Jane K. Dickinson, RN, PhD, CDCES , Susan J. Guzman, PhD, and Jennalee S. Wooldridge, PhD
The Science of Diabetes Self-Management and Care2023, Vol. 49(3) 193–205© The Author(s) 2023Article reuse guidelines:sagepub.com/journals-permissionsDOI: 10.1177/26350106231168326journals.sagepub.com/home/tde
Abstract
Purpose: Explore the emotional experience of people with diabetes as they encounter words and phrases that have been previously identified as problematic and evaluate potential differences in their emotional impact based on type of diabetes and demographic characteristics.
Methods: A cross-sectional descriptive study employing an online survey of 107 adults with type 1 diabetes and 110 adults with type 2 diabetes. A semantic differential scale was used to examine feeling states associated with negative diabetes language. Descriptive statistics including means, standard deviations, and frequencies were calculated for all study variables. For each target word, frequencies of participants who endorsed a positive, neutral, or negative affective response on the sematic differential scale are reported.
Results: People with diabetes reported feeling blamed, misunderstood, hopeless, judged, not motivated, and not trusting in response to “noncompliant,” “unmotivated,” “in denial,” “preventable,” “failed,” “should,” “uncontrolled,” “what did you do wrong,” and “you could end up blind or on dialysis.” Participants who have type 1 diabetes and are female, White, more educated, and younger reported more negative feelings about the target words.
Conclusion: People with diabetes experience highly negative affective responses when they read and hear previously identified words and phrases considered to be judgmental and unhelpful.
Use of negative words and phrases when talking to and about people with diabetes (PWD) is an increasingly recognized problem.1-3 Negative language encompasses messages containing labels that are experienced as judgmental and stigmatizing.3-5 Such language does not reflect an understanding of the emotional and behavioral challenges of living with diabetes.1 In response, prominent diabetes organizations worldwide have published position statements to address the way health care professionals talk about and to PWD.4,5 These publications point to the need for an ongoing effort to change the language used in diabetes care, focusing on less judgmental and more personcentered, fact-based messages. For example, these guidelines suggest substituting neutral and factual messages like “the person’s A1C is 9.5%” rather than using labels such as “uncontrolled.”4 Despite this, outdated and ineffective language continues to be used in health care communication, research articles and presentations, and the media. Resistance to making a change may be due to the perception that it is too difficult and not important.5
For PWD, words and messages used by health care professionals (HCPs) influence their motivation, selfmanagement behaviors, and health outcomes.3 When PWD experience negative words and messages in clinical encounters, they can become disengaged in health care and daily self-management2 and are less likely to return for follow-up visits.6,7 Even students preparing for careers in the health professions8 and public health campaigns9 have expressed negative attitudes toward PWD, based on negative words and stereotypes.
From Department of Health & Behavior Studies, Teachers College Columbia University, Steamboat Springs, Colorado (Dr Dickinson); Behavioral Diabetes Institute, San Diego, California (Dr Guzman); Mental Health Service, VA San Diego Healthcare System, San Diego, California (Dr Wooldridge); and Department of Psychiatry, University of California San Diego, California (Dr Wooldridge).
