Abstract
Background and Purpose:
Illness uncertainty is prevalent in patients with chronic life-limiting conditions. We described high levels of illness uncertainty in patients awaiting liver transplant and examined relationships between uncertainty and person factors and the antecedents of uncertainty,
Methods:
The Mishel Uncertainty in Illness scale was used to measure illness uncertainty. We used modes and interquartile range (IQR) to describe illness uncertainty levels in 115 patients awaiting liver transplant. Multiple logistic and linear regression models estimated the associations of uncertainty with hypothesized antecedents.
Results:
High total illness uncertainty score was reported by 15.6% of the patients. After adjusting for all variables, illness uncertainty was associated with two antecedents of uncertainty, low social well-being (OR=0.816; p=.025) and low self-efficacy (OR=0.931; p=.013). Complexity was negatively associated with social well-being; ambiguity and inconsistency were negatively associated with self-efficacy.
Conclusions:
One in seven patients experienced high illness uncertainty. Social well-being and self-efficacy were negatively related to illness uncertainty.
Keywords: illness uncertainty, end-stage liver disease, transplants, chronic disease
Illness uncertainty is the inability of individuals to determine the meaning of symptoms or events related to an illness, or to accurately predict outcomes (Mishel, 1988). Uncertainty in chronic illness comes from the unpredictability of symptoms, continual concerns about disease exacerbation, and an unknown future due to physically limiting problems. Illness uncertainty consists of four subdomains: (1) Ambiguity refers to unclear or ever-changing bodily cues about the state of the illness that may be confused with other illness concerns; (2) Complexity is defined as difficulty understanding one’s treatments or the healthcare system; (3) Inconsistency is characterized by frequently changing or inconsistent messages from the healthcare provider; and (4) Unpredictability refers to the incongruence between the present and previous illness experience (Mishel, 1997).
Illness uncertainty has been associated with depression and poorer quality of life in patients with chronic illnesses including liver disease (Bailey Jr. et al., 2010; Colagreco et al., 2014; Kimbell et al., 2015). Patients living with end-stage liver disease experience considerable illness uncertainty. For example, patients are uncertain about if or when they will receive a liver transplant. Therefore end-stage liver disease serves as an exemplar of a chronic life-limiting condition to explore the phenomenon of illness uncertainty (Bailey Jr. et al., 2017).
Illness Uncertainty and Waiting for a Transplant
In patients with end-stage liver disease waiting for liver transplant, illness uncertainty is exacerbated (Lasker et al., 2010). Patients waiting for liver transplant experience deteriorating physical conditions including fatigue, muscle weakness, decreased appetite, nausea, weight loss, and skin problems. End-stage liver disease patients can develop encephalopathy with memory loss and diminished cognitive ability. Eventually, they die unless a donor organ becomes available (Montagnese & Bajaj, 2019). The awareness that without a liver transplant they may die, making the wait even more intolerable (Larson, 2015).
The waiting period for a liver transplant is difficult, fraught with uncertainty and associated with significant morbidity and mortality (Golfieri et al., 2019; Pérez-San-Gregorio et al., 2012). Patients reported the waiting period as the most stressful time of their lives (Teixeira et al., 2016). However, little attention has been paid to the factors that influence illness uncertainty among patients waiting for liver transplant (Bailey Jr. et al., 2017).
One study (n=96) examined differences in illness uncertainty experienced by women waiting for liver transplant (n=24) compared to post-transplant women (n=72) (Lasker et al., 2010). Women on the waiting list experienced significantly higher levels of illness uncertainty and ambiguity than the post-transplant women. However, the investigators did not quantify high uncertainty levels.
Antecedents of illness uncertainty were identified from the literature relevant to the Uncertainty in Illness Theory. While most empirical studies have focused predominantly on two groups of antecedents: stimuli frame and structure providers, we considered the following antecedents as conceptually relevant to patients waiting for liver transplant. Stimuli frame included bodily function (Mishel & Murdaugh, 1987; Nelson, 1996) and severity of illness (Braden, 1990). Within the other group of antecedents: cognitive capacities, we identified control over disease and symptoms (Affleck et al., 1987), strategies to control emotions (McCain & Cella, 1995) and self-efficacy (Braden, 1991). Lastly, we considered the structure providers including social roles and support as well as education (Fleury et al., 1995; Mishel & Braden, 1988) (Figure 1).
Figure 1:
Model of illness uncertainty
Purpose
The purpose of this study was to use the Mishel Uncertainty in Illness Scale Version A (MUIS-A) to quantify a high level of illness uncertainty. The aims of this study were to: (1) describe illness uncertainty levels in a sample of patients with end-stage liver disease awaiting liver transplant and (2) explore whether and how high total illness uncertainty relates to person factors and the antecedents of uncertainty based on the Uncertainty in Illness Theory. Furthermore, as illness uncertainty consists of four sub-domains: ambiguity, complexity, inconsistency, and unpredictability, we sought to explore whether and how each illness uncertainty subdomain relates to person factors and the antecedents.
