Being diagnosed with a chronic illness is a life-altering experience that can be shaped, for better or worse, by psychological factors. How patients think about their illness—their core beliefs about what it means and what it might bring such as whether it is catastrophic, manageable, or even an opportunity—can influence how they respond and adapt. This research introduces the concept of illness mindsets and presents the initial validation of the Illness Mindset Inventory (IMI), a new tool designed to assess these beliefs and their implications for health and well-being. Study 1 examines the factor structure, internal reliability, and discriminant validity of the 9-item IMI in N = 201 healthy participants and N = 200 participants with cancer, diabetes, cardiovascular disease, and/or chronic pain. Study 2 investigates cancer patients (N = 463) with different degrees of illness severity and tests the pre-registered hypothesis that the IMI will account for variability in functioning over and above measures of illness severity. In Study 1, illness mindsets were associated with between 5.7 and 12.1% additional variance in physical, social, and emotional functioning, above and beyond disease status. In Study 2, illness mindsets accounted for between 6.9 and 12.0% additional variance in physical functioning, social functioning, and emotional distress in people diagnosed with cancer above and beyond cancer stage, cancer status, trait optimism, and self-efficacy. Illness mindsets may help account for variance in individual functioning beyond disease status and disease severity. Future research can probe the IMI’s utility in supporting patient care; in predicting functioning before, during, and after a diagnosis; and as a potential target for intervention.
For many patients, the diagnosis of a chronic illness is a life-altering experience that marks the beginning of years of disease management (Stanton et al., 2007). While diseases like arthritis, cancer, cardiovascular disease, and diabetes are physiologically complex, psychological factors can also shape how a chronic illness impacts an individual’s functioning (Miller et al., 2009). As a result of psychological factors, patients with the same illness and similar pathology may have very different lived experiences, health outcomes, and even life expectancies (Giltay et al., 2006).
Understanding the patient’s perspective in the context of illness is increasingly considered critical to delivering high-value, patient-centered care. Indeed, since the Patient Protection and Affordable Care Act was enacted in 2010 in the United States, researchers and healthcare providers have increasingly looked for ways to capture the perspective of patients. Although measures of patients’ reports on their own health status (e.g., their physical, emotional, and social functioning; sometimes referred to as Patient-Reported Outcome Measures, or PROs) are increasingly used, measures capturing patients’ general orientations toward common illnesses, such as cancer, diabetes, cardiovascular disease, or chronic pain syndromes, are not yet considered. In this paper, we introduce the concept of patients’ illness mindsets—their core assumptions about the nature of their illness, such as whether it is catastrophic, manageable, or an opportunity for positive change.
Mindsets have proven to be an important component in shaping behavior in other domains (e.g., intelligence, stress) and thus may be similarly useful in the context of illness. As such, mindsets may be an important—yet currently overlooked—piece of the patient experience puzzle (see Fig. 1). Although illness mindsets may play a key role in shaping health and behavior, we currently do not have a tool to measure them. Here, we present a brief 9-item scale, the Illness Mindset Inventory (IMI), and test its ability to account for variability in patient physical, social, and emotional functioning.
Fig. 1
Psychological assessments, clinical assessments, and functioning outcomes described in this study. Examples of clinical assessments are provided. Examples of measures of functioning are shown from the PROMIS-29 scale. Example items for measures of optimism (Life Orientation Test Revised [LOT-R]- 10 items); self-efficacy (General Self Efficacy [GSE]—6 items); Illness Mindsets (Illness Mindset Inventory [IMI]—9 items); and appraisals (Brief Illness Perception Questionnaire [IPQ-brief]—9 items) are shown. Connections between clinical assessments, psychological assessments, and functioning indicate hypothesized bidirectional relationships
Mindsets are core assumptions that people make about a conceptual domain or category. While the terms mindset and belief are sometimes used interchangeably, mindsets are specific types of beliefs that have motivational value and orient people to particular sets of expectations, attributions, and goals (Dweck, 2019). Our mindsets help us organize and simplify complex information in ways that create meaning (e.g., Why is this happening?), make predictions (e.g., What will happen next?), and motivate action (e.g., What should I do next?).
Decades of research have highlighted the impact of mindsets in the domain of intelligence (e.g., “intelligence is fixed” vs. “intelligence is malleable”) on academic performance and educational achievement (Blackwell et al., 2007; Paunesku et al., 2015). Another line of research on mindsets about health-related constructs like stress (e.g., “stress is debilitating” vs. “stress can be enhancing”) and healthy eating (e.g., “healthy foods are disgusting and depriving” vs. “healthy foods can be delicious”) has revealed how mindsets can also influence health and health-related behaviors (Boles et al., 2021; Crum et al., 2011, 2013, 2017).
Regardless of the domain, mindsets can evoke changes in emotion, attention, motivation, and physiology that can lead to self-fulfilling outcomes (Walton & Wilson, 2018). For example, an individual who is oriented toward the debilitating (rather than the enhancing) aspects of stress may attend to the physiological symptoms of a stress response, worry that their stress will impact their work performance, and avoid the very behaviors that could mitigate the external stressor. As a result of these processes, the individual’s stress response may increase, their performance might suffer, and the situation could provide further evidence to support the mindset that stress is indeed debilitating (Crum et al., 2020).
Mindsets about illness
In the present work, we introduce a new category of mindsets: mindsets about illness. Mindsets about illness are core assumptions people hold about the nature of chronic illness in general or about a specific chronic illness, such as diabetes, cancer, or hypertension. While many mindsets about illness may exist, we describe three in this paper: the mindset that “illness is a catastrophe”, the mindset that “illness is manageable”, and the mindset that “illness can be an opportunity” (hereafter referred to as the “catastrophe mindset, “manageable mindset”, and “opportunity mindset”, for brevity). The catastrophe mindset describes an association between the concept of a chronic illness and a globally negative impact on one’s life. The manageable mindset captures the understanding that while a chronic illness may have negative impacts, it can be accepted, dealt with, and controlled to a meaningful degree. Finally, the opportunity mindset reflects a nuanced understanding of a chronic illness as a major life event that can serve as a catalyst for positive change, such as emotional growth or enhanced personal connections.
In addition to drawing from mindset theory, our concept of illness mindsets builds on the well-established literature on illness perceptions. Illness perceptions, often assessed using tools like the Illness Perception Questionnaire (IPQ), reflect individuals’ appraisals of their specific illness—such as its causes, timeline, controllability, or consequences. While illness perceptions provide valuable information about how a person evaluates the personal impact of their condition, illness mindsets operate at a higher level of abstraction: they capture broad, organizing assumptions about the nature and meaning of chronic illness itself (e.g., “illness is a catastrophe” vs. “illness can be an opportunity”). This theoretical distinction parallels differences between mindset research and appraisal theories in other domains, such as stress and emotion, and allows for generalization across illnesses and timepoints. We discuss these distinctions in more detail below.
Importantly, illness mindsets—similar to mindsets about stress or intelligence—are not assessments of fact or experience, and are therefore not necessarily true or false, or right or wrong. Instead, they are subjective mental frameworks regarding the essence of a category—in this case an illness—that simplify and guide people through a complex and uncertain reality. Regardless of the objective nature of an illness, as assessed via clinically diagnosed disease presence (e.g., whether or not a patient has been diagnosed with cancer) or severity (e.g., the clinically diagnosed stage of that cancer), all illnesses could be perceived as more or less of a catastrophe, more or less manageable, and more or less of an opportunity. However, the mindsets people hold may lead them down very different paths. For example, consider the impact of viewing a particular illness as “a catastrophe” on emotions, attention, and behavior. That mindset might lead people to feel a sense of dread or despair, and to be overly focused on all the ways their illness is, in fact, a catastrophe. Behaviorally, this mindset might lead them to withdraw from life and/or fail to adhere to treatment protocols.
The utility of illness mindsets relative to other measures of patient psychology
We propose that illness mindsets capture a valuable yet currently overlooked dimension of patient psychology—one that can help clinicians, healthcare providers, public health officials, and researchers better understand and support patients. To situate this construct, it is important to clarify how illness mindsets relate to, and diverge from other well-established psychological constructs in health research. Illness mindsets are related to, but distinct from: (1) more specific patient appraisals or perceptions about illness, such as the beliefs about its causes, timeline, consequences and controllability; cognitive distortions like catastrophizing and personalization; and coping appraisals, and (2) more general psychological dispositions or personality traits, such as optimism, self-efficacy, and locus of control (Baum & Posluszny, 1999; Lau & Hartman, 1983; Leventhal et al., 1980; Petrie et al., 2007; Stanton et al., 2007). Figure 1 illustrates how illness mindsets sit at a middle-level of abstraction between more specific appraisals and more general personality traits.
