimination in health care was associated with higher odds of occupational functioning (i.e., a person’s ability to participate in various activities such as employment or school). Discrimination in health care was not significantly associated with physical independence, mobility, social integration, life satisfaction, general perceived health status, or current perceived health status.
(Myaskovskyet al., 2012) Race- or ethnicity- discriminationa General health care settings 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always) a A binary measure was created to categorize responses as ‘any’ versus ‘none’ of the seven experiences. Cronbach’s alpha = 0.90 Participants who reported any discrimination in health care took significantly longer to be accepted for a kidney transplant than those who did not report any discrimination in health care. None reported.
(Myaskovsky et al., 2017) Race- or ethnicity- discriminationa General health care settings 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always)a A binary measure was created to categorize responses as ‘ever’ versus ‘never.’ Cronbach’s alpha = 0.92 Overall, multivariable analysis results showed previous experience of discrimination was significantly associated with lower service satisfaction. None reported.
Discrimination in health care was not included in multivariable analyses of physical component summary of health, the mental component summary of health, and general health status due to nonsignificant associations found in bivariate analyses.
(Padela et al., 2014) Religion-based discrimination General health care settings 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always) a Each response category was assigned a numerical value between 0 and 10. An average score was then computed if the respondent answered more than 50% of the items. Cronbach’s alpha = 0.93 None reported. Discrimination in health care was not significantly associated with cervical cancer screening status.
(Peek et al., 2011) Discrimination based on race, ancestry, or national origin General health care settings 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always) The factor structure of measure items was explored by treating the measure items as continuous variables, and descriptive statistics were computed by taking the mean score across items. Exploratory factor analysis results supported a single factor solution (eigenvalue = 4.36, accounting for 62% of the variance).

All items had standardized factor loadings > 0.49.

Cronbach’s alpha = 0.89 (original sample) & 0.85 (retest sample)

Test-retest reliability = 0.58, p < .001

Correlations with other variables indicated measure’s convergent validity and discriminant validity.		 None reported. None reported.
(Sheppard et al., 2017) Race- or ethnicity-based discrimination a General health care settings 7 items rated on a binary scale (Yes/No)a A binary measure was created to categorize responses as ‘any’ versus ‘none’ of the experiences. Cronbach’s alpha for entire sample = 0.87, for Black women = 0.86, and for White women = 0.81a None reported. Discrimination in health care was not significantly associated with chemotherapy use.
(Sheppard et al., 2011) Race- or ethnicity-based discriminationa General health care settings a 7 items rated on a binary scale (Yes/No)a Mean score across items was computed. Cronbach’s alpha = 0.81 None reported. None reported.
(Sheppard, O’Neill, et al., 2015) Race- or ethnicity-based discrimination a General health care settings 7 items rated on a binary scale (Yes/No)a A binary measure was created to categorize responses as ‘any’ versus ‘none’ of the experiences. Cronbach’s alpha for entire sample = 0.87, for Black women = 0.86, and for White women = 0.81a None reported. Discrimination in health care was not included in multivariable analyses of RS testing and chemotherapy use due to nonsignificant associations found in bivariate analysis. None reported.
Discrimination in health care was not included in multivariable analyses of RS testing and chemotherapy use of blood pressure control due to nonsignificant association found in bivariate analysis.
(Sheppard, Oppong, et al., 2015) Race- or ethnicity-based discrimination a General health care settings 7 items rated on a binary scale (Yes/No)a A binary measure was created to categorize responses as ‘any’ versus ‘none’ of the experiences. Cronbach’s alpha for entire sample = 0.87, for Black women = 0.86, and for White women = 0.81a None reported. Discrimination in health care was not included in multivariable analyses of time to surgery due to nonsignificant association found in bivariate analysis. None reported.
Discrimination in health care was not included in multivariable analyses of time to surgery due to nonsignificant association found in bivariate analysis.
(Sheppard et al., 2008) Discrimination based on Latino/Hispanic ethnicity General health care settings 6 items rated on a binary scale (Yes/No) A binary measure was created to categorize responses as ‘any’ versus ‘none’ of the experiences. Cronbach’s alpha = 0.60 Participants who had not experienced discrimination in health were more likely to report satisfaction with health care relationships. None reported.
Discrimination in health care was not included in multivariable analyses of recent mammography due to nonsignificant association found in bivariate analysis.
(Sheppard et al., 2014) Ethnicity-based discrimination General health care settings 6 items rated on a binary scale (Yes/No) A binary measure was created to categorize responses as ‘any’ versus ‘none’ of the experiences. Not reported Participants whose home language was Spanish and English (versus Spanish only) had higher adjusted odds of discrimination in health care. Additionally, higher health care satisfaction and better communication with providers were each associated with lower adjusted odds of discrimination in health care. Discrimination in health care was not significantly associated with education level, health insurance status, provider ethnicity (i.e., Latino or non-Latino), or trust in provider.
(Vu et al., 2016) Religion-based discrimination General health care settings 7 items rated on a 5-point scale (Never, Rarely, Sometimes, Most of the Time, Always) a Each response category was assigned a numerical value between 0 and 10. An average score was then computed if the respondent answered more than 50% of the items. Cronbach’s alpha = 0.93 None reported. Discrimination in health care was not significantly associated with delayed care seeking due to a perceived lack of female clinicians.
Open in a new tab
a