METHOD
Design
This was a descriptive study using data from the parent study, which was a randomized controlled trial testing the efficacy of an Uncertainty Self-Management Intervention (SMI) compared to Liver Disease Education (LDE), tailored for patients awaiting liver transplant (Bailey et al., 2017). Baseline data (pre-intervention) collected from 115 patients, intervention and control groups combined, were analyzed for this paper.
Sample
One hundred and fifteen patients were recruited from four US transplant centers: Private University Medical Center in the Southeast (n=52), Public University Medical Center in the Southeast (n=2), Public University Medical Center in the Northeast (n=44), and a Public University Medical Center in the Midwest (n=17). Eligible patients were: at least 18 years old, able to read and speak English, formally enrolled on a liver transplant list, had not received a prior transplant (any organ), had no significant cognitive impairment. Institutional Review Boards at all four participating sites approved the protocol. Informed consent was obtained from all patients using procedures established at each site. Participants completed questionnaires at baseline and at two and four weeks post intervention.
Measures
Illness Uncertainty.
Illness uncertainty, the primary outcome, was measured by the MUIS-A (Mishel, 1981). The MUIS-A was designed to capture the patient illness experience and has been used widely in studies involving cancer and chronic illness samples (Mishel, 1997). The total scale consists of 33 items and has a high level of internal consistency, with a Cronbach’s alpha of .91. The 33 items can be totaled or scored in 4 subscales, in which case 32 items are scored (item 15 is removed), each representing a distinct type of uncertainty (Mishel, 1997). Examples of items include “My symptoms continue to change unpredictably” and “The results of my tests are inconsistent.” Responses are indicated on a 5-point Likert scale ranging from strongly disagree to strongly agree. Scores on the MUIS-A range from 33 to 165 and are calculated as the sum of all items, reverse-coded wherever needed, so that higher scores indicate greater levels of illness uncertainty. Mishel did not define levels of uncertainty with cut-off scores. In this study, we calculated the sample mean plus one sample standard deviation of the total uncertainty scores and, therefore, chose a cut off score of 100 for identifying high levels of illness uncertainty.
Subscales of Uncertainty.
The MUIS-A scale has four subscales. (1) Ambiguity refers to unclear or ever-changing bodily cues about the state of the illness that may be confused with other illness concerns.; (2) Complexity was defined as difficulty understanding one’s treatments or the healthcare system; (3) Inconsistency was characterized by frequently changing or inconsistent messages from the healthcare provider; and (4) Unpredictability refers to the incongruence between the present and previous illness experience (Mishel, 1997). Each subscale was calculated as the average of all items so that subscales had the same scale range (1 – 5) and comparable scores.
Person factors.
Person factors included gender, age, and race, which were self-reported.
The following antecedents of uncertainty were derived from the theory.
Bodily function was measured by the subscales of Physical Well-Being and Functional Well-Being from the Functional Assessment of Cancer Therapy: General (FACT-G). The tool is a 27-item questionnaire that is valid and reliable (Cella et al., 1993). The tool measures quality of life in four domains: physical well-being, functional well-being, social/family well-being, and emotional well-being (Cella et al., 1993). Cronbach’s alpha was .92 and validity has been established (Cella et al., 1993).
Illness severity was measured by the Model for End-Stage Liver Disease (MELD) rates illness severity for adult liver transplant candidates. The range is from 6 (less ill) to 40 (gravely ill). The number was calculated using the most recent laboratory tests (Larson & Curtis, 2006)
Uncertainty Management was measured by two subscales from the Self-Control Schedule (SCS): problem solving and cognitive reframing (Rosenbaum, 1983). Problem solving was defined as the ability to identify and define concerns and generate solutions. Cognitive reframing was defined as the ability to address concerns from a positive point of view. The Cronbach’s alpha of the 11-item problem solving subscale was .91 and that of the 10-item cognitive reframing scale was .92 in a sample of men with localized prostate cancer (Mshel et al., 2002). The validity of both subscales has been established (Braden, 1990).
Self-Efficacy for Symptom Management.
Perceived level of self-efficacy was measured by a 12-item scale, modified from a standard self-efficacy measure for managing arthritis (Lorig et al., 1989). The scale uses ratings from 10 (very uncertain) to 100 (very certain). The self-efficacy score was the average of the 12 items. Cronbach’s alpha was .95; prior studies demonstrated evidence of internal consistency and construct validity (Porter et al., 2008). Prior studies used this instrument to assess self-efficacy in cancer patients (Porter et al., 2011).
Social roles and support were measured by the subscale of Social Well-Being from the FACT-G (Cella et al., 1993).
Education level was self-reported by the participants.
Analysis
Data were analyzed using Stata Version 14.2 (StataCorp LLC, College Station, TX, USA 2017). To describe illness uncertainty levels, we calculated frequencies and percentages for high vs. moderate/low total uncertainty categories; we also calculated means and standard deviations for the total uncertainty scale and all uncertainty subscales, including ambiguity, complexity, inconsistency, and unpredictability. Furthermore, we calculated mode and interquartile range (IQR) for each illness uncertainty item (5-point Likert scale), grouped by subscales.