One useful comparison is with the Illness Perception Questionnaire (IPQ), which assesses individuals specific perceptions regarding their illness across nine dimensions (timeline, consequences, causes, personal control, treatment control, identity, concern, understanding, emotional response) (Petrie et al., 2007). While illness mindsets may superficially resemble certain IPQ items (“How much does the illness affect your life?”; “How concerned are you about your illness?”), they are conceptually distinct. Illness mindsets are not focused on specific perceptions of how the illness is affecting their life currently but instead reflect broader, more generalized assumptions about the nature and meaning of illness (e.g., “Cancer can be an opportunity to make positive life changes”; “Cancer ruins most aspects of a person’s life”). Similarly, cognitive distortions like catastrophizing are specific ways of thinking about the illness, likely driven by the more general mindset about the nature of the illness as a “catastrophe”, but not redundant constructs.
Illness mindsets also relate to stress and coping theories, particularly Lazarus and Folkman’s transactional model of stress. In this framework, individuals appraise stressors as either threats or challenges (primary appraisal) and assess their ability to cope with those stressors (secondary appraisal). The “catastrophe” mindset, for instance, may relate to threat appraisals, while the “manageable” and “opportunity” mindsets may relate to challenge appraisals or positive secondary appraisals. However, illness mindsets differ in that they operate at a more general level of meaning making which can organize cognition, emotion and physiology in ways more specific situational appraisals do not (see Jamieson et al., 2018, for a detailed review of the distinction between mindsets and appraisals).
Illness mindsets may also intersect with constructs in the Commonsense Model of Self-Regulation (CSM), such as illness coherence (i.e., how well an individual believes they understand their illness) and the integration of illness into one’s life. However, illness mindsets go beyond coherence to capture the general stance a person takes toward illness—whether they see it as fundamentally catastrophic, manageable or even an opportunity for growth. As such, illness mindsets may color or change the meaning of any more specific understanding or integration of an illness.
As a second point of distinction, while illness mindsets are related to broader traits like optimism and self-efficacy, they are more specific in topic. For example, optimism reflects a general expectation that good things will happen across all or many domains of one’s life (e.g., “Overall, I expect more good things to happen to me than bad”) while illness mindsets reflect domain-specific beliefs about illness itself (e.g., “Illness can lead to positive life changes”). Similarly, while Health Locus of Control (HLC) theory distinguishes between internal and external attributions for health outcomes, it focuses on one’s dispositional assessment of who has control, whereas illness mindsets capture core assumptions about the meaning or nature of illness itself (e.g., whether illness is an opportunity or a catastrophe). Because they are more localized to the domain or category of a particular illness, illness mindsets may be a more relevant measure when considering impacts on health outcomes in the context of illness.
Because they are domain-specific yet broadly framed, illness mindsets occupy a unique level of abstraction that lends itself to intervention (Walton & Crum, 2021). Where appropriate, rather than attempting to persuade a patient that their illness is not actually affecting them, a provider might share evidence from research or other patients that support the notion that a given illness could be manageable or even an opportunity. From a practical standpoint, measuring mindsets also has advantages, as they can be measured early in a diagnosis or even before someone is ill (before one could ask a person how much a treatment is working or how much the illness has affected their life). Indeed, as illustrated in the studies that follow, illness mindsets can be measured and compared across time points of a disease (before, during, and after a cancer diagnosis) and across illness types and severities (e.g., between people who have cancer vs. diabetes; between people who have stage I vs. stage III cancer) (Walton & Crum, 2021).
In sum, while illness mindsets overlap conceptually with illness perceptions, coping appraisals and personality traits, they offer a distinct and clinically actionable perspective. We do not propose that illness mindsets replace existing constructs, but rather that they complement them—offering new insight into how people interpret and respond to illness and therefore offering another potential target for change.
Scope of current research and hypotheses
The goal of this paper is to introduce and validate a measure to capture illness mindsets. After briefly reviewing our process for item generation and scale development, we present results from two studies. Study 1 confirms the factor structure and internal reliability of the scale and explores the relationship between the IMI, functioning, and measures of discriminant validity (affect, coping, and illness perceptions) in a sample of N = 201 healthy subjects and N = 200 subjects with one of four chronic illnesses (cancer, diabetes, cardiovascular disease, or chronic pain). As detailed below, Study 2 investigates our pre-registered hypotheses about the relationship of the IMI to measures of illness severity (cancer stage and cancer status), functioning, and measures of discriminant validity (optimism and self-efficacy) in a sample of N = 463 cancer patients and survivors.
More specifically, in seeking to establish the initial reliability, validity, and utility of the scale, we explore four important topics. First, we explore the scale’s internal structure and reliability to answer the questions: Are these mindsets distinct from or related to each other? Do the items for each mindset hold together well?
Second, we explore the scale’s discriminant validity. How do illness mindsets relate to psychological constructs, such as more situationally and personally specific illness perceptions and more general personality traits, such as optimism and self-efficacy? As discussed in more detail below, we conceptualize the IMI as being related to these constructs but not redundant with them. We aim to test whether this theoretical distinction also holds empirically.
Third, we explore concurrent validity, or the relationship between the IMI and other important metrics, such as the clinically diagnosed presence and severity of disease and patient reported functioning. This, in turn, raises three main questions. First, do the mindsets of people who have a particular disease (e.g., cancer) differ from those who have a different disease or who do not currently have a disease? Second, does greater severity of the diagnosis correlate with more “pessimistic” mindsets? Third, what is the relationship between the IMI and measures of physical, emotional, and social functioning—that is, are more “pessimistic” mindsets correlated with lower levels of functioning?
Finally, we seek to explore the incremental validity of the IMI in helping to account for variability in a patient’s physical, social, and emotional functioning over and above the variability accounted for by the severity of disease. How much variability in a person’s physical, social, and emotional functioning is associated with their mindset, compared to or controlling for their clinically diagnosed disease severity? Put differently, for people with roughly equal clinical diagnoses and severities, does the IMI account for additional variance in their functioning?
If a few short questions can reveal the essence or “gist” of how a person feels about the nature of a particular illness, the implications are numerous. Should the IMI hold up empirically as a distinct construct accounting for variability in physical, emotional, and social functioning, as we predict it will, it could serve as a useful variable in (1) supporting patient care (revealing otherwise unknown/unmeasured information that could be used to direct treatment or flag a patient for referrals); (2) predicting functioning before, during, and after a diagnosis; and (3) offering a potential target for intervention.
Item generation, scale development, and initial piloting
The final 9-item Illness Mindset Inventory (IMI) was developed using a theory-driven process informed by existing work on mindsets and illness perceptions (Blackwell et al., 2007; Broadbent et al., 2006; Crum et al., 2013), benefit finding (Helgeson et al., 2006), pain catastrophizing (Sullivan et al., 1995), and post-traumatic growth (Tedeschi & Calhoun, 1996). Initially, a larger pool of items was created—approximately four to five items per mindset domain (catastrophe, manageable, and opportunity). An exploratory factor analysis of the nine items was then conducted on a pilot sample of N = 220 participants (N = 93 healthy; N = 127 with a chronic illness). A three-factor structure fit the data best and explained 59% of the total variance (χ2(18, 220) = 25.94; p = 0.101). This structure also aligned with and supported our theoretical and conceptual distinctions between the three mindsets. Scale items and factor loadings for the three-factor model are included in Table 1. Full details on the original item pool, decision criteria for item retention, and factor loadings for the full and final versions are provided in the Supplementary Materials.
Table 1
Illness mindset inventory items & factor loadings
Factor loading
Factor 1
Factor 2
Factor 3
Catastrophe mindset
1. A chronic illness negatively impacts nearly all aspects of life
0.80
0.04
0.01
6. Having a chronic illness spoils most parts of life
0.76
0.01
0.05
8. A chronic illness ruins most aspects of life
0.84
− 0.02
0.00
Manageable Mindset
3. A chronic illness can be managed so that you can live a normal life
0.05
0.11
0.79
5. A chronic illness is something that can be dealt with
0.02
− 0.07
0.88
9. You can live a relatively normal life with a chronic illness
− 0.30
0.04
0.51
Opportunity Mindset
2. A chronic illness can be an opportunity to make positive life changes
− 0.02
0.61
0.22
4. Having a chronic illness can allow you to find more meaning in life
− 0.04
0.67
− 0.12
7. Having a chronic illness is a challenge that can make you stronger
0.04
0.79
− 0.03
Individual items and factor loadings for three illness mindsets. Item numbers indicate the order in which items should be administered. Each item is rated on a 6-point Likert scale from Strongly Disagree (1)—Strongly Agree (6). The term ‘a chronic illness’ can be replaced with the name of a specific illness (e.g., ‘Cancer’). Summary are calculated for each factor by taking the average of the individual items. No items are reverse scored. Factor loadings from an exploratory factor analysis of N = 220 participants (N = 93 healthy; N = 127 chronically ill) from an initial pilot study. This study is described in detail in the Supplementary Materials
Bolded values represent factor loadings to indicate which factor items loaded most strongly to. For instance the three items from the catastrophe mindset loaded most strongly onto Factor 1, so the highest values are bolded here
Study 1
The aim of Study 1 was to confirm the factor structure and internal reliability of the scale and explore the relationships among the IMI, functioning, and measures of convergent/divergent validity (affect, coping, and illness perceptions).