Corresponding Author:
Jane K. Dickinson, Department of Health & Behavior Studies, Teachers College Columbia University, 525 W. 120th St. New York, NY 10027, USA. Email: dickinson@tc.columbia.edu
There is growing evidence that negative diabetes language (eg, “diabetic,” “poorly controlled,” “noncompliant”) contributes to suboptimal clinical outcomes through emotional processes. People interpret the words they hear and read; they apply meaning to those words depending on their own perspectives, context, and perceptions. Words, therefore, create a person’s reality through the meaning they assign. That meaning, in turn, influences their thoughts, feelings, and behaviors.5 For example, trust is interconnected with feeling supported by the HCP and feeling confident in self-management.10 Language used in diabetes clinical encounters affects trust,5,11 and PWD report they would have more trust in their HCPs if they changed their language.1 When PWD trust the HCP, they are more likely to share information and more closely and consistently follow the care plan.12
A narrative review by Skinner et al10 suggested that the way HCPs speak to PWD may exacerbate or even lead to diabetes distress. Diabetes distress, the negative emotions experienced as a result of living with and managing diabetes,13 is a pervasive challenge for PWD. Scare tactics, such as threats of severe diabetes-related complications, can lead to or further increase diabetes distress, which is associated with less engagement in self-management10 and elevated A1C.14,15
Qualitative research suggests PWD often feel judged and blamed by HCPs when they hear negative messages, such as “noncompliant” and “uncontrolled,”1 and perceptions of blame can erode the patient-provider relationship.2,16“When people feel blamed, judged, or stigmatized for having diabetes, they stop talking about it and they may start hiding it.”5 Words that blame and judge can contribute to a power imbalance between the HCP and the patient, placing the PWD in a passive role where they are the recipient of, rather than an active participant in, diabetes care.17
Language guidance has relied on an assumption that problematic words used in diabetes contexts have a substantial influence on the experience of diabetes4,5; however, to date, no studies have explicitly and systematically measured the affective response to what is called “negative language” in diabetes. Given that diabetes messages often do not distinguish between diabetes type, it is unclear whether those with type 1 diabetes (T1DM) and type 2 diabetes (T2DM) experience negative language similarly. The purpose of this study, therefore, was to explore (1) the emotional experience of PWD as they encounter words and phrases previously identified as problematic and (2) evaluate potential differences in their emotional impact based on type of diabetes and demographic characteristics. This study is an initial effort to understand the emotional experience associated with negative language and document its scope among PWD.
The authors used a descriptive cross-sectional design to gather information on participants’ emotional responses and use the findings to inform diabetes care and education. The study was approved by the Institutional Review Board (Protocol No. 20-166) at a university in the northeast US.
In the current study, a convenience sample of 217 adults with diabetes (T1DM: n = 107; T2DM: n = 110) was recruited from the Taking Control of Your Diabetes (TCOYD) privacy-protected online diabetes research registry. Members of the registry are people with diabetes who have attended a TCOYD event and agreed to participate in research. Access to the registry was granted through an existing relationship between the Behavioral Diabetes Institute and TCOYD. Adults, age 18 and older, with T1DM or T2DM, who can read and respond in English at an 8th-grade reading level, and who had access to the Internet were eligible to participate. Registry members were contacted through email blasts inviting them to participate in a study about the experiences PWD have with words and messages about diabetes. Six invitation email blasts were sent over the course of 3 weeks in February 2020, until the target sample of 100 adults with T1DM and 100 adults with T2DM was reached. Participants completed an anonymous, online survey through Qualtrics (Provo, UT); the first item in the survey was informed consent, and those who consented and completed all survey measures were compensated with a $20 e-gift card.
Demographic and descriptive characteristics. Participants self-reported age (in years), gender (male/female/nonbinary), diabetes type (T1DM/T2DM), diabetes duration (in years), most recent self-reported A1C, race/ethnicity (White, Hispanic/Latino, Black/African American, Asian/Pacific Islander, other), education (in years), and income (ranges from <$25 000 to >$250 000).
Affective responses to target problem words in diabetes. The authors developed a semantic differential scale examining feeling states associated with negative diabetes language. Semantic differential scales measure feelings or affective responses to objects, events, or concepts.18 A semantic differential scale presents “target” words followed by pairs of bipolar adjectives separated by a 5- or 7-point scale. The study participant chooses a point on the scale between the bipolar adjectives to describe their affective response to the target word.