To explore the relationships between different levels of illness uncertainty (high vs. moderate/low uncertainty categories) and demographic (age, education), illness severity (MELD score) and psychosocial characteristics (FACT-G functional well-being, physical well-being and social well-being scores, SCS problem solving and cognitive reframing scores, and self-efficacy scores), we conducted bivariate analysis for each of the possible characteristics: t-test for continuous independent variables, chi-square test for categorical independent variables, and we calculated unadjusted odds ratios for all independent variables. In order to understand how the independent variables were associated with each of the uncertainty subscales, we built a simple linear model for each of the subscales.
To calculate the adjusted odds ratios for high level of illness uncertainty, we built a multiple logistic regression model. We used the same independent variables for the model of high total uncertainty and for the model of each subscale so we could draw conclusions on how each subscale within the total uncertainty scale and patient characteristics were related. We included all of the independent variables that were significantly associated with the binary outcome (high vs. moderate or low total uncertainty level) or any of the continuous outcomes (uncertainty subscale scores) in the final model. We also built multiple regression models of the four subscales (ambiguity, unpredictability, complexity, and inconsistency) using the same independent variables included in the logistic regression model for the total uncertainty outcome.
RESULTS
Among 115 patients, internal consistency of total illness uncertainty was 0.77. As shown in Table 1, the actual range of the total illness uncertainty scores was 49 – 125, with 15.6% of patients (n=18) reporting scores higher than or equal to 100, or high illness uncertainty; 84.4% of patients (n=97) reporting scores lower than 100, i.e. middle or low illness uncertainty. For the four subscales, the mean scores for ambiguity and unpredictability were 3.06 and 3.38 respectively, which were higher than the mean scores for complexity, 2.12, and inconsistency, 2.29. The standard deviations for all four subscales were similar, ranging from 0.44 to 0.66. In Table 2, we report the mode and inter-quartile range (IQR) for each of the illness uncertainty items. The modes for all items were either “agree” or “disagree” with the respective illness-related statements, with no mode being “undecided”, indicating patients took a position.
Table 1.
Total Illness Uncertainty Categories and Illness Uncertainty Subscores
| Number of items |
Range |
||||
|---|---|---|---|---|---|
| Potential | Actual | Mean | SD | ||
| Total uncertainty | 32 | [32, 160] | [49, 125] | 87.50 | 13.97 |
| Mean uncertainty subscale scores | |||||
| Ambiguity | 13 | [1, 5] | [1.38, 4.46] | 3.06 | 0.66 |
| Complexity | 7 | [1, 5] | [1, 4.00] | 2.12 | 0.44 |
| Inconsistency | 7 | [1, 5] | [1, 3.57] | 2.29 | 0.56 |
| Unpredictability | 5 | [1, 5] | [1.80, 5] | 3.38 | 0.60 |
Table 2.
Illness Uncertainty Item by Item
| Item | Summarized question | IQR | Mode |
|---|---|---|---|
| Ambiguity | |||
| 3 | Unsure if getting better/worse | [2, 4] | 2 |
| 4 | Unclear about pain | [2, 4] | 4 |
| 8 | Don't know when things will be done | [2, 4] | 4 |
| 9 | Symptoms change unpredictably | [2, 4] | 4 |
| 13 | Treatment too complex | [2, 3] | 2 |
| 14 | Difficult to know if treatment helps | [2, 4] | 2 |
| 16 | Cannot plan for the future | [2, 4] | 4 |
| 17 | Illness changing with good and bad days | [4, 4] | 4 |
| 18 | Unclear about managing at home | [2, 4] | 2 |
| 20 | Don't know what will happen to me | [2, 4] | 2 |
| 23 | Treatment benefit unknown | [2, 4] | 2 |
| 24 | Don't know about caring for myself | [2, 4] | 4 |
| 26 | Treatment changes what I can do | [2, 4] | 4 |
| Complexity | |||
| 6 | Purpose of treatment clear | [4, 5] | 4 |
| 7 | I understand what pain means | [3, 4] | 4 |
| 10 | Understand everything | [4, 4] | 4 |
| 28 | Treatment will be successful | [3, 4] | 4 |
| 31 | Nurses are available for care | [4, 4] | 4 |
| 32 | Illness seriousness is known | [4, 4] | 4 |
| 33 | Providers use words I know | [4, 4] | 4 |
| Inconsistency | |||
| 1 | Unsure what is wrong | [1, 2] | 2 |
| 2 | Many unanswered questions | [2, 3] | 2 |
| 5 | Vague explanations from providers | [2, 3] | 2 |
| 11 | Providers use words with many meanings | [2, 4] | 2 |
| 19 | Different opinions about what is wrong | [2, 2] | 2 |
| 22 | Inconsistent test results | [2, 3] | 2 |
| 29 | No specific diagnosis | [1, 2] | 2 |
| Unpredictability | |||
| 12 | Predict how long illness will last | [1, 2] | 2 |
| 21 | I know when days are good | [2, 4] | 2 & 4 |
| 25 | Predict my illness course | [2, 4] | 2 |
| 27 | Nothing else is wrong with me | [2, 3] | 2 |
| 30 | Physical distress is predictable | [2, 4] | 2 |
Note. IQR: interquartile range. Response values: 1 Strongly disagree, 2 Disagree, 3 Undecided, 4 Agree, 5 Strongly agree.