Methods
This study was reviewed and approved by the Stanford University Institutional Review Board and Administrative Panel for the Protection of Human Subjects.
Participants and procedure. We recruited N = 401 participants using Qualtrics Panels, an online service that facilitates recruitment of demographically representative populations. These were recruited in two groups: N = 200 participants who indicated a current diagnosis of one of four chronic illnesses: cancer (N = 50), diabetes (N = 50), cardiovascular disease (N = 50), or chronic pain (N = 50), as well as N = 201 participants with no current or lifetime history of any chronic illness (see Table C1 in the Supplementary Materials demographic details by group). Participants completed a survey that included questions about health history, demographics, the nine-item IMI, and the measures of convergent/divergent validity (e.g., illness perceptions) and criterion validity (e.g., physical, social, and emotional functioning) described in more detail below. To concretize and constrain the domain participants were judging, IMI items were phrased using the specific illness names (e.g., “cancer ruins most aspects of life”) rather than the generic term “chronic illness”. Participants in the chronic illness group(s) responded to scale items that corresponded with their indicated illness. Participants in the healthy condition were randomly assigned to one of the four versions of the scale to match the illness-specific wording of the clinical sample.
Measures
In addition to the nine-item IMI, participants completed measures of illness perceptions, coping strategies, affect, physical functioning, social functioning, and emotional distress.
Illness perceptions were measured using The Brief Illness Perception Questionnaire (IPQ-B), which assesses nine cognitive representations of a person’s experience with their own illness (e.g., consequences, timeline, understanding, etc.) based upon the theoretical framework outlined in the Self-Regulatory Model of Illness (Broadbent et al., 2006). Items were assessed using a numeric slide scale ranging from 0 to 10 with item-specific anchors.
Coping strategies were measured using the short version of the Coping Strategies Inventory (CSI-SF). The CSI-SF consists of 16 items assessing four factors of general coping styles pertaining to how a person is coping with a particular illness: problem-focused engagement (e.g., “I make a plan of action and I follow it”); problem-focused disengagement (e.g., “I try to put the problem out of my mind”); emotion-focused engagement (“I try to let my emotions out”); and emotion-focused disengagement (e.g., “I keep my thoughts and feelings to myself”) (Addison et al., 2007). Items were assessed using a five-point Likert scale ranging from Never (1)—Very Often (5).
Affect was measured using the Positive and Negative Affect Scale (PANAS). The scale consists of 20 items, each of which describes an affective state. Participants were asked to rank the extent to which they felt each item on a five-point Likert scale from 1 (very slightly or not at all) to 5 (extremely) over the past week.(Crawford & Henry, 2004).
Functioning was measured across three domains: physical, social, and emotional. Physical functioning was measured using the physical functioning sub-scale of the Patient-Reported Outcomes Measurement Information System—29 (PROMIS-29) (Cella et al., 2010). Social functioning was measured using the ability to participate in social roles sub-scale of the PROMIS-29. Emotional functioning was measured using the depression and anxiety subscales of the PROMIS-29, which were combined to create an emotional distress variable (as suggested in the literature; e.g., (Hays et al., 2018)). Each subscale consists of four individual items rated on a five-point Likert scale. Means, SDs, and alphas for the scales and subscales (when appropriate) are included in Table 2.
Table 2
Correlations with measures of convergent, discriminant, and criterion validity in participants with chronic illness (N-200) in Study 1
Variable
M
SD
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Illness mindsets
1, Catastrophe
3.31
1.33
(.90)
2. Manageable
4.52
1.04
− .56**
(.88)
3. Opportunity
4.44
1.08
− .41**
.65**
(.85)
Illness perceptions
4. Consequence
6.38
2.52
.65**
− .42**
− .28**
–
5. Timeline
7.83
2.80
.17*
− .18*
− .32**
.28**
–
6. Pcrs. Control
5.56
2.63
− .36**
.44**
.42**
− .29**
− .21**
–
7. Tx. Control
7.11
2.46
− .29**
.44**
.31**
− .21**
− .21**
.56**
–
8. Identity
5.77
2.84
.56**
− .47**
− .27**
.71**
.35**
− .27**
− .28**
–
9. Concern
7.58
2.58
.30**
− .26**
− .13
.44**
.28**
− .26**
− .12
.40**
–
10. Understanding
7.55
2.06
− .14
.22**
.22**
.06
.03
.21**
.24**
.04
.05
–
11. Emo. Response
6.21
2.96
.55**
− .48**
.29**
.58**
.19**
− .38**
− .18*
.53**
.48**
− .06
–
Affect
12. Pos. Affect
29.84
9.83
− .33**
.44**
.46**
− .20**
− .11
.41v
− .30**
− .20**
− .07
.18**
− .28**
(.94)
13. Ncg. Affect
21.87
9.38
.36**
− .22**
− .11
.39**
.05
− .12
− .09
.35**
.22**
− .08
.52**
− .23**
(.93)
Coping strategies
14. Prob. Foe. Eng
3.53
0.65
− .24**
.30**
.46**
− .11
− .08
.35**
.25**
− .07
− .07
.29**
− .23**
.59**
− .15*
(.73)
15. Prob. Foe. Discng
3.00
0.70
.20**
− .08
− .02
.16*
.08
.05
− .03
.10
.00
.04
.15*
.06
.20**
.06
(.52)
16. Emot. Foe. Eng
2.98
0.83
− .04
.08
.26**
.16*
− .08
.28**
.20**
.15*
.12
.13
.13
.39**
.15*
.34**
.11
(.82)
17. Emot. Foe. Discng
3.20
0.80
.38**
− .21**
− .17*
.21**
.04
− .16*
− .14
.22**
.12
− .21**
.32**
− .30**
.45**
− .24**
.15*
− .24**
(.70)
Functioning
18. Phys. Functioning
3.65
1.06
− .51**
.44**
.27**
− .54**
− .23**
.24**
.36**
− .54**
− .24**
− .00
− .32**
.30**
− .25**
.11
− .06
− .01
− .09
(.90)
19. Soc. Functioning
3.11
1.05
− .64**
.47**
3 Q**
− .58**
− .22**
30**
.28**
− .62**
− .31**
.04
− .53**
− .40**
.10
− .17*
− .06
− .29**
.73**
(.95)
20. Emo. Distress
2.41
1.06
.49**
− .30**
− .18**
.45**
.06
− .11
− .11
.41**
.21**
− .03
.58**
− .32**
.77**
− .25**
.21**
.11
.54**
− .32**
− .50**
(.94)
*p < .05 **p < .01. Mand SD are used to represent mean and standard deviation, respectively. Values in parentheses on the diagonal are Cronbach’s Alpha, a measure of internal consistency for non-single item subscales. Illness Mindsets measured with the Illness Mindset Inventory (IMI); Illness Perceptions measured with the Illnesses Perceptions Questionnaire—Brief Version (IPQ-Brief); Affect measured with the Positive and Negative Affect Scale (PANAS); Coping Strategies measured with the Coping Strategies Inventory—Short Form (CSI-SF); Wellbeing measured with the Patient Reported Outcomes Measurement Information System—29 (PROMIS-29)
Analytical approach
To test the degree to which the IMI is an internally consistent and reliable measure, we conducted a confirmatory factor analysis (CFA) using lavaan version 0.5-23(Rosseel, 2012) in R version 3.3.1 (R Core Team: R: A Language & Environment for Statistical Computing., 2020) and Cronbach’s Alpha to measure internal reliability of the items (Table 1). To examine whether illness mindsets would relate to—but not be redundant with—existing measures of illness perceptions, coping behaviors, and affect, we explored the convergent and discriminant validity of the IMI with a series of correlational analyses (Table 2).
To examine whether mindsets would differ between individuals who have or do not have a chronic disease, we examined whether mindsets differed for each disease category between those who were healthy and those who had the illness using linear regression models that controlled for demographic variables (age, race, sex, education, income). To concretize and constrain the domain participants were judging, IMI items were phrased using the specific illness names (e.g., “cancer ruins most aspects of life”) rather than the generic term “chronic illness”. Participants in the chronic illness group responded to scale items that corresponded with their indicated illness. Participants in the healthy condition were randomly assigned to one of the four versions of the scale to match the illness-specific wording of the clinical sample.
To examine whether mindsets would differ for people with different diseases (cancer, hypertension, diabetes, and chronic pain), we also examined differences in the IMI between the four disease types using ANCOVAs and controlling for age, gender, race, educational attainment, and annual household. Tukey’s test of multiple comparisons was used to further probe for any differences. To examine the concurrent validity of the IMI with functioning, we ran a series of pairwise correlations between the mindsets and measures of functioning (physical, social, and emotional).