Target words/phrases “noncompliant,” “unmotivated,” “in denial,” “preventable,” “failed,” “should,” “uncontrolled,” “what did you do wrong,” and “you could end up blind or on dialysis”) were selected by 2 of the authors (JKD, SJG) to be representative of commonly noted negative diabetes language.1,4,5 These 11 words/phrases were chosen because they were the most frequently identified problem words in a focus group study on diabetes language, and the bipolar affective adjectives were selected based on how these focus group participants commonly felt in response to these words.1 In the current study, participants responded to semantic differential scales in which they rated these 11 target words on six, 7-point bipolar scales (hopeful – hopeless; trusting – not trusting; encouraged – blamed; understood – misunderstood; accepted – judged; motivated – not motivated). Participants were given the following instructions:
We would like to learn how you feel, as a person with diabetes, when you read the bolded words below. For the first 11 questions, read the statement in quotes, then click on the circle that is closest to how you feel for each of the word pairs. For example, think about the word pair hopeful-hopeless. If the bolded word makes you feel hopeful, click circle #1, 2, or 3, where circle #1 is the most hopeful. If the bolded word makes you feel hopeless, click circle #5, 6, or 7, where circle #7 is the most hopeless. If the bolded word does not make you feel hopeful or hopeless, click circle #4, which is neutral.
The scale was developed for the purposes of the current study and was tested for usability and understanding among 10 people with T1DM and 10 people with T2DM prior to use in the current sample. Semantic differential scales in general have demonstrated validity in terms of assessing affective associations.18 Additionally, mean responses were reported across all affective scales for each target word. Mean scores above 4 indicated a negative affective response.
Descriptive statistics including means, standard deviations, and frequencies were calculated for primary study variables. To facilitate interpretation, responses were collapsed across each item such that responses of 1 to 3 were categorized as “positive,” a response of 4 was categorized as “neutral,” and responses of 5 to 7 were categorized as “negative.” Additionally, words were considered most problematic if more than 50% of participants indicated a negative affective response. The researchers selected this threshold to be sensitive enough to identify the relative negative affective response among words that were previously identified by PWD1 as having a negative connotation in the context of diabetes. Chi-square analyses were used to examine relationships between perceptions of target words and diabetes types. For chi-square difference tests, mean responses to each semantic differential item were coded as being <4 or >4. Bivariate correlation analyses were used to examine relationships between target words and participant characteristics. All analyses were conducted in IBM SPSS 27.0.
Overall demographic characteristics of the sample (N = 217) and comparisons by T1DM (n = 107) and T2DM (n = 110) are reported in Table 1. On average, participants were 53.76 years old (SD = 16.66) and had an A1C of 6.98 (53 mmol/mol; SD = 12.0). About two thirds of the sample identified as female (75.1%), and most participants identified as non-Hispanic White (78.8%) and reported obtaining a college degree or higher (84.3%). Participants with T1DM were significantly younger (MT1DM = 44.20, SD = 16.10 vs MT2DM = 63.00, SD = 11.10), had a longer duration of diabetes (MT1DM = 21.40, SD = 16.20 vs MT2DM = 16.16, SD = 9.30), and were more likely to identify as White (92.5% vs. 65.5%) than participants with T2DM (see Table 1). There were no other significant differences in demographic characteristics by diabetes type.
Affective responses to target words/phrases are reported in Figure 1 for both diabetes types by means and percentage of responses in the collapsed categories of positive, neutral, and negative. Participants generally reported the most negative affective responses to the following target words: “failed,” “what did you do wrong,” “you could end up blind or on dialysis,” “denial,” “uncontrolled,” “should,” “unmotivated,” “noncompliant,” and “preventable.” Participants generally reported more neutral responses to “suffered” and “diabetic.” Frequencies of negative affective responses indicate that the majority of participants endorsed feeling hopeless, misunderstood, judged, blamed, not trusting, and not motivated in relation to problematic words and phrases.