Characteristics of the study sample, including the person factors and antecedents possibly related to illness uncertainty, are reported both for the whole sample and by different total uncertainty levels. In Table 3, patients were mostly male (61%), with a disproportionally lower percentage of the high illness uncertainty group being male (39% vs 65%; p = 0.037). The vast majority of patients were non-Hispanic White (94%), and more than half of the patients had received education at the college level or above (57%). In Table 4, the study sample was on average 56 years of age, and the group with high uncertainty was significantly younger than the group with moderate or low uncertainty (51 vs 57; p = 0.017).
Table 3.
Description of Sample using Categorical Measures (N=115)
| Uncertainty |
||||
|---|---|---|---|---|
| Characteristic | Total | High | Mid/low | |
| % (n) | % (n) | % (n) | p | |
| Gender | .037 | |||
| Female | 39.1% (45) | 61.1% (11) | 35.1% (34) | |
| Male | 60.9% (70) | 38.9% (7) | 64.9% (63) | |
| Race | .240 | |||
| Non-Hispanic White | 93.9% (108) | 100.0% (18) | 92.8% (90) | |
| Other* | 6.1% (7) | 0.0% (0) | 7.2% (7) | |
| Highest level of education | .446 | |||
| Less than college | 43.5% (50) | 55.6% (10) | 41.2% (40) | |
| College | 47.0% (54) | 33.3% (6) | 49.5% (48) | |
| Above college | 9.6% (11) | 11.1% (2) | 9.3% (9) | |
Other race category includes: 1 Hispanic White, 5 Black, and 1 more than one race.
Table 4.
Description of Sample using Continuous Measures (N=115)
| Uncertainty |
||||
|---|---|---|---|---|
| Characteristic | Total | High | Mid/low | |
| Mean (SD) | Mean (SD) | Mean (SD) | p | |
| Age | 56.0 (9.9) | 50.9 (11.1) | 56.9 (9.5) | .017 |
| FACT-G functional well-being | 15.4 (5.9) | 10.4 (5.6) | 16.4 (5.5) | .000 |
| FACT-G physical well-being | 19.0 (6.3) | 14.0 (6.4) | 19.9 (5.9) | .000 |
| MELD score (illness severity) | 15.4 (4.2) | 15.3 (2.4) | 15.5 (4.5) | .864 |
| FACT-G social well-being | 20.4 (5.0) | 16.0 (5.6) | 21.3 (4.4) | .000 |
| SCS problem solving | 78.3 (15.5) | 71.9 (15.3) | 79.4 (15.4) | .058 |
| SCS cognitive reframing | 76.5 (15.1) | 71.3 (15.6) | 77.4 (14.9) | .116 |
| Self-efficacy | 63.1 (18.7) | 44.4 (15.0) | 66.6 (17.2) | .000 |
Note. MELD: Model for End-Stage Liver Disease; FACT-G: Functional Assessment of Cancer Therapy – General; SCS: Self Control Schedule.
As shown in Table 4, patients with high illness uncertainty levels reported lower FACT-G well-being scores: low functional well-being (10.4 vs 16.4), low physical well-being (14.0 vs 19.9), and low social well-being (16.0 vs 21.3), as well as low self-efficacy (44.4 vs 66.6), compared to patients with moderate or low illness uncertainty levels; these differences were statistically significant (p < 0.001). Patients with high illness uncertainty levels also reported lower problem solving scores (71.9 vs 79.4) with a trend towards statistical significance (p = 0.058).
We report results from the multiple regression model of high total illness uncertainty in Table 5, including the adjusted odds ratio for each of the independent variables: education, age, functional well-being, physical well-being, problem solving, cognitive reframing, self-efficacy, and social well-being. After adjusting for all of the control variables, low social well-being (odds ratio = 0.816, with 95% CI [0.683, 0.975]) and low self-efficacy (odds ratio = 0.931, with 95% CI [0.881, 0.985]) were significantly associated with higher odds of reporting a high total illness uncertainty score.
Table 5.
Multivariate Associations of High Total Illness Uncertainty
| Independent variable | AOR | 95% CI | p |
|---|---|---|---|
| Age | 0.947 | [0.881, 1.017] | 0.135 |
| FACT-G functional well-being | 1.012 | [0.840, 1.219] | 0.901 |
| FACT-G physical well-being | 0.954 | [0.823, 1.106] | 0.534 |
| SCS problem solving | 1.002 | [0.921, 1.091] | 0.954 |
| SCS cognitive reframing | 1.019 | [0.935, 1.112] | 0.663 |
| Self-efficacy | 0.931 | [0.881, 0.985] | 0.013 |
| FACT-G social well-being | 0.816 | [0.683, 0.975] | 0.025 |
| Education: Less than college (ref: college) | 1.262 | [0.297, 5.363] | 0.753 |
| Education: Above college (ref: college) | 2.411 | [0.209, 27.865] | 0.481 |
Note. AOR: adjusted odds ratio; CI: confidence interval.
FACT-G: Functional Assessment of Cancer Therapy – General; SCS: Self Control Schedule.