To test the incremental validity of the IMI, we ran three separate hierarchical linear regression models to explore whether illness mindsets explain additional variance in these three facets of functioning (physical, social, and emotional) above and beyond the presence of a chronic illness (Table 3). In step 1, disease status (healthy vs. chronic illness) was included as the sole predictor of the outcome variable of interest. The three illness mindsets were added individually in three separate steps to determine the independent contribution of each mindset. In step 2a, we included the catastrophe mindset; in step 2b, we included the manageable mindset; and in step 2c, we included the opportunity mindset. In step 3, we included all three mindsets to quantify the amount of variance explained by all three illness mindsets. This same series of steps was conducted separately for physical functioning, social functioning, and emotional distress as dependent variables. In all steps, we controlled for age, gender, race, educational attainment, and annual household income.
Table 3
Incremental validity of illness mindsets in predicting physical functioning, social functioning and emotional distress beyond disease status in Study I
Step 1
Step 2a
Step 2b
Step 2c
Step 3
Physical functioning
Intercept
4.94***(4.63–5.26)
5.89*** (5.51–6.27)
3.70*** (3.25–4.16)
4.15*** (3.70–4.60)
4.80*** (4.18–5.41)
Disease Status
− 1.18*** (− 1.36–1.01)
− 1.16*** (− 1.33–1.00)
− 1.20*** (− 1.36–1.03)
− 1.27*** (− 1.44–1.09)
− 1.20*** (− 1.36–1.03)
Illness Catastrophe
− 0.25*** (− 0.31–0.19)
− 0.19*** (− 0.25–0.12)
Illness Manageable
0.28*** (0.20–0.35)
0.15** (0.05–0.25)
Illness Opportunity
0.19*** (0.11–0.26)
0.04 (− 0.05–0.14)
Observations
399
399
399
399
399
F Statistic
36.43** (df = 6; 392)
45.03*** (df = 7; 391)
42.48*** (df = 7; 391)
36.30*** (df = 7; 391)
38.66*** (df = 9; 389)
R2
0.36
0.45
0.43
0.39
0.47
∆R2 (From Step 1)
0.088***
0.074***
0.036***
0.114***
Social functioning
Intercept
4.50*** (4.14–4.85)
5.65*** (5.23–6.07)
3.08*** (2.58–3.59)
3.49*** (2.99–3.99)
4.44*** (3.77–5.12)
Disease Status
− 1.57*** (− 1.76–1.37)
− 1.54*** (− 1.72–1.36)
− 1.58*** (− 1.77–1.40)
− 1.68*** (− 1.87–1.48)
− 1.60*** (− 1.78–1.41)
Illness Catastrophe
− 0.30*** (− 0.37–0.24)
− 0.24*** (− 0.31–0.16)
Illness Manageable
0.32*** (0.23–0.40)
0.13* (0.02–0.24)
Illness Opportunity
0.24*** (0.15–0.32)
0.09 (− 0.01–0.19)
Observations
399
399
399
399
399
F Statistic
45.56*** (df = 6; 392)
57.51*** (df = 7; 391)
51.75*** (df = 7; 391)
46.28*** (< tf = 7; 391)
49.00*** (df = 9; 389)
R2
0.41
0.51
0.48
0.45
0.53
A R2 (From Step 1)
0.096***
0.070***
0.042***
0.121***
Emotional distress
Intercept
2.33*** (1.99–2.67)
1.60*** (1.18–2.02)
3.08*** (2.58–3.58)
2.76*** (2.28–3.25)
2.08*** (1.38–2.77)
Disease Status
1.11*** (0.92–1.29)
1.09*** (0.91–1.27)
1.11*** (0.93–1.30)
1.15*** (0.96–1.34)
1.10*** (0.92–1.29)
Illness Catastrophe
0.19*** (0.12–0.26)
0.16*** (0.09–0.24)
Illness Manageable
− 0.17*** (− 0.25–0.08)
− 0.07 (− 0.19–0.04)
Illness Opportunity
− 0.10* (− 0.19–0.02)
− 0.01 (− 0.12–0.09)
Observations
399
399
399
399
399
F Statistic
25.80*** (df = 6; 392)
28.20*** (< df = 7; 391)
25.06*** (df − 7; 391)
23.23*** (df = 7; 391)
22.30*** (df = 9; 389)
R2
0.28
0.34
0.31
0.29
0.34
A R2 (From Step 1)
0.052***
0.027***
0.011*
0.057***
Asterisks indicate significance such that *p < 0.05;**p < 0.01; ***p < 0.001. Standardized Beta’s and 95% CPs are presented in parentheses Emotional distress was assessed using the anxiety and depression subscales of the PROMIS-29. All models controlled for age, race, gender, education, and annual household income
Results
Confirmatory factor analysis and internal consistency
The confirmatory factor analysis yielded measures of absolute fit (RMSEA = 0.078; SRMR = 0.034) and relative fit (TLI = 0.952; CFI = 0.968) that fell within acceptable ranges, which supported our hypothesized three-factor structure and confirmed the findings from previous exploratory factor analyses (see Table C2 in Supplementary Materials). Internal consistency, measured by Cronbach’s Alpha, was strong and ranged from α = 0.85 − 0.90 for each of the three mindsets. Although there were few cross-loadings among the factors, correlations between the mindsets were moderate to high and in expected directions (catastrophe mindset was negatively correlated with manageable (r = − .56) and opportunity (r = − .41) mindset; manageable and opportunity mindset were positively correlated (r = .65)).
Convergent/divergent validity
Illness mindsets and psychological variables
On average, moderate correlations were observed between mindsets and the subscales of the IPQ. The catastrophe mindset was associated with percptions of greater consequences (r = .65), timeline (r = .17), identity (r = .56), concern (r = .30), and emotional response (r = .55), whereas the manageable and opportunity mindsets were associated with perceptions of greater personal control (r = .44; r = .42), treatment control (r = .44, r = .31), and understanding (r = .22, r = .22). See Table 2 for a complete list of correlations.
The catastrophe mindset related to fewer problem-focused “engaged” coping strategies (r = − .24), more problem-focused and emotion-focused “disengaged” coping strategies (r = .20; r = .38), lower positive affect (r = − .33), and greater negative affect (r = .66). The manageable mindset displayed the opposite trends. Like the manageable mindset, the opportunity mindset was associated with more problem-focused coping (r = .46) and more positive affect (r = .46); however, it was also associated with more emotion-focused coping (r = .26) and was not significantly correlated with negative affect (r = − .11). See Table 2 for additional details.
Relationship between IMI and disease status and type
We explored differences in mindsets across disease status using multiple linear regression models that controlled for age, gender, race, educational attainment, and annual household income. Interestingly, people who had a chronic illness did not differ from those who were healthy with respect to the catastrophe (ß = 0.08 p = 0.56) and manageable (ß = 0.04; p = 0.69) mindsets. However, people who were ill were significantly more likely to view their illness as an opportunity (ß = 0.45; p < 0.001). We also report on differences in significant variability in mindsets across illness type within the chronic illness subgroup. Analyses of covariance (ANCOVA) controlling for age, gender, race, educational attainment, and annual household income showed significant differences in the catastrophe (F = 69.38; p < 0.001), manageable (F = 17.54; p < 0.001), and opportunity (F = 44.50; p < 0.001) mindsets (Fig. 2). Tukey’s test of multiple comparisons revealed that much of this variance was driven by patients with chronic pain and cancer. Patients in the chronic pain condition reported significantly greater agreement with the catastrophe mindset and significantly lower agreement with the manageable and opportunity mindsets compared with all other illness conditions. Patients in the cancer condition reported significantly different levels of agreement with the catastrophe and opportunity mindsets compared with the other illness conditions but differed only from the chronic pain condition in their agreement with the manageable mindset. Full details on the comparisons across disease status and illness condition are reported in the supplement.
Fig. 2
Illness mindsets across disease status in the entire sample (N = 401) are shown in panel (a). Illness mindsets across diagnosis type within the chronic illness subgroup (N = 200) are shown in panel (b). Results of significance testing in panel a (t-test) and panel b (ANOVA) indicated with asterisks such that *p < 0.05 **p < 0.01 ***p < 0.001
Concurrently, the catastrophe mindset was significantly negatively correlated with physical functioning (r = − .51) and social functioning (r = − .64), and positively correlated with emotional distress (r = .49). The manageable and opportunity mindsets exhibited the opposite pattern (r = − .18–.47). See Table 2 for additional details.
Incremental validity
Relative value of illness mindsets in accounting for variance in functioning
Across three separate stepwise regression models, we found that mindsets were associated with significantly more variance in physical functioning (Table 3), social functioning (Table 3), and emotional distress (Table 3) than disease status alone. Of the three mindsets, the catastrophe mindset explained the greatest amount of variance in physical functioning (∆R2 = 0.087***), social functioning (∆R2 = 0.096***), and emotional distress (∆R2 = 0.052***), beyond disease status. In total, the IMI accounted for 11.44% more variance in physical functioning, 12.11% more variance in social functioning, and 5.7% more variance in emotional distress than disease status.