Chi-square difference tests, shown in Table 2, indicated that participants with T1DM reported more negative affective responses to the words “noncompliant,” “denial,” “preventable,” “failed,” “uncontrolled,” and “you could end up blind or on dialysis” than participants with T2DM. There were no significant differences in affective response between participants with T1DM and T2DM for “unmotivated,” “diabetic,” “should,” “suffer,” and “what did you do wrong.”
Correlations among mean affective response to target words and participant characteristics are shown in Table 3. Overall, mean negative affective responses to 6 target words/phrases had a statistically significant negative correlation with age: “noncompliant,” “denial,” “prevent,” “uncontrolled,” “you could end up blind or on dialysis,” and “failed.” Associations between negative affective responses and age were small to moderate. “Should” was the only target word in which the mean negative affective response significantly correlated with diabetes duration, and the association was small. Education was significantly positively correlated with mean negative affective response for 9 of the target words/phrases: “noncompliant,” “suffer,” “unmotivated,” “denial,” “failed,” “should,” “uncontrolled,” “what did you do wrong,” and “you could end up blind or on dialysis.” Associations were small to moderate. With the exception of “diabetic” and “suffered,” identifying as a non-White racial or ethnic group was small to moderately associated with less negative affective response to target words/phrases. Identifying as a female was associated with greater negative affective response to all target words. Strength of associations ranged from small to moderate. Mean affective response was not significantly correlated with income or A1C for any of the target words/phrases.
This study represented an initial effort to describe how adults with diabetes feel about common words or phrases that have been identified as problematic in previous qualitative work and national language guidance in diabetes care (eg, uncontrolled, noncompliant, failed). A large majority of the study participants with either T1DM or T2DM reported that these problematic words were associated with feelings of hopelessness, misunderstanding, judgement, blame, not trusting, and not motivated. These findings support previous qualitative research1 in which adults with diabetes reported that words and phrases they frequently heard in health care visits (eg, “uncontrolled,” “should,” and “noncompliance”) made them feel judged, blamed, frustrated, and angry.1
The target words or phrases that PWD reported to be emotionally negative were similar regardless of type of diabetes. Participants with either T1DM or T2DM experienced “what did you do wrong,” “noncompliant,” “unmotivated,” and “should” as similarly emotionally negative. However, where there were statistically significant differences between people with T1DM and T2DM, those with T1DM experienced the words as more affectively negative. Affective responses to a few words suggested that specific words may have a differential impact based on type of diabetes. For example, when hearing commonly used nonspecific diabetes messages such as “diabetes is preventable,” the emotional experience of those with T1DM was feeling more misunderstood, judged, blamed, and not trusting than those with T2DM.
There are several potential explanations for why more of those with T1DM experience these problematic words and phrases as affectively negative than those with T2DM. Research on diabetes stigma suggests that people with T1DM may experience higher levels of negative social perceptions around having diabetes because T1DM is more “visible” to others.19 People with T1DM experience common misunderstandings, biases, and stereotypes often attached to T2DM (eg, “it’s your fault you got diabetes”), contributing to feeling more misunderstood. People with T1DM are also less likely to reach A1C target levels than people with T2DM and may have more frequent interactions with HCPs where elevated A1C is discussed.20,21 Therefore, those with T1DM may have more conversations about increased risk of complications and may hear more commonly used problematic words and phrases, such as “uncontrolled” and “you could end up blind or on dialysis.” Those with T2DM may experience the target words as less affectively negative due to self-stigmatization, an internalization of common misbeliefs and stereotypes of diabetes,22 for example, “I have uncontrolled diabetes because I am noncompliant.” It may be that when people with T2DM agree with the problematic word or phrase, they do not feel as discouraged, judged, or blamed when experiencing such words.