Table 6 shows the multiple regression results for the four uncertainty subscales. Low social well-being scores were significantly related to high complexity scores (beta = −0.026; p < 0.01). Low self-efficacy scores were significantly related to high ambiguity scores (beta = −0.014; p < 0.001) and high inconsistency scores (beta = −0.012; p < 0.001), as well as related to high complexity scores with a trend towards significance (beta = −0.005; p = 0.09). Additionally, though not significantly related to high total illness uncertainty, functional well-being scores were significantly related to ambiguity scores (beta = −0.033; p < 0.01). Having not attended college, compared to having attended college as the highest education level, was significantly related to higher ambiguity scores (beta = 0.209; p = 0.04) and higher inconsistency scores (beta = 0.215; p = 0.03). Interestingly, having received education beyond college, compared to having attended college as the highest education level, was significantly related to higher unpredictability scores (beta = 0.358; p = 0.02).
Table 6.
Multivariate Associations of Illness Uncertainty Sub scale scores: (Ambiguity, Complexity, Inconsistency, Unpredictability)
| Independent variable | Ambiguity | Complexity | ||||
|---|---|---|---|---|---|---|
| AC | 95% CI | p | AC | 95% CI | p | |
| Age | −0.0033 | [−0.0117, 0.0050] | 0.429 | 0.0058 | [−0.0029, 0.0145] | 0.189 |
| FACT-G functional well-being | −0.0329 | [−0.0565, −0.0093] | 0.006 | −0.0009 | [−0.0198, 0.0179] | 0.921 |
| FACT-G physical well-being | 0.0009 | [−0.0209, 0.0227] | 0.933 | 0.0003 | [−0.0156, 0.0161] | 0.974 |
| SCS problem solving | −0.0024 | [−0.0104, 0.0056] | 0.551 | 0.0052 | [−0.0034, 0.0139] | 0.232 |
| SCS cognitive reframing | 0.0029 | [−0.0057, 0.0115] | 0.501 | −0.0077 | [−0.0171, 0.0017] | 0.103 |
| Self-efficacy | −0.0145 | [−0.0208, −0.0083] | 0.000 | −0.0047 | [−0.0102, 0.0008] | 0.093 |
| FACT-G social well-being | −0.0081 | [−0.0284, 0.0123] | 0.432 | −0.0262 | [−0.0454, −0.0071] | 0.007 |
| Education: Less than college (ref: college) | 0.2086 | [0.0093, 0.4079] | 0.038 | −0.0991 | [−0.2703, 0.0722] | 0.251 |
| Education: Above college (ref: college) | −0.1786 | [−0.5291, 0.1719] | 0.312 | 0.0453 | [−0.2133, 0.3039] | 0.728 |
| Independent variable | Inconsistency | Unpredictability | ||||
| AC | 95% CI | p | AC | 95% CI | p | |
| Age | −0.0017 | [−0.0128, 0.0095] | 0.767 | −0.0027 | [−0.0148, 0.0094] | 0.659 |
| FACT-G functional well-being | −0.0089 | [−0.0339, 0.0161] | 0.479 | −0.0099 | [−0.0388, 0.0190] | 0.499 |
| FACT-G physical well-being | 0.0125 | [−0.0095, 0.0344] | 0.260 | −0.0081 | [−0.0305, 0.0142] | 0.471 |
| SCS problem solving | −0.0012 | [−0.0112, 0.0087] | 0.804 | 0.0047 | [−0.0063, 0.0157] | 0.396 |
| SCS cognitive reframing | 0.0009 | [−0.0097, 0.0115] | 0.864 | −0.0054 | [−0.0170, 0.0063] | 0.360 |
| Self-efficacy | −0.0125 | [−0.0189, −0.0061] | 0.000 | −0.0034 | [−0.0137, 0.0069] | 0.510 |
| FACT-G social well-being | −0.0193 | [−0.0415, 0.0029] | 0.085 | 0.0022 | [−0.0257, 0.0300] | 0.876 |
| Education: Less than college (ref: college) | 0.2153 | [0.0218, 0.4089] | 0.027 | 0.0983 | [−0.1527, 0.3492] | 0.437 |
| Education: Above college (ref: college) | 0.0470 | [−0.2306, 0.3246] | 0.737 | 0.3584 | [0.0586, 0.6581] | 0.018 |
Note. AC: adjusted coefficient; CI: confidence interval.
FACT-G: Functional Assessment of Cancer Therapy – General; SCS: Self Control Schedule.
DISCUSSION
This study described illness uncertainty in patients waiting for liver transplant, using three levels of measures: the total uncertainty levels both as continuous scores and as the prevalence of high total uncertainty, the uncertainty subscale scores, and the itemized illness uncertainty levels. Our findings contribute to the further development and application of the illness uncertainty measure (MUIS-A) by identifying a cut-off score of 100 for high uncertainty levels. However, a recent study of patients undergoing hematopoietic stem cell transplant reported cut points for the MUIS-A and established that a score of more than 88 was indicative of high illness uncertainty reflecting a score 7 points above the total scale mean of 82.5. They reported a mean illness uncertainty score of 95.46 (SD 9.5) with a range of 77–118 (Adarve & Osorio, 2019).