Information was obtained directly from authors.

Only five of the articles presented the complete wording of the question stem and all items in either the text or tables, although many implied that the measure was used with few or no modifications. We obtained the wording of the questions and the items directly from authors for an additional 14 articles, as well as clarification about the wording for another article (see Appendix). Most studies used the original question stem and seven items with minor changes (Bird & Bogart, 2001; Bird et al., 2004). Two studies dropped the following measure item: “Had a doctor or nurse act as if he or she was afraid of you” (Sheppard et al., 2008, 2014). One study by Jabson and colleagues (2011) used only the first four items of the original measure (Bird et. al, 2004), as well as an item from two of the other original studies about assumptions regarding multiple sexual partners (Bird et al., 2003).

Nine studies allowed for binary responses (yes/no) while others (n = 13) used 4 or 5-point response options. Half of the studies combined responses to individual items to create a binary measure of never versus ever experiencing discrimination (Table 2). Other studies created a sum or mean score. The measure’s psychometric properties were not reported in seven articles; for five of those, we obtained this information from the authors. For the studies where we could report this information, the majority indicated that the measure had strong internal consistency, with Cronbach’s alphas of .89 or higher. The lowest reliabilities were reported for measures of discrimination based on Latino/Hispanic ethnicity (0.60; Sheppard et al., 2008) and sexual orientation (0.75; Jabson et al., 2011). One study reported test-retest reliability of .58 (p < .001; Peek et al., 2011). Only one study reported results of analyses to assess construct validity (Peek et al., 2011).

3.3. Associations between the Discrimination in Health Care Measure and Health-Related Variables

As Table 2 shows, 15 articles reported results from multivariable analyses examining the associations between discrimination in health care and other variables. Statistically significant (p < .05) multivariable results demonstrated that household language (e.g., Spanish and English versus Spanish only) was associated with ethnicity-based discrimination in health care (Sheppard et al., 2014). Race-based discrimination in health care was positively associated with health care mistrust (Greer et al., 2014).

Multivariable analyses results additionally suggest significant associations of discrimination based on race or ethnicity with the following health-related factors (Table 2): poor patient-provider communication (Hausmann et al., 2011; Sheppard et al., 2014), problems accessing needed health care (Gonzales et al., 2013; Hausmann et al., 2010), lack of preventive care receipt (Gonzales et al., 2013, 2014), greater length of time to acceptance for kidney transplant (Myaskovsky et al., 2012), low treatment adherence (Greer et al., 2014), higher hemoglobin A1C levels (Gonzales et al., 2014), higher occupational functioning (Myaskovsky et al., 2011), or lower health care satisfaction (Myaskovsky et al., 2017; Sheppard et al., 2008, 2014). SES-based discrimination in health care was related to less positive patient-provider communication (Hausmann et al., 2011). In addition, discrimination based on sexual orientation was associated with quality of life (Jabson et al., 2011). Null findings are also summarized in Table 2 (Burgess et al., 2014; Gonzales et al., 2013, 2014; Hausmann et al., 2010, 2011; Jabson et al., 2011; Manze et al., 2010; Myaskovsky et al., 2011; Sheppard et al., 2014, 2017; Vu et al., 2016).

4. Discussion and Conclusion

Although major strides in understanding individuals’ experiences of discrimination when seeking or receiving health care have been made, guidance on how to best measure discrimination in health care is still limited (Bastos et al., 2010; Paradies et al., 2015; Priest et al., 2013; Shavers et al., 2012). We set out to contribute to filling that gap by systematically evaluating the use and adaptation of the Discrimination in Health Care Measure, a scale which has been included in numerous studies since it was first published (Bird & Bogart, 2001). Principal findings, future directions and limitations, and practical implications including those for research are discussed.