Discussion
This study offers the first test of the reliability and validity of the IMI in a sample of N = 401 healthy and chronically ill participants. The results provide preliminary answers to our four questions. First, regarding our initial question of internal validity, the results of Study 1 suggest the scale has a three-factor structure and that the internal reliability within each factor is high. These results suggest that the catastrophe and manageable mindsets are not mere opposites, and that the opportunity and manageable mindsets are not the same thing.
Regarding convergent/divergent validity, as expected, the mindsets were significantly correlated with some of the specific items on the IPQ-B. The largest correlations were between the catastrophe mindset and the IPQ-B consequences item (How much does your illness affect your life?), the identity item (How much do you experience symptoms from your illness?), and the emotional response item (How much does your illness affect you emotionally?). Correlations between the mindsets and other items of the BIPQ were smaller in magnitude.
With respect to affect and coping, people who endorsed the catastrophe mindset experienced more negative emotions and fewer positive emotions, and generally were more disengaged from coping with their emotions or problems. People with the manageable mindset and the opportunity mindset experienced more positive patterns of affect and coping in general, with a few notable differences: whereas the manageable mindset was primarily associated with problem-focused coping, the opportunity mindset was associated with both problem-focused coping and emotion-focused coping, and was uncorrelated with negative emotions. This suggests that having a cancer-is-an-opportunity mindset does not necessarily mean one avoids negative feelings or emotions. Rather, holding the mindset that cancer is an opportunity means that positive effects can arise as a result of going through or even embracing difficult times. This finding is similar to what is seen in the stress mindset research.
On average, the correlations between illness perceptions, affect, and coping strategies were weaker in magnitude than were the correlations between mindsets, affect, and coping strategies. Taken together, these correlations support our conceptualization that—while mindsets might relate to perceptions, affect, and coping in response to an illness—they are not redundant constructs, either statistically or conceptually. While the IPQ-B remains a widely used and well-validated tool for understanding individuals’ perceptions of a specific illness, the IMI offers a complementary perspective. The IPQ-B focuses on individual, illness-specific appraisals across dimensions like timeline, controllability, and emotional response. By contrast, the IMI captures broader beliefs about the meaning of illness as a life phenomenon. For example, while the IPQ-B asks, “How much does your illness affect your life?” the IMI asks participants to agree or disagree with statements like “Chronic illness can be an opportunity to grow”. The IMI’s brief format and general phrasing may make it particularly useful in early stages of illness, in preventive care, or across diagnostic groups where standard condition-specific tools may not yet be applicable.
Regarding whether mindsets differ by disease status and subtype, we found that only the opportunity mindset differed significantly between those with and those without a chronic illness. Interestingly, the directionality of this difference suggests that people with chronic illnesses may have more of an opportunity mindset about their disease than healthy individuals do. Within the chronic illness subgroup, endorsement of the three illness mindsets differed across illness type. These differences were driven by individuals with chronic pain, who indicated comparatively greater endorsement of the catastrophe mindset and less endorsement of the manageable and opportunity mindsets. These findings suggest that there is some degree of illness-specific variation in these mindsets. However, in general, illness mindsets do not appear to be contingent on health status. In other words, people can have mindsets about illnesses with which they have not been diagnosed. These mindsets are likely shaped by a variety of social and cultural factors (e.g., experience with close others with the disease, society’s portrayal of the illness).
Finally, regarding our questions about concurrent and incremental validity, we found that mindsets were strongly correlated with physical, social, and emotional functioning, and provided added incremental validity in predicting functioning over and above and beyond disease status. This suggests that while having a chronic illness can certainly have a negative impact on multiple facets of functioning, the mindset an individual adopts statistically accounts for a significant amount of additional variance in these critical outcomes. This was especially true for the catastrophe mindset, which had a particularly negative relationship with all three domains of functioning. The direction of causality remains to be established.
One limitation of the current analysis is that while we assessed illness perceptions and compared them to illness mindsets at the correlational level, we did not include illness perceptions as covariates in our regression models. This decision reflects our theoretical view that Illness Mindsets operate at a broader, more abstract and upstream level, therefore shaping more specific interpretations and appraisals. Controlling for illness perceptions would statistically remove variance that is conceptually downstream of mindsets, potentially obscuring their explanatory role. Future research should examine how interventions targeting illness mindsets might lead to adaptive changes in downstream perceptions and appraisals, and whether shifting mindsets offers a more efficient or scalable approach to improving patient outcomes compared with targeting these downstream appraisals.
Study 2
Study 1 explored the distribution of illness mindsets across disease status and illness type, and their relationship to illness perceptions, coping, affect, and functioning. Study 2 sought to explore the relationship of the IMI with physical, social, and emotional functioning in a sample of N = 463 cancer patients and survivors. Building on our findings from Study 1, we first hypothesized that there would be no significant differences in the distribution of mindsets across cancer type, status, or stage. Second, we hypothesized that mindsets would be associated with physical, social, and emotional functioning over and above these indices. Finally, Study 2 also sought to further confirm the discriminant validity of the IMI, in terms of its distinctness from more trait-like measures of optimism and self-efficacy.
Methods
This study was reviewed and approved by the Stanford University Institutional Review Board and Administrative Panel for the Protection of Human Subjects. Detailed hypotheses and analyses were agreed upon in a signed and dated collaborative agreement between the Stanford Mind & Body Lab and the Cancer Support Community before accessing data and were registered on OSF [Link in Title Page].
Participants and procedure
Cancer patients with and without current evidence of disease (EoD) were recruited through the Cancer Support Community's nationwide registry to complete an online survey about their experiences, health, and functioning as part of a larger survey examining cancer patient experiences and perspectives on cancer. Complete data were collected from N = 463 participants. The sample was, on average, 59.93 (SD = 10.55) years old, 79.3% female, and 91.9% White, and 60.8% had at least a four-year college degree.
Participants were eligible if they had a self-reported history of a cancer diagnosis. No exclusion criteria were applied based on time since diagnosis, treatment type, or treatment status. As a result, the sample included individuals who were in active treatment, had recently completed treatment, or were long-term survivors. This approach allowed us to examine illness mindsets across a broad spectrum of experiences with cancer.
A diverse range of cancer types were represented in the sample. The most prevalent diagnostic categories were breast cancer (37.1%), cancers of the blood (12.1%), lung cancer (9.3%), prostate cancer (9.3%), gynecological cancers (5.8%), and lymphoma (5.6%). Most participants had been diagnosed with localized (stage 0-III) disease (64.7%), compared with metastatic (stage IV) cancer (17.7%). The remaining 17.6% of participants were either unsure of their cancer stage or had a form of cancer that did not use the TMN staging nomenclature. Participants with an initial diagnosis (16.6%) or recurrence/relapse (21.7%) were categorized as demonstrating evidence of disease (EoD; 38.3%). Participants who had been diagnosed with cancer in the past but currently showed no indication of active disease were categorized as having no evidence of disease (NED; 61.6%).
Measures
The survey included measures of clinical history, illness mindsets, trait optimism, self-efficacy, and physical, social, and emotional functioning.
To ensure that the mindsets were not simply proxies for optimism or self-efficacy, measures of these traits were administered. Trait optimism was measured by the 10-item Revised Life Orientation Test (LOT-R), which assesses individual differences in generalized optimism versus pessimism (Scheier et al., 1994). Individual items are rated on a 5-point Likert scale ranging from 0 (strongly disagree) to 4 (strongly agree). A total score was calculated by summing the 10 items.
Self-efficacy was measured by the short version of the General Self Efficacy Scale (GSE), a six-item measure that assesses beliefs about one’s ability to cope with a variety of difficult demands in life (Romppel et al., 2013). Items are rated on a scale ranging from 1 (not at all true) to 4 (exactly true). A total score was calculated by summing the six items.
Further, to help establish that the mindsets were not simply an indication of how well people were functioning in the moment, we measured three facets of functioning: physical, social, and emotional. As with Study 1, physical and social functioning were measured using corresponding subscales of the PROMIS-29 (Hays et al., 2018). Each subscale consisted of four items rated on a five-point Likert scale. Summary scores were calculated by taking the average score of the four items. Emotional distress was assessed using the emotional wellbeing subscale from CancerSupportSource,® a validated, multidimensional distress screening tool for cancer patients (Buzaglo et al., 2020). Items were rated on a five-point Likert scale ranging from 0 to 5.