We found that 15.6% of the study sample, or one in seven patients, experienced high levels of illness uncertainty. Of interest were the participant responses to items on the ambiguity and complexity subscales. Mean responses on the 13 item Ambiguity subscale were 3.06 with respondents endorsing a high level of uncertainty on items such as “Unsure if I am getting better or worse” and “I cannot plan for the future”. Similar to those responses, participants responded with a high degree of certainty to 7 items on the Complexity subscale (subscale mean score 2.12). This subscale included items such as: “Purpose of treatment is clear” and “Providers use words I know”.
Reducing ambiguity and complexity while waiting for a liver transplant is challenging, however possible. Future interventions might target ambiguity by helping patients integrate uncertainty into their daily lives and teaching them how to cognitively reframe the way they view their illness. Incorporating uncertainty into one’s daily life could include strategies such as finding joy in helping others, time spent with friends and family, and satisfaction with hobbies. Cognitively reframing events would include tactics such as reexamining events related to the illness and then talking about them with family members, close friends, and healthcare provider. Patients could be encouraged to focus on the positives in their lives. These strategies were beneficial in men with prostate cancer electing a course of watchful waiting (Bailey Jr. et al., 2004). Complexity might be addressed by explaining how the United Network for Organ Sharing (UNOS) functions to allocate livers for life-saving transplant.
We also described the characteristics of patients with high uncertainty levels. We identified a priori these characteristics (independent variables in our model) of uncertainty based on the Uncertainty in Illness Theory, then estimated the associations between high uncertainty and these variables. Low social well-being and low self-efficacy were significantly related to high levels of illness uncertainty. Patients with lower social well-being received less support from family and experienced higher concerns in family communication as it related to their illness. Patients with low self-efficacy were less confident in their abilities to manage their illness related symptoms.
The important role of family support in assisting patients to manage end-stage liver disease and waiting for a transplant has been noted as crucial to the patient’s integration of their illness in their life (Wainwright, 1995). In theory, this strategy to include a family member should have addressed the patient participants’ social well-being and self-efficacy. Future studies might include caregiver support strategies as a booster to the uncertainty management strategies or consider the use of a social network lens that would include the use of social media or online communities to support patients during this period of waiting (Allen et al., 2020; Grumme & Gordon, 2016).
A strength of our study was the use of theory to guide the selection of variables. However, our study was limited by the lack of ethnic diversity as 94% of the sample was non-Hispanic white. This limits the generalization of our findings to the population of individuals with life-limiting illnesses. Another limitation was the sample size for our analyses. We need to cautiously interpret the adjusted odds ratio and association between our variables and outcomes, as well as further validate the conclusions in other studies with similar populations.
The implications of our findings could be useful for clinicians in the identification of those seriously ill patients most in need of intervention. Clinicians could use of a brief illness uncertainty measure as part of the clinical encounter (Northouse et al., 2013). This would give the them additional information that could then be used to address the patient’s concerns. Clinicians could also ask questions to understand what kinds of support is available and determine how confident the patient is in managing their illness related concerns. These data be used to develop tailored intervention strategies to address the needs of patients awaiting liver transplant as well as other patients with chronic illnesses that opt to take a “wait and see” or active surveillance approach, e.g. men with early stage prostate cancer and patients with early stage gliomas (Kazer et al., 2011; Khan et al., 2016).
In sum, addressing the negative impact of high levels of illness uncertainty particularly in the areas of patient understanding about their illness and communication with the clinical care team are important. Assessing the social support needs of the patient that they receive from family and their self-efficacy or confidence in their abilities to manage symptoms should be part of the clinical encounter. This has the potential to lead to the co-production of care between providers and patients that are needed during a period of waiting.
Acknowledgments:
We sincerely thank Karen Stechuchak who prepared data for the analyses.
Funding:
This study was supported by grant P01 NR010948 from NIH/NINR.
Footnotes
Declaration of Conflicting Interests: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Contributor Information
Donald E Bailey, Jr, Duke University School of Nursing, Durham, NC, USA.
Jia Yao, Center for Health Policy & Inequities, Duke University, Durham, NC, USA.
Qing Yang, Duke University School of Nursing, Durham, NC, USA.