Regarding the study’s first objective, we found that the Discrimination in Health Care Measure has generally been used in cross-sectional studies with small sample sizes (< 300 participants) that have been published during the past 10+ years. The measure has been used in studies employing different data collection modes (e.g., self-administered survey, in-person interview) and with diverse populations (e.g., African American, breast cancer survivors). In relation to the study’s second objective, we found that the Discrimination in Health Care Measure has mostly been used to assess race- or ethnicity-based discrimination in health care in general, though it has been used to measure discrimination in specific health care contexts as well. Most researchers used the measure without making major modifications. Furthermore, in studies that assessed the measure’s psychometric properties, sound measure reliability was found. Last, in terms of the study’s third objective, we found that the studies included in this review suggest significant associations between discrimination in health care and an array of health-related outcomes (e.g., poor patient-provider communication, delays accessing needed care, lack of preventive care receipt).

Findings from this review suggest several future research directions. First, changes to the measure’s wording can impact responses and the reliability and validity of data. Information on the measure’s psychometric properties, with few exceptions, included only Cronbach’s alpha, which has limitations (Sijtsma, 2009). Additional research is needed to more fully evaluate the validity of the Discrimination and Health Care measure. Future studies that provide additional evidence of convergent and discriminant validity will be important additions to the literature. Studies that adapt the measure to assess discrimination based on factors other than race or ethnicity and in specific health care settings are also needed to increase knowledge regarding the measure’s psychometric properties and utility more broadly. Research that examines discrimination experiences with specific providers and subsequent behaviors with those same providers would also provide stronger evidence of causality.

Our study has limitations. We might have missed studies due to our search strategy, the databases used, or the timeframe. We also might have missed studies because of how the measure was cited in articles. In addition, we limited our study to published literature, and publication bias may have resulted in the exclusion of studies in which the measure was found to have low reliability or non-significant relationships with other variables.

4.1. Conclusion

The purpose of this systematic search and review was to identify and review studies that used the Discrimination in Health Care Measure, with a focus on synthesizing information about the study characteristics and the measure itself. The results demonstrate that this relatively short, seven-item measure has been used or adapted for diverse populations, different types of health care discrimination, and different health care settings. In some included studies, the measure also demonstrated sound internal consistency reliability and statistically significant multivariable associations with a variety of health-related outcomes. For these reasons, researchers may consider using the Discrimination in Health Care Measure when designing studies that will examine individuals’ discriminatory experiences when seeking or receiving health care. Use or adaptation of the Discrimination in Health Care Measure may additionally be useful to health professionals and administrators in assessing the effects of quality improvement and educational initiatives targeting provider bias and structural racism.

4.2. Practice Implications

Based on this review, the Discrimination in Health Care Measure has several strengths in terms of its practice implications including practice that involves research and/or evaluation related to health system quality improvement and provider education initiatives. First, given the complexity of how discrimination may be experienced, multi-item measures may have better predictive validity than single-item measures, as noted earlier (Hausmann et al., 2010). Another strength of the Discrimination in Health Care Measure is the ease with which the measure can be modified for specific contexts. For example, the phrase “when getting health care” in the question from the 2001 study by Bird and Bogart was replaced with more specific contexts such as HIV treatment (Bird et al., 2004; Burgess et al., 2014), breast cancer treatments (Jabson et al., 2011), hospitals and clinics (Padela et al., 2014; Vu et al., 2016), and a specific health care facility (Greer et al., 2011). Similarly, our findings suggest the measure can be used as originally written with diverse study populations or easily adapted for specific groups. Although the measure has most often been used to assess race- and/or ethnicity-based discrimination, the question stem may be revised to assess other types of discrimination (Jabson et al., 2011; Padela et al., 2014; Vu et al., 2016). Furthermore, the first version of the measure consisted of only seven items, so new items that are expected to be more common for a particular population or in a specific setting (e.g., based on stereotypes or on anecdotal reports) could be added without substantially lengthening the measure (Bird & Bogart, 2001).

The use or adaptation of the Discrimination in Health Care Measure may be particularly timely given the current emphasis on combatting structural racism and other types of bias in the delivery of health care. That is, the Discrimination in Health Care Measure may be easily used or adapted by health systems evaluating quality improvement and/or provider education initiatives aimed at reducing bias and discrimination. By doing so, health systems may be better able to gauge their progress in reducing discrimination from the perspectives of diverse patient groups and help to decrease health inequities in the longer term.