Analytic approach
To further explore convergent and discriminant validity (the relationship of IMI with measures of self-efficacy and optimism) and concurrent validity (the relationship of IMI with functioning), we conducted a series of correlational analyses. To test the hypothesis that there would be no significant differences in the distribution of mindsets across cancer type, status, or stage, we conducted a multivariate aof nnalysis of covariance (MANCOVA) that included three categorical independent variables (cancer type, stage at diagnosis, and current cancer status) and three continuous dependent variables (illness catastrophe mindset, illness manageable mindset, and illness opportunity mindset). Age, race, gender, and educational attainment were included as demographic covariates in all models. Finally, to test incremental validity, we conducted three separate stepwise linear regression models with physical, social, and emotional functioning as the dependent variables of interest, respectively. Each model included the same independent variables added in the same series of steps. In step 1, stage at diagnosis (localized vs. metastatic) and current cancer status (EoD vs. NED) were included as measures of illness severity and presence. In step 2, self-efficacy and optimism were included as individual differences of importance. In steps 3a-c, we added the catastrophe, manageable, and opportunity mindsets, respectively. In step 3d, we included all three illness mindsets to report the total incremental validity of the nine-item IMI, above and beyond clinical and psychological variables.
Results
Descriptive statistics and internal consistency
Descriptive statistics, including mean scores, standard deviations, Cronbach’s Alphas, and Pearson’s correlations for mindsets, individual differences, and functioning variables are included in Table 4. All three mindsets demonstrated strong internal consistency, ranging from α = 0.77 − 0.85.
Table 4
Correlations between mindsets, psychological variables, and wellbeing in Study 2
Variables
M
SD
1
2
3
4
5
6
7
Illness mindsets
1. Catastrophe
3.69
1.04
(0.77)
2. Manageable
4.27
0.85
− .44**
(0.81)
3. Opportunity
4.19
0.99
− .22**
.55**
(0.79)
Psychological variables
4. Self-Efficacy
17.57
4.26
− .24**
.36**
.21**
(0.89)
5. Optimism
15.40
6.17
− .34**
.41**
.29**
.42**
(0.86)
Functioning
6. Physical Functioning
3.58
1.08
− .23**
.23**
.07
.26**
.20**
(0.91)
7. Social Functioning
4.17
0.95
− .32**
.32**
.13**
.32**
.31**
.75**
(0.95)
8. Emotional Distress
1.48
0.95
.44**
− .43**
− .21**
− .35**
− .51**
− .33**
− .41** (0.89)
*p < .05 **p < .01. M and SD are used to represent mean and standard deviation, respectively. Values in parentheses on the diagonal are Cronbach’s Alpha, a measure of internal consistency for non-single item subscales. Illness Mindsets measured with the Illness Mindset Inventory (IMI); Self-efficacy was measured by the short version of the General Self Efficacy Scale (GSE); Trait optimism was measured by the 10-item Revised Life Orientation Test (LOT-R); physical and social functioning were measured with the corresponding subscales of the PROMIS-29; Emotional distress was assessed using the emotional wellbeing subscale from CancerSupportSource®
Discriminant validity
The catastrophe mindset was negatively correlated with self-efficacy (r = − .24) and optimism (r = − .34), whereas the manageable and opportunity mindsets showed the opposite pattern of correlations (rs = .21–.41) (Table 4). These patterns suggest, as theorized, that mindsets are related to broad personality traits but not redundant with them. In other words, optimists may be slightly less likely to view illness as a catastrophe and more likely to view it as an opportunity; however, such mindsets are not inevitable and therefore are potentially malleable, when desirable, despite personality dispositions.
Illness mindsets across clinical characteristics
Perhaps surprisingly, though as hypothesized, we found no significant differences in illness mindsets across levels of clinical stage at diagnosis, current cancer status, or cancer type (Fig. 3). The complete results are reported in the Supplement.
Fig. 3
Illness mindsets across a cancer stage, b status of disease, and c diagnostic category in study 2
Illness mindsets were significantly correlated with physical, social, and emotional functioning (Table 4), and explained significantly more variance in physical functioning (∆R2 = 0.114***), social functioning (∆R2 = 0.21***), and emotional distress (∆R2 = 0.057***) than models that included clinical health variables (e.g., cancer stage, cancer status) and psychological variables (e.g., optimism, self-efficacy) alone, controlling for demographic variables. These effects were driven by the catastrophe and manageable mindsets but not the opportunity mindset. Results for these three models are presented in Table 5.
Table 5
Study 2—Additional variance in physical functioning explained by illness mindsets
Step 1
Step 2
Step 3a
Step 3 b
Step 3c
Step 3d
Physical functioning
Intercept
2.75** (0.98–4.52)
2.73** (0.99–4.47)
3.69** (1.95–5.43)
1.98*(0.23–3.73)
2.48** (0.71–4.26)
3.05** (1.20–4.89)
Cancer Stage
− 0.22 (− 0.47–0.03)
− 0.18 (− 0.44–0.07)
− 0.15 (− 0.39–0.10)
− 0.20 (− 0.45–0.04)
− 0.19 (− 0.44–0.06)
− 0.17(− 0.41–0.08)
Metastatic
Cancer Status: EoD
− 0.21 (− 0.43–0.00)
− 0.19 (− 0.40–0.02)
− 0.22** (− 0.42–0.01)
− 0.18 (− 0.38–0.03)
− 0.19 (− 0.40–0.02)
− 0.20 (− 0.40–0.01)
Optimism
0.00 (− 0.02–0.02)
− 0.01 (− 0.02–0.01)
− 0.00 (− 0.02–0.02)
0.00 (− 0.02–0.02)
− 0.01 (− 0.03–0.01)
Self-Efficacy
0.04*** (0.02–0.07)
0.03* (0.01–0.06)
0.03* (0.00–0.06)
0.04*** (0.01–0.07)
0.03 (0.00–0.06)
Illness Catastrophe
− 0.23*** (− 0.33–0.13)
− 0.17** (− 0.28–0.07)
Illness Manageable
0.24*** (0.11–0.36)
0.17** (0.01–0.33)
Illness Opportunity
0.07 (− 0.03–0.18)
− 0.03 (− 0.16–0.09)
Observations
312
312
312
312
312
312
R2
0.07
0.10
0.16
0.14
0.11
0.17
A R2 (From Step 2)
0.055***
0.040***
0.005
0.069***
Social functioning
Intercept
1.50 (− 0.58–3.58)
1.43 (− 0.59–3.46)
2.92*** (0.95–4.90)
0.25 (− 1.74–2.25)
1.06(− 1.00–3.12)
1.91 (− 0.16–3.98)
Cancer Stage
− 0.37* (− 0.66–0.07)
− 0.31* (− 0.60–0.02)
− 0.26 (− 0.53–0.02)
− 0.34* (− 0.62–0.06)
− 0.33* (− 0.62–0.04)
− 0.28* (− 0.56–0.01)
Metastatic
Cancer Status: EoD
− 0.32* (− 0.57–0.06)
− 0.29* (− 0.53–0.04)
− 0.32** (− 0.56–0.09)
− 0.26* (− 0.50–0.02)
− 0.29* (− 0.54–0.04)
− 0.29* (− 0.53–0.06)
Optimism
0.01 (− 0.01–0.03)
− 0.00 (− 0.03–0.02)
0.00 (− 0.02–0.02)
0.01 (− 0.02–0.03)
− 0.01 (− 0.03–0.01)
Self-Efficacy
0.06*** (0.03–0.09)
0.04** (0.01–0.07)
0.04* (0.00–0.07)
0.06** (0.02–0.09)
0.03* (0.00–0.06)
Illness Catastrophe
− 0.35*** (− 0.46–0.24)
− 0.27*** (− 0.39–0.15)
Illness Manageable
0.37*** (0.23–0.52)
0.28** (0.10–0.46)
Illness Opportunity
0.11 (− 0.01–0.23)
− 0.06 (− 0.20–0.08)
Observations
311
311
311
311
311
311
R2
0.08
0.14
0.23
0.21
0.15
0.26
A R2 (From Step 2)
0.095***
0.070***
0.009
0.120***
Emotional distress
Intercept
2.50** (0.70–4.30)
2.82*** (1.20–4.44)
1.52 (− 0.04–3.08)
3.84*** (2.25–5.43)
3.22*** (1.57–4.86)
2.42** (1.78–4.05)
Cancer Stage
0.16 (− 0.10–0.42)
0.09 (− 0.15–0.32)
0.04 (− 0.18–0.26)
0.11 (− 0.11–0.34)
0.10 (− 0.13–0.33)
0.06 (− 0.15–0.28)
Metastatic
Cancer Status: EoD
0.12 (− 0.10–0.34)
0.08 (− 0.11–0.28)
0.12 (− 0.06–0.30)
0.07 (− 0.12–0.25)
0.09 (− 0.11–0.29)
0.10 (− 0.08–0.28)
Optimism
− 0.05*** (− 0.07–0.04)
− 0.04*** (0.06–0.02)
− 0.05*** (− 0.06–0.03)
− 0.05*** (0.07–0.03)
− 0.04*** (− 0.05–0.02)
Self-Efficacy
− 0.04*** (− 0.07–0.02)
− 0.03* (− 0.05–0.00)
− 0.02 (− 0.05–0.00)
− 0.04** (− 0.06–0.01)
− 0.02 (− 0.04–0.01)
Illness Catastrophe
0.30—(0.22 − 0.39)
0.30*** (0.22–0.39)
0.24** (0.14–0.33)
Illness Manageable
− 0.32**** (− 0.44–0.21)
− 0.22** (0.36–0.07)
Illness Opportunity
− 0.11* (− 0.21–0.02)
0.02 (− 0.09–0.13)
Observations
312
312
312
312
312
312
R2
0.10
0.28
0.37
0.35
0.29
0.39
A R2 (From Step 2)
0.094***
0.068***
0.013*
0.117***
Asterisks indicate significance such that *p < 0.05;**p < 0.01; ***p < 0.001. Optimism was measured using the Life Orientation Test-Revised (LOT-R), Self-Efficacy was measured using the short version of the General Self Efficacy Scale (GSE). Emotional distress was assessed using the emotional wellbeing subscale from CancerSupportSource screening tool. All models controlled for age, race, gender, and education
Discussion
Illness mindsets demonstrated strong internal consistency again in Study 2, lending additional support for the IMI as a reliable measure. As hypothesized, mindsets did not differ across clinically diagnosed variables of diagnosis type, stage at diagnosis, and current cancer status. This means that patients who have more severe or active cancer do not necessarily have more negative mindsets.