REFERENCES
- Adarve SE, & Osorio JH (2019). Factors Associated With Uncertainty in Patients Scheduled to Undergo Hematopoietic Stem Cell Transplantation. Cancer Nursing. 10.1097/ncc.0000000000000773 [DOI] [PubMed] [Google Scholar]
- Affleck G, Tennen H, Pfeiffer C, & Fifield J (1987). Appraisals of control and predictability in adapting to a chronic disease. Journal of Personality and Social Psychology, 53(2), 273–279. [DOI] [PubMed] [Google Scholar]
- Allen C, Vassilev I, Kennedy A, & Rogers A (2020). The work and relatedness of ties mediated online in supporting long-term condition self-management. Sociology of Health and Illness, 42(3), 579–595. 10.1111/1467-9566.13042 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bailey DE Jr., Hendrix CC, Steinhauser KE, Stechuchak KM, Porter LS, Hudson J, Olsen MK, Muir A, Lowman S, DiMartini A, Salonen LW, & Tulsky JA (2017). Randomized trial of an uncertainty self-management telephone intervention for patients awaiting liver transplant. Patient Educ Couns, 100(3), 509–517. 10.1016/j.pec.2016.10.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bailey DE Jr., Barroso J, Muir AJ, Sloane R, Richmond J, McHutchison J, Patel K, Landerman L, & Mishel MH (2010). Patients with chronic hepatitis C undergoing watchful waiting: Exploring trajectories of illness uncertainty and fatigue. Research in Nursing and Health, 33(5), 465–473. 10.1002/nur.20397 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bailey DE Jr., Hendrix C, Steinhauser K, Stechuchak KM, Porter LS, Hudson J, Olsen M, Muir AJ, Lowman SG, DiMartini A, Salonen LW, & Tulsky JA (2017). Randomized trial of an uncertainty self-management telephone intervention for patients awaiting liver transplant. Patient Education and Counseling, 100(3), 509–517. 10.1016/j.pec.2016.10.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bailey DE Jr., Mishel MH, Belyea M, Stewart JL, & Mohler J (2004). Uncertainty intervention for watchful waiting in prostate cancer. Cancer Nursing, 27(5), 339–346. http://www.ncbi.nlm.nih.gov/pubmed/15525860 http://graphics.tx.ovid.com/ovftpdfs/FPDDNCIBCFOJOI00/fs047/ovft/live/gv024/00002820/00002820-200409000-00001.pdf [DOI] [PubMed] [Google Scholar]
- Braden CJ (1990). A test of the Self-Help Model: learned response to chronic illness experience. Nursing Research, 39(1), 42–47. http://www.ncbi.nlm.nih.gov/pubmed/2294514 [PubMed] [Google Scholar]
- Braden CJ (1991). Patterns of change over time in learned response to chronic illness among participants in a systemic lupus erythematosus self-help course. Arthritis Care and Research, 4(4), 158–167. [DOI] [PubMed] [Google Scholar]
- Cella DF, Tulsky DS, Gray G, Sarafian B, Linn E, Bonomi A, Silberman M, Yellen SB, Winicour P, Brannon J, & et al. (1993). The Functional Assessment of Cancer Therapy scale: development and validation of the general measure. Journal of Clinical Oncology, 11(3), 570–579. http://www.ncbi.nlm.nih.gov/pubmed/8445433 http://jco.ascopubs.org/content/11/3/570.full.pdf [DOI] [PubMed] [Google Scholar]
- Colagreco JP, Bailey DE Jr., Fitzpatrick JJ, Afdhal NH, & Lai M (2014). Watchful waiting: role of disease progression on uncertainty and depressive symptoms in patients with chronic hepatitis C. Journal of Viral Hepatitis, 21(10), 727–733. 10.1111/jvh.12207 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fleury J, Kimbrell LC, & Kruszewski MA (1995). Life after a cardiac event: women's experience in healing. Heart and Lung, 24(6), 474–482. [DOI] [PubMed] [Google Scholar]
- Golfieri L, Gitto S, Vukotic R, Andreone P, Marra F, Morelli MC, Cescon M, & Grandi S (2019). Impact of psychosocial status on liver transplant process [Review]. Annals of Hepatology, 18(6), 804–809. 10.1016/j.aohep.2019.06.011 [DOI] [PubMed] [Google Scholar]
- Grumme VS, & Gordon SC (2016). Social Media Use by Transplant Recipients for Support and Healing. Computers, Informatics, Nursing, 34(12), 570–577. 10.1097/CIN.0000000000000276 [DOI] [PubMed] [Google Scholar]
- Kazer MW, Bailey DE Jr., Sanda M, Colberg J, & Kelly WK (2011). An Internet intervention for management of uncertainty during active surveillance for prostate cancer. Oncology Nursing Forum, 38(5), 561–568. 10.1188/11.ONF.561-568 [DOI] [PubMed] [Google Scholar]
- Khan OH, Mason W, Kongkham PN, Bernstein M, & Zadeh G (2016). Neurosurgical management of adult diffuse low grade gliomas in Canada: a multi-center survey. Journal of Neuro-Oncology, 126(1), 137–149. 10.1007/s11060-015-1949-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kimbell B, Boyd K, Kendall M, Iredale J, & Murray SA (2015). Managing uncertainty in advanced liver disease: a qualitative, multiperspective, serial interview study. BMJ Open, 5(11), e009241. 10.1136/bmjopen-2015-009241 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Larson AM (2015). Palliative care for patients with end-stage liver disease [Review]. Current Gastroenterology Reports, 17(5), 440. 10.1007/s11894-015-0440-6 [DOI] [PubMed] [Google Scholar]