Supplementary Material

Lindly_Appendix

Highlights.

  • The Discrimination in Health Care Measure has been primarily used to assess racial/ethnic discrimination in multiple studies of clinical care.

  • The measure shows good reliability across studies.

  • Discrimination in health care, using this measure, has been associated with adverse health-related outcomes.

Acknowledgments

We have no known conflict of interest to disclose. The first author thanks Dr. Laura Bogart, Rand Corporation, for her collaboration on the original studies that included the Discrimination in Health Care Measure. She also thanks Dr. Larissa Myaskovsky, University of New Mexico, for her encouragement to write a paper about the measure. We thank Abigail Arias for her assistance with formatting and the references. Dr. Lindly’s effort was supported in part by grant number T32HS000063 from the Agency for Healthcare Research and Quality and the Southwest Health Equity Research Collaborative at Northern Arizona University (U54MD012388), which is sponsored by the National Institute on Minority Health and Health Disparities (NIMHD). The content is solely the responsibility of the authors and does not necessarily represent the views of the Agency for Healthcare Research and Quality.

Contributor Information

Sheryl Thorburn, School of Social and Behavioral Health Sciences, Oregon State University.

Olivia J. Lindly, Department of Health Sciences, Northern Arizona University.

References

  1. American Psychological Association. (n.d.). PsychINFO. Retrieved May 18, 2021 from, http://www.apa.org/pubs/databases/psycinfo/
  2. Bastos JL, Celeste RK, Faerstein E, & Barros AJD (2010). Racial discrimination and health: A systematic review of scales with a focus on their psychometric properties. Soc Sci Med, 70(7), 1091–99. [DOI] [PubMed] [Google Scholar]
  3. Bird ST & Bogart LM (2001). Perceived race-based discrimination and socioeconomic status (SES)-based discrimination in interactions with health care providers. Ethn Dis, 11(3), 554–63. [PubMed] [Google Scholar]
  4. Bird ST, & Bogart LM (2003). Birth control, conspiracy beliefs, perceived discrimination, and contraception among African Americans: An exploratory study. J Health Psychol, 8(2), 263–76 [DOI] [PubMed] [Google Scholar]
  5. Bird S,T, Bogart LM, & Delahanty DL (2004). Health-related correlates of perceived discrimination in HIV care. AIDS Patient Care STDS, 18(1), 19–26. [DOI] [PubMed] [Google Scholar]
  6. Bogart LM (2001). Relationship of stereotypic beliefs about physicians to health care- relevant behaviors and cognitions among African-American women. J Behav Med, 24(6), 573–86. [DOI] [PubMed] [Google Scholar]
  7. Burgess DJ, Taylor BC, Phelan S, Spoont M, Ryan MV, Hausmann LRM, Do T, & Gordon HS (2014). A brief self-affirmation study to improve the experience of minority patients. Appl Psychol Health Well Being, 6(2), 135–50. [DOI] [PubMed] [Google Scholar]
  8. Clark R, Anderson NB, Clark VR, & Williams DR (1999). Racism as a stressor for African Americans: A biopsychosocial model. Am Psychol, 54(10), 805–16. [DOI] [PubMed] [Google Scholar]
  9. Dolezsar CM, McGrath JJ, Herzig AJM, & Miller SB (2014). Perceived racial discrimination and hypertension: A comprehensive systematic review. Health Psychol, 33(1), 20–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Elsevier BV (n.d.). Scopus. Retrieved May 18, 2021, from https://www.scopus.com/
  11. Gonzales KL, Harding AK, Lambert WE, Fu R, & Henderson WG (2013). Perceived experiences of discrimination in health care: A barrier for cancer screening among American Indian women with type 2 diabetes. Women’s Health Iss, 23(1), e61–e67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gonzales KL, Lambert WE, Fu R, Jacob M, & Harding AK (2014). Perceived racial discrimination in health care, completion of standard diabetes services, and diabetes control among a sample of American Indian women. Diabetes Educ, 40(6), 747–55. [DOI] [PubMed] [Google Scholar]
  13. Greer TM, Brondolo E, & Brown P (2014). Systemic racism moderates effects of provider racial biases on adherence to hypertension treatment for African Americans. Health Psychol, 33(1), 35–42. [DOI] [PubMed] [Google Scholar]