Also as hypothesized, the catastrophe and manageable mindsets were significantly associated with physical, social, and emotional functioning. This held true even when controlling for severity (i.e., cancer stage and status) and broader psychological traits (i.e., optimism and self-efficacy). These findings suggest that while a cancer diagnosis is associated with significant distress in many or most patients (Carlson et al., 2004; Zabora et al., 2001), mindsets about cancer were associated with differences in physical, social, and emotional functioning, suggesting that illness mindsets may be an important psychological factor to consider alongside clinical variables. Contrary to our hypotheses, the opportunity mindset was a weaker or, in some cases, not a significant predictor of functioning in the hierarchical regressions. While the manageable and catastrophe mindsets seem to account for most of the variability in the outcomes we measured here, future research could explore whether opportunity mindsets may prove to be a valuable measure in accounting for or engendering variance in a broader range of outcome measures, such as meaning in life or life satisfaction.
General Discussion
This paper introduces the concept of illness mindsets—people’s core assumptions about the nature and meaning of an illness—and provides initial data suggesting that illness mindsets can be reliably measured in both healthy and chronically ill populations using the nine-item IMI. Extensive pilot-testing supported the three-factor structure of the scale (measuring catastrophe, manageable, and opportunity mindsets, respectively), and subsequent studies demonstrated strong internal consistency of each of these three factors. Study 1 found that illness mindsets were associated with between 5.7 and 12.1% additional variance in physical, social, and emotional functioning, above and beyond disease status, controlling for demographic variables. In Study 2, illness mindsets statistically accounted for between 6.9 and 12.0% additional variance in physical functioning, social functioning, and emotional distress, above and beyond cancer stage, cancer status, trait optimism, and self-efficacy, controlling for demographic variables. Illness mindsets were correlated with—but not redundant with—more situationally specific illness perceptions and more general measures of trait optimism and self-efficacy.
Of note, the correlational nature of the present studies limits our ability to infer direction of causality. Our results provide robust evidence of an association between illness mindsets and important patient outcomes, the cross-sectional nature of our data prevents us from drawing definitive conclusions about causality. It is possible that illness mindsets influence functioning, as theorized, but it is also possible that one’s current level of functioning informs or reinforces their mindset. To clarify this directionality, future research should use longitudinal designs and experimental methods to manipulate mindsets and examine subsequent changes in functioning over time. Establishing this causal link is essential for confirming the IMI’s value as a target for clinical intervention.
Implications for use as a patient-reported measure in clinical care
Regardless of the direction of causal influence, taken together, these initial studies provide preliminary evidence suggesting that illness mindsets, as measured by the brief nine-item IMI, may be a clinically useful patient-reported measure that can help account for variance in individual functioning, over and above clinically diagnosed disease status. The IMI can be an important addition to the accumulating repertoire of PROs, for several reasons. First, illness mindsets are not simply rebranding of existing constructs; we propose that they operate at a level of conceptual abstraction that has unique clinical utility and can easily measure an aspect of patient experience that is not currently captured by existing PROMs and psychological measures. Second, illness mindsets are efficient: They can be measured briefly and reliably with just nine questions, thus placing little additional burden on the patient. By contrast, many existing PROMs require extensive surveying and burden on the part of patients, with measures consisting of up to 128 questions (Maharaj et al., 2020). Third, illness mindsets are precise: They capture a clearly defined construct that can get at the heart of patients’ orientations toward their illness.
In summary, the use of the IMI in clinical care may be a relatively cost- and time-efficient way to assist providers in understanding their patients, which in turn may promote better communication and shared decision making. The IMI may also help providers understand a potential source of decreases (or increases) in functioning and help them monitor the effects of disease progression (Chen et al., 2013). While we provide initial theoretical justification for illness mindsets as a construct, we also acknowledge their conceptual proximity to existing appraisal-based and illness perception frameworks. Future work should aim to systematically compare illness mindsets with stress appraisals, coping profiles, and illness perceptions to determine the relative benefit of each in predicting important outcomes and as potential targets for change.
To fully tap the IMI’s utility in research and clinical care, the following three questions are critical to explore in future research.
First, where do illness mindsets come from? As discussed in the introduction, mindsets are conceptually distinct from more specific illness perceptions because they are intended to reflect core beliefs about the nature of an illness. Indeed, the current research provides preliminary support for the notion that illness mindsets are not merely a direct reaction to an individual’s clinically diagnosed disease status or diagnosed severity. In other words, being diagnosed with a more severe illness does not necessarily predict that the individual is more likely to hold the mindset that illness is more catastrophic, less manageable, or less of an opportunity. Across two studies, we found only negligible differences across disease status.
But if mindsets are not simply a reaction to an individual’s current experience with a disease, where do they come from? Although mindsets are measured at the individual level, they are informed by cultural and social factors, such as how society portrays illness in the media (movies, books, social media), how providers or health-care entities describe diagnoses, and one’s experiences with others who are ill (Crum & Zuckerman, 2017; Markus & Kitayama, 2010). Future work should aim to explore these sources of mindsets, being careful not to blame individuals for the mindsets they currently hold.
Second, how do mindset effects unfold over time? A particularly important task of future work is to use these measures to improve understanding of how and why mindsets may be related to an individual’s functioning. Past research has shown that mindsets can influence health outcomes through affective, behavioral, and physiological mechanisms.(Crum et al., 2011, 2013) For example, mindsets about illness could influence levels of interleukin-6, a pro-inflammatory cytokine that previously has been shown to respond to psychological changes in affect (Stellar et al., 2015), mental health (Soygur et al., 2007), and social functioning (Creswell et al., 2012). It is also possible that mindsets shape motivation to engage in health behaviors like adherence to medication and diet/exercise guidelines, which can improve functioning over time. Future research should explore how mindsets may influence downstream functioning through these behavioral and potentially physiological pathways.
Third, can illness mindsets be changed? Perhaps the most important question for future work is to determine if mindsets can be shaped—that is, offered in a way that helps patients or potential patients understand mindsets’ utility. The greatest strength of mindsets may be their potential for intervention—whether before, at, or after the diagnosis of an illness, past research suggests. Because people may be particularly convinced of and invested in specific illness perceptions (such as the degree to which they are experiencing symptoms or how much the illness is affecting them emotionally), countering these specific perceptions may be ineffective. By contrast, people may be more open to information that leads them to modify their mindsets. Guiding a patient toward a more adaptive mindset may simply help them view their illness from a different perspective without minimizing their concerns. Mindsets are also a more specific and tangible lever for change than sweeping personality traits, such as optimism, that span multiple domains of life, are generally stable over time, and are more difficult to intervene upon. Brief, highly targeted interventions have been found to shape mindsets in numerous domains and often elicit meaningful and sustained effects (See Walton & Crum, 2020, for a review) and some work has already shown that illness mindsets can be changed in ways that inspire improvements in health related quality of live (Zion et al., 2023). Future work is needed to explore and develop interventions that efficiently, ethically, and effectively change illness mindsets in the service of improving patient health.
Conclusion
The diagnosis of a chronic illness is a life-altering experience that can negatively affect a person’s social, emotional, and physical functioning. While chronic illnesses carry objective physiological signatures that define their presence and severity (e.g., elevated glucose for diabetes, presence of malignant cells for cancer, heightened blood pressure for hypertension), the negative repercussions on functioning are not entirely dictated by these biological signatures. An individual’s functioning may also be profoundly shaped by psychological factors. This series of studies introduces illness mindsets as a novel psychological variable that may help us understand variability in physical, social, and emotional functioning above and beyond the presence and severity of disease. In so doing, this research lays the groundwork for much future research aimed at further understanding the nuanced ramifications of these mindsets and, potentially, effectively leveraging mindsets to improve the experience of individuals coping with illness and disease.