- Larson AM, & Curtis JR (2006). Integrating palliative care for liver transplant candidates: "too well for transplant, too sick for life". JAMA, 295(18), 2168–2176. 10.1001/jama.295.18.2168 [DOI] [PubMed] [Google Scholar]
- Lasker JN, Sogolow ED, Olenik JM, Sass DA, & Weinrieb RM (2010). Uncertainty and liver transplantation: women with primary biliary cirrhosis before and after transplant. Women and Health, 50(4), 359–375. 10.1080/03630242.2010.498750 [DOI] [PubMed] [Google Scholar]
- Lorig K, Chastain RL, Ung E, Shoor S, & Holman HR (1989). Development and evaluation of a scale to measure perceived self-efficacy in people with arthritis. Arthritis and Rheumatism, 32(1), 37–44. https://www.ncbi.nlm.nih.gov/pubmed/2912463 [DOI] [PubMed] [Google Scholar]
- McCain NL, & Cella DF (1995). Correlates of stress in HIV disease. Western Journal of Nursing Research, 17(2), 141–155. 10.1177/019394599501700203 [DOI] [PubMed] [Google Scholar]
- Mishel MH (1981). The measurement of uncertainty in illness. Nursing Research, 30(5), 258–263. http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=ovfta&NEWS=N&AN=00006199-198109000-00002 [PubMed] [Google Scholar]
- Mishel MH (1988). Uncertainty in illness. Image - the Journal of Nursing Scholarship, 20(4), 225–232. 10.1111/j.1547-5069.1988.tb00082.x [DOI] [PubMed] [Google Scholar]
- Mishel MH (1997). Uncertainty in Illness Scales manual. Available from Mishel M at the University of North Carolina-Chapel Hill, School of Nursing. [Google Scholar]
- Mishel MH, & Braden CJ (1988). Finding meaning: antecedents of uncertainty in illness. Nursing Research, 37(2), 98–103. [PubMed] [Google Scholar]
- Mishel MH, & Murdaugh CL (1987). Family adjustment to heart transplantation: redesigning the dream. Nursing Research, 36(6), 332–338. [PubMed] [Google Scholar]
- Montagnese S, & Bajaj JS (2019). Impact of Hepatic Encephalopathy in Cirrhosis on Quality-of-Life Issues [Review]. Drugs, 79(Suppl 1), 11–16. https://doi.org/ 10.1007/s40265-018-1019-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mshel MH, Belyea M, Germino B, Stewart JL, Bailey DE Jr., Roberton C, & Mohler J (2002). Helping patients with localized prostate carcinoma manage uncertainty and treatment side effects: nurse-delivered psychoeducational intervention over the telephone. Cancer, 94(6), 1854–1866. [DOI] [PubMed] [Google Scholar]
- Nelson JP (1996). Struggling to gain meaning: Living with the uncertainty of breast cancer. Advances in Nursing Science, 18(3), 59–76. [DOI] [PubMed] [Google Scholar]
- Northouse LL, Mood DW, Schafenacker A, Kalemkerian G, Zalupski M, LoRusso P, Hayes DF, Hussain M, Ruckdeschel J, Fendrick AM, Trask PC, Ronis DL, & Kershaw T (2013). Randomized clinical trial of a brief and extensive dyadic intervention for advanced cancer patients and their family caregivers [Multicenter Study Randomized Controlled Trial Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't]. Psycho-Oncology, 22(3), 555–563. https://doi.org/ 10.1002/pon.3036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pérez-San-Gregorio MÁ, Martín-Rodríguez A, Domínguez-Cabello E, Fernández-Jiménez E, Borda-Más M, & Bernardos-Rodríguez Á (2012). Mental health and quality of life in liver transplant and cirrhotic patients with various etiologies. International Journal of Clinical and Health Psychology, 12(2), 203–218. http://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=2012-11670-002&site=ehost-live&scope=site anperez@us.es [Google Scholar]
- Porter LS, Keefe FJ, Garst J, Baucom DH, McBride CM, McKee DC, Sutton L, Carson K, Knowles V, Rumble M, & Scipio C (2011). Caregiver-assisted coping skills training for lung cancer: results of a randomized clinical trial [Article]. Journal of Pain and Symptom Management, 41(1), 1–13. 10.1016/j.jpainsymman.2010.04.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Porter LS, Keefe FJ, Garst J, McBride CM, & Baucom D (2008). Self-efficacy for managing pain, symptoms, and function in patients with lung cancer and their informal caregivers: associations with symptoms and distress. Pain, 137(2), 306–315. 10.1016/j.pain.2007.09.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rosenbaum M. (1983). Learned resourcefulness as a behavioral repertoire for the self-regulation of internal events: Issues and speculations. In Rosenbaum M, Franks CM, & Jaffe Y (Eds.), Perspectives on Behavior Therapy in the Eighties (pp. 54–73). Springer Pub. Co. [Google Scholar]
- Teixeira HR, Marques DM, Lopes AR, Ziviani LC, Magro JT, Mente ED, Castro ESO, Galvao CM, & Mendes KD (2016). Anxiety and Stress Levels on Liver Transplantation Candidates. Transplantation Proceedings, 48(7), 2333–2337. 10.1016/j.transproceed.2016.06.031 [DOI] [PubMed] [Google Scholar]
- Wainwright SP (1995). The transformational experience of liver transplantation. Journal of Advanced Nursing, 22(6), 1068–1076. 10.1111/j.1365-2648.1995.tb03107.x [DOI] [PubMed] [Google Scholar]