  14. Hausmann LRM, Hannon MJ, Kresevic DM, Hanusa BH, Kwoh CK, & Ibrahim SA (2011). Impact of perceived discrimination in healthcare on patient-provider communication. Med Care, 49(7), 626–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hausmann LRM, Kressin NR, Hanusa BH, & Ibrahim SA (2010). Perceived racial discrimination in health care and its association with patients’ healthcare experiences: Does the measure matter? Ethn Dis, 20(1), 40–7. [PubMed] [Google Scholar]
  16. Jabson JM, Donatelle RJ, & Bowen D (2011). Breast cancer survivorship: The role of perceived discrimination and sexual orientation. J Cancer Surviv, 5(1), 92–101. [DOI] [PubMed] [Google Scholar]
  17. King G, & Williams DR (1995). Race and health: A multidimensional approach to African-American health. In Amick BC, Levine S, Tarlov AR, & Walsh DC (Eds.), Society and Health (pp. 93–130). Oxford University Press [Google Scholar]
  18. Klonoff EA, & Landrine H (1999). Do blacks believe that HIV/AIDS is a government conspiracy against them? Prev Med, 28(5), 451–7. [DOI] [PubMed] [Google Scholar]
  19. Kressin NR, Orner MB, Manze M, Glickman ME, & Berlowitz D (2010). Understanding contributors to racial disparities in blood pressure control. Circ Cardiovasc Qual Outcomes, 3(2), 173–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Krieger N (1994). Epidemiology and the web of causation: Has anyone seen the spider? Soc Sci Med, 39(7), 887–903. [DOI] [PubMed] [Google Scholar]
  21. Krieger N (2001). Theories for social epidemiology in the 21st Century: An ecosocial perspective. Int J Epidemio, 30(4), 668–77. [DOI] [PubMed] [Google Scholar]
  22. Krieger N, & Sidney S (1996). Racial discrimination and blood pressure: The CARDIA study of young black and white adults. Am J Public Health, 86(10), 1370–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. LaVeist TA, Nickerson KJ, & Bowie JV (2000). Attitudes about racism, medical mistrust, and satisfaction with care among African American and white cardiac patients. Med Care Res Rev, 57(Suppl. 1), 146–61. [DOI] [PubMed] [Google Scholar]
  24. Lillie-Blanton M, Brodie M, Rowland D, Altman D & McIntosh M (2000). Race, ethnicity, and the health care system: Public perceptions and experiences. Med Care Res Rev. 57(Suppl. 11), 218–35. [DOI] [PubMed] [Google Scholar]
  25. Manze M, Rose AJ, Orner MB, Berlowitz DR, & Kressin NR (2010). Understanding racial disparities in treatment intensification for hypertension management. J Gen Intern Med, 25(8), 819–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Myaskovsky L, Burkitt KH, Lichy AM, Ljungberg IH, Fyffe DC, Ozawa H, Switzer GE, Fine MJ, & Boninger ML (2011). The association of race, cultural factors, and health-related quality of life in persons with spinal cord injury. Arch Phys Med Rehabil, 92(3), 441–48. [DOI] [PubMed] [Google Scholar]
  27. Myaskovsky L, Doebler DA, Posluszny DM, Dew MA, Unruh M, Fried LF, Switzer GE, Kim S, Chang CC, Ramkumar M, & Shapiro R (2012). Perceived discrimination predicts longer time to be accepted for kidney transplant. Transplantation, 93(4), 423–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Myaskovsky L, Gao S, Hausmann LRM, Bornemann KR, Burkitt KH, Switzer GE, Fine MJ, Phillips S. l., Gater D, Spungen AM & Boninger ML (2017). How are race, cultural, and psychosocial factors associated with outcomes in veterans with spinal cord injury? Arch Phys Med Rehabi, 98(9), 1812–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Padela AI, Peek ME, Johnson-Agbakwu CE, Hosseinian Z, & Curlin F (2014). Associations between religion-related factors and cervical cancer screening among Muslims in Greater Chicago. J Low Genit Tract Dis, 18(4), 326–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Paradies Y (2006). A Systematic review of empirical research on self-reported racism and health. Int J Epidemiol, 35(4), 888–901. [DOI] [PubMed] [Google Scholar]
  31. Paradies Y, Ben J, Denson N, Elias A, Priest N, Pieterse A, Gupta A, Kelaher M, & Gee G (2015). Racism as a determinant of health: A systematic review and meta-analysis. PLoS One, 10(9), Article 0138511. 10.1371/journal.pone.0138511 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pascoe EA, & Richman LS (2009). Perceived discrimination and health: A meta-analytic review. Psychol Bull, 135(40), 531–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Peek ME, Nunez-Smith M, Drum M, & Lewis TT (2011). Adapting the everyday discrimination scale to medical settings: Reliability and validity testing in a sample of African American patients. Eth Dis, 21(4), 502–9. [PMC free article] [PubMed] [Google Scholar]