Declarations
Conflict of interest
Dr. Alexandra K. Zaleta has received institutional research funding provided to the Cancer Support Community (prior employer) for research where author was PI, unrelated to this project from Astellas Pharma, Boston Scientific Foundation, Gilead Sciences, Novartis, and Seagen; and institutional advisory funding provided to Cancer Support Community (prior employer) for author participation in advisory boards unrelated to this project from Beigene and Bristol Myers Squibb. Other others have no competing interests to declare that are relevant to the content of this article.
Ethical approval
This study was reviewed and approved by the Stanford University Institutional Review Board and Administrative Panel for the Protection of Human Subjects.
Consent to participate
All participants provided their consent to participate this research and signed an IRB approved informed consent form.
Consent for publication
Not Applicable.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Addison, C. C., Campbell-Jenkins, B. W., Sarpong, D. F., et al. (2007). Psychometric evaluation of a coping strategies inventory short-form (CSI-SF) in the Jackson Heart Study cohort. International Journal of Environmental Research and Public Health,4, 289–295.CrossRefPubMedPubMedCentral
Baum, A., & Posluszny, D. M. (1999). Health psychology: Mapping biobehavioral contributions to health and illness. Annual Review of Psychology,50, 137–163.CrossRefPubMed
Blackwell, L. S., Trzesniewski, K. H., & Dweck, C. S. (2007). Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and an intervention. Child Development,78, 246–263.CrossRefPubMed
Boles, D. Z., DeSousa, M., Turnwald, B. P., et al. (2021). Can exercising and eating healthy be fun and indulgent instead of boring and depriving? Targeting mindsets about the process of engaging in healthy behaviors. Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2021.745950CrossRefPubMedPubMedCentral
Broadbent, E., Petrie, K. J., Main, J., & Weinman, J. (2006). The brief illness perception questionnaire. Journal of Psychosomatic Research,60, 631–637.CrossRefPubMed
Buzaglo, J. S., Zaleta, A. K., McManus, S., Golant, M., & Miller, M. F. (2020). CancerSupportSource®: Validation of a revised multi-dimensional distress screening program for cancer patients and survivors. Supportive Care in Cancer,28, 55–64.CrossRefPubMed
Cella, D., Riley, W., Stone, A., et al. (2010). The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008. Journal of Clinical Epidemiology,63, 1179–1194.CrossRefPubMedPubMedCentral
Chen, J., Ou, L., & Hollis, S. J. (2013). A systematic review of the impact of routine collection of patient reported outcome measures on patients, providers and health organisations in an oncologic setting. BMC Health Services Research,13, 1–24.CrossRef
Crawford, J. R., & Henry, J. D. (2004). The positive and negative affect schedule (PANAS): Construct validity, measurement properties and normative data in a large non-clinical sample. British Journal of Clinical Psychology,43, 245–265.CrossRefPubMed
Creswell, J. D., Irwin, M. R., Burklund, L. J., et al. (2012). Mindfulness-based stress reduction training reduces loneliness and pro-inflammatory gene expression in older adults: A small randomized controlled trial. Brain, Behavior, and Immunity,26, 1095–1101.CrossRefPubMedPubMedCentral
Crum, A. J., Akinola, M., Martin, A., & Fath, S. (2017). The role of stress mindset in shaping cognitive, emotional, and physiological responses to challenging and threatening stress. Anxiety Stress and Coping,30(4), 379–395.CrossRefPubMed
Crum, A. J., Corbin, W. R., Brownell, K. D., & Salovey, P. (2011). Mind over milkshakes: Mindsets, not just nutrients, determine ghrelin response. Health Psychology,30, 424.CrossRefPubMed
Crum, A. J., Jamieson, J. P., & Akinola, M. (2020). Optimizing stress: An integrated intervention for regulating stress responses. Emotion,20, 120–125.CrossRefPubMedPubMedCentral
Crum, A. J., Salovey, P., & Achor, S. (2013). Rethinking stress: The role of mindsets in determining the stress response. Journal of Personality and Social Psychology,104, 716–733.CrossRefPubMed
Dweck, C. S. (2019). The choice to make a difference. Perspectives on Psychological Science,14, 21–25.CrossRefPubMed
Giltay, E. J., Kamphuis, M. H., Kalmijn, S., Zitman, F. G., & Kromhout, D. (2006). Dispositional optimism and the risk of cardiovascular death: The Zutphen Elderly Study. Archives of Internal Medicine,166, 431–436.PubMed
Hays, R. D., Spritzer, K. L., Schalet, B. D., & Cella, D. (2018). PROMIS-29 v2.0 profile physical and mental health summary scores. Quality of Life Research,27, 1885–1891.CrossRefPubMedPubMedCentral
Jamieson, J. P., Crum, A. J., Goyer, J. P., Marotta, M. E., & Akinola, M. (2018). Optimizing stress responses with reappraisal and mindset interventions: an integrated model. Anxiety Stress and Coping,31(3), 245–261.CrossRefPubMed
Lau, R. R., & Hartman, K. A. (1983). Common sense representations of common illnesses. Health Psychology,2, 167–185.CrossRef
Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common sense representation of illness danger. Contrib to Med Psychol,2, 7–30.
Maharaj, A. D., Samoborec, S., Evans, S. M., et al. (2020). Patient-reported outcome measures (PROMs) in pancreatic cancer: A systematic review. HPB,22, 187–203.CrossRefPubMed
Markus, H. R., & Kitayama, S. (2010). Cultures and selves. Perspectives on Psychological Science,5, 420–430.CrossRefPubMed
Miller, G., Chen, E., & Cole, S. W. (2009). Health psychology: Developing biologically plausible models linking the social world and physical health. Annual Review of Psychology,60, 501–524.CrossRefPubMed
Paunesku, D., Walton, G. M., Romero, C., Smith, E. N., Yeager, D. S., & Dweck, C. S. (2015). Mind-set interventions are a scalable treatment for academic underachievement. Psychological Science,26, 784–793.CrossRefPubMed
Petrie, K. J., Jago, L. A., & Devcich, D. A. (2007). The role of illness perceptions in patients with medical conditions. Current Opinion in Psychiatry,20, 163–167.CrossRefPubMed
R Core Team: R (2020) A Language and Environment for Statistical Computing.
Romppel, M., Herrmann-Lingen, C., Wachter, R., et al. (2013). A short form of the general self-efficacy scale (GSE-6): Development, psychometric properties and validity in an intercultural non-clinical sample and a sample of patients at risk for heart failure. GMS Psycho-Social-Medicine
Rosseel, Y. (2012). lavaan}: An {R package for structural equation modeling. Journal of Statistical Software,48, Article 1–36.CrossRef
Scheier, M.F., Carver, C.S., & Bridges, M.W. (1994). Life Orientation Test – Revised (LOT-R). J Pers Soc Psychol.
Soygur, H., Palaoglu, O., Akarsu, E. S., et al. (2007). Interleukin-6 levels and HPA axis activation in breast cancer patients with major depressive disorder. Progress in Neuro-Psychopharmacology & Biological Psychiatry,31, 1242–1247.CrossRef
Stanton, A. L., Revenson, T. A., & Tennen, H. (2007). Health psychology: Psychological adjustment to chronic disease. Annual Review of Psychology,58, 565–592.CrossRefPubMed
Stellar, J. E., John-Henderson, N., Anderson, C. L., Gordon, A. M., McNeil, G. D., & Keltner, D. (2015). Positive affect and markers of inflammation: Discrete positive emotions predict lower levels of inflammatory cytokines. Emotion,15, 129–133.CrossRefPubMed
Sullivan, M., Bishop, S. R., & Pivik, J. (1995). The pain catastrophizing scale: Development and validation. Psychological Assessment,7, 524.CrossRef
Tedeschi, R. G., & Calhoun, L. G. (1996). The posttraumatic growth inventory: Measuring the positive legacy of trauma. Journal of Traumatic Stress,9, 455–471.CrossRefPubMed
Walton, G. M. (2014). The new science of wise psychological interventions. Current Directions in Psychological Science,23, 73–82.CrossRef
Walton, G. M., & Crum, A. J. (2021). Handbook of wise interventions. Guilford Publications.
Walton, G. M., & Wilson, T. D. (2018). Wise interventions: Psychological remedies for social and personal problems. Psychological Review,125, 617–655.CrossRefPubMed
Zabora, J., BrintzenhofeSzoc, K., Curbow, B., Hooker, C., & Piantadosi, S. (2001). The prevalence of psychological distress by cancer site. Psycho-Oncology,10, 19–28.CrossRefPubMed
Zion, S. R., Schapira, L., Berek, J. S., Spiegel, D., Dweck, C. S., & Crum, A. J. (2023). Changing cancer mindsets: A randomized controlled feasability and efficacy trial. Psychooncology, 32(9), 1433–1332. https://doi.org/10.1002/pon.6194.