The word/phrases “you could end up blind or on dialysis,” “failed,” and “what did you do wrong” were associated with the highest levels of feeling hopeless, blamed, and judged (see Figure 1). Feeling hopeless about developing complications is an item in the most highly endorsed subscale (Powerlessness) of the Diabetes Distress Scale for people with T1DM.23 Similarly, people with T2DM highly endorse worrying about developing serious complications.24 In previous research,19 having an elevated A1C was associated with greater feelings of blame and judgment. Results from the current study raise a serious concern for people with elevated A1C, who may be even more likely to hear these messages than this study sample, who were largely meeting A1C targets.
To “motivate” PWD, HCPs and public health campaigns often rely on fear-based tactics that are counterproductive. Emphasizing the risk of negative outcomes may paradoxically result in PWD feeling decreased motivation for health behavior change.25 Consistent with prior research, this study revealed that PWD feel not motivated when they hear messages implying they are unmotivated and when they hear directives (eg, “you should . . . ”). They also responded as not motivated when they hear scare tactics (ie, “you could go blind or end up on dialysis”) and when they are told they have failed. This is consistent with research showing that negatively framed messages are associated with lower motivation.26,27
Judged, misunderstood, and not trusting were themes identified in a 2015 focus group study on diabetes and language.1 Study findings are consistent with those themes in that more than 50% of respondents rated problem words in diabetes that elicit feelings of not trusting, misunderstood, or judged. Trust can facilitate open communication, which is critical to effective patient-provider relationships and care delivery.28 In one study, over 80% of adults reported withholding medically relevant information from HCPs because they did not want to be lectured or judged.29
This study was an initial effort to describe how adults with T1DM or T2DM feel about commonly used problematic language in diabetes. Examining differences in the emotional response between various problem words/phrases by type of diabetes and demographic characteristics allows for a more nuanced understanding of the emotional experience of problematic language. This cross-sectional survey study relied on self-reported diabetes status and A1C. While self-reported diabetes status and A1C are not as ideal as having objective data, evidence has shown that self-reported diabetes status is highly accurate.30 Study findings showed that participants who are female, White, more educated, and younger reported more negative feelings about the target words. However, these associations must be interpreted with caution because these results may reflect sample limitations (largely female, educated, White, A1C at/near treatment target). Additionally, participants in the study sample with T1DM tended to be younger and were more likely to be White than participants with T2DM. Thus, the observed differences between participants with T1DM and T2DM may be in part explained by differences in demographic characteristics. Future studies should examine a more diverse sample around gender, education level, race/ethnicity, health inequities (eg, health insurance status), and A1C levels. Furthermore, the negative words in diabetes examined in this study were not exhaustive, and additional words and phrases as well as positive and neutral words could be examined in relation to affective response in future research.
Future research could further examine relationships among feelings about problematic language and diabetes distress, perceptions of stigma, and engagement with diabetes self-management.
These results underscore the need to move away from problematic language and messages. There are several practical approaches applicable to diabetes care supported by the current study. Discussions of elevated risk factors for complications can be made more effective by focusing on the benefits of treatment recommendations and not labeling people or threatening negative health outcomes. The diabetes lexicon includes labels commonly used in electronic health records (eg, uncontrolled, nonadherent) about people who are not at treatment targets. The use of words that are neutral and fact based may help people feel less blamed. For example, report A1C objectively (A1C = 10.9%, 96 mmol/mol or A1C > 9%, 75 mmol/mol) rather than using a subjective descriptor (poorly controlled). Furthermore, HCPs can facilitate willingness for and engagement in treatment by avoiding “should” and other directives and instead support and talk about the person’s own reasons and means for behavior change.
Overall, the results of this study indicate that a large percentage of PWD experience negative feelings about commonly used problematic words and phrases in diabetes. These findings add empirical support to the recommendations posed by diabetes language guidance statements and suggest that efforts to change how diabetes is discussed are important and relevant.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr. Wooldridge was supported in part by a Career Development Award [1IK2RX003634] from the United States (U.S.) Department of Veterans Affairs Rehabilitation Research and Development Service.
Jane K. Dickinson https://orcid.org/0000-0003-0732-8116