  34. Priest N, Paradies Y, Trenerry B, Truong M, Karlsen S, & Kelly S (2013). A systematic review of studies examining the relations between reported racism and health and wellbeing for children and young people. Soc Sci Med, 95, 115–27. [DOI] [PubMed] [Google Scholar]
  35. ProQuest. (n.d.). Sociological abstracts. Retrieved May 18, 2021, from http://www.proquest.com/products-services/socioabs-set-c.html
  36. Ren XS, Amick BC, & Williams DR (1999). Racial/ethnic disparities in health: the interplay between discrimination and socioeconomic status. Ethn Dis, 9(2), 151–65. [PubMed] [Google Scholar]
  37. Shavers VL, Fagan P, Jones D, Klein WM, Boyingtom J, Moten C, & Rorie E (2012). The state of research on racial/ethnic discrimination in the receipt of health care. Am J Public Health, 102(5), 953–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sheppard VB, Cavalli LR, Dash C, Kanaan YM, Dilawari AA, Horton S, & Makambi KH (2017). Correlates of triple negative breast cancer and chemotherapy patterns in black and white women with breast cancer. Clin Breast Cancer, 17(3), 232–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sheppard VB, Davis K, Boisvert M, Jennings Y, & Montalvo B (2011). Do recently diagnosed black breast cancer patients find questions about cancer fatalism acceptable? A preliminary report. J Cancer Educ, 26(1), 5–10. [DOI] [PubMed] [Google Scholar]
  40. Sheppard VB, O’Neill SC, Dilwari A, Horton S, Hirpa FA, & Isaacs C (2015). Patterns of 21-gene assay testing and chemotherapy use in black and white breast cancer patients. Clin Breast Cancer, 15(2), e83–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sheppard VB, Oppong BA, Hampton R, Snead F, Horton S, Hirpa F, Brathwaite EJ, Makambi K, Onyewu S, Boisvert M, & Willey S (2015). Disparities in breast cancer surgery delay: The lingering effect of race. Ann Surg Oncol, 22(9), 2902–11. [DOI] [PubMed] [Google Scholar]
  42. Sheppard VB, Wang J, Yi B, Harrison TM, Feng S, Huerta EE, & Mandelblatt JS (2008). Are health-care relationships important for mammography adherence in Latinas? J Gen Intern Med, 23(12), 2024–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sheppard VB, Williams KP, Wang J, Shavers V, & Mandelblatt JS (2014). An examination of factors associated with healthcare discrimination in Latina immigrants: The role of healthcare relationships and language. J Natl Med Assoc, 106(1), 15–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sijtsma K (2009). On the use, the misuse, and very limited usefulness of Cronbach’s alpha. Psychometrika, 74(1), 107–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Thomson R (n.d.). Web of science. Clarivate Analytics. Retrieved May 18, 2021 from http://apps.webofknowledge.com
  46. Thorburn S, & Bogart LM (2005). African American women and family planning services: Perceptions of discrimination. Women Health, 42(1), 23–39. [DOI] [PubMed] [Google Scholar]
  47. US National Library of Medicine and National Institutes of Health. (n.d.). PubMed Central (PMC). National Center for Biotechnology Information, U.S. National Library of Medicine. Retrieved May 18, 2021 from, https://www.ncbi.nlm.nih.gov/ [Google Scholar]
  48. Vu M, Azmat A, Radejko T, & Padela AI (2016). Predictors of delayed healthcare seeking among American Muslim women. J Women’s Health, 25(6), 586–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Williams DR (1996). Racism and Health: A research agenda. Ethn Dis, 6(1–2), 1–6. [PubMed] [Google Scholar]
  50. Williams DR, Yu Y, Jackson JS, & Anderson NB (1997). Racial differences in physical and mental health: Socio-economic status, stress and discrimination. J Health Psychol, 2(3), 335–51. [DOI] [PubMed] [Google Scholar]
  51. World Health Organization. (2017, June 27 ). Joint united nation’s statement on ending discrimination in health care settings. http://www.who.int/mediacentre/news/statements/2017/discrimination-in-health-care/en/

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Lindly_Appendix
NIHMS1840471-supplement-Lindly_Appendix.docx (26.7KB, docx)

RESOURCES