The Impact of Pruritus on Quality of Life: The Skin Equivalent of Pain

Objective To compare the impact of chronic pruritus and chronic pain on quality of life (QoL) using directly elicited health utility scores.

Design Cross-sectional study.

Setting Convenience sample of patients attending the Emory Dermatology Clinic, Emory Spine Center, and Emory Center for Pain Management, Atlanta, Georgia.

Participants Adult men and women (aged ≥18 years) experiencing chronic pain or pruritus for 6 weeks or more.

Main Outcome Measures The mean utility score of patients with chronic pruritus was compared with that of patients with chronic pain. A regression analysis was performed to determine the impact of the primary predictor variable —symptom type —on the primary outcome variable —mean utility score (a metric representing the impact on QoL).

Results The study included 73 patients with chronic pruritus and 138 patients with chronic pain. The mean (SD) utility among patients with pruritus was 0.87 (0.27) compared with 0.77 (0.31) for patients with pain (P   <  .01). After symptom severity, duration, and demographic factors were controlled for, only symptom severity (0.03 [P   <  .05]) and single marital status ( −0.12 [P  =  .02]), but not symptom type (P  =  .43), remained significant predictors of the mean symptom utility score.

Conclusions Chronic pruritus has a substantial impact on QoL, one that may be comparable to that of pain. The severity of symptoms and the use of support networks are the main factors that determine the degree to which patients are affected by their symptoms. Addressing support networks in addition to developing new therapies may improve the QoL of itchy patients.

Several studies have attempted to demonstrate that chronic pruritus has a significant effect on health-related quality of life (QoL).^1-3 Because chronic pruritus is often not amenable to treatments for acute pruritus, it can result in a debilitating course, including the development of symptoms of depression, global distress, and impairment of sleep.³ Although few large-scale epidemiological studies have been performed to determine the prevalence of chronic pruritus, there is evidence to suggest that it is a common problem: a cross-sectional study of skin disease in France found that pruritic cutaneous conditions may affect 20% to 30% of the French population.⁴

Despite the reported widespread and debilitating effect of chronic pruritus, research relating to this symptom has not been as well established as in other chronic conditions such as pain. Chronic pain and pruritus share many similarities: both are complex, subjective symptoms that have been demonstrated to have a compelling effect on health-related QoL, including the development of symptoms of depression and the impairment of activities of daily living.^5-7 However, unlike chronic pruritus, pain syndromes have been well studied in health services and outcomes research, resulting in a better understanding of this complex condition and the development of novel treatments. Given the similarities between these 2 symptoms, we believe that chronic pruritus has a substantial QoL impact, comparable to that of pain. The purpose of this study was to determine the extent of the QoL impact of these 2 conditions by measuring health utility scores —a metric that represents a subjective level of satisfaction with a certain health state —in patients with chronic pruritus and pain.

The institutional review board of Emory University, Atlanta, Georgia, approved this cross-sectional study. The survey materials included paper-based questionnaires regarding demographics, clinical parameters, and symptom severity. In addition to these paper-based surveys, each consenting patient completed a face-to-face time trade-off (FTF TTO) interview to assess the QoL impact of their symptom (pain or pruritus) by generating a utility score for their respective health state. The FTF TTO method is a standard method with which to elicit health utility scores.⁸ Utility scores are numeric values that are expressed on a continuous scale anchored at death (utility score, 0) to perfect health (utility score,  1). Health utility scores closer to 0 indicate a greater burden of disease than that of a health state closer to 1. Previously reported mean health utility scores for other dermatological conditions include 0.640 for bullous diseases, 0.867 for mycosis fungoides, and 0.938 for acne vulgaris.⁹

In the FTF TTO interview, we used a computerized utility instrument to provide patients with a hypothetical decision: living the rest of their life in their current health state or living for a shorter period in perfect health. In other words, patients could exchange a portion of their future survival time in order to live in perfect health during a shortened life span. The length of time in perfect health was varied until the patient was indifferent about the decision. The FTF TTO –derived utility score is the ratio of the time remaining after the trade to the life expectancy of the individual. For example, a patient with a life expectancy of 75 years who was willing to give up 3 years to live without pruritus would have a utility score of 72/75, or 0.960, for the health state of chronic pruritus. Simply put, this patient would be willing to forfeit 4% of his or her life expectancy to live without pruritus. The eAppendix details the elicitation method.

From June 2007 to April 2008, we recruited adult patients (aged ≥18 years) with chronic pain or pruritus (symptom duration >6 weeks) from the Emory Spine Center and the Center for Pain Management or from the Emory Dermatology Clinic, respectively, Atlanta, Georgia. Potential patients were recruited via flyers placed in their respective clinic offices or were informed of the study by their treating physician. We excluded patients who demonstrated an inability to speak or read English or any other disability that would prevent them from completing both the questionnaire and the FTF TTO. Informed consent was obtained from all patients before they were included in the study. We included patients only once in the study, regardless of the number of times that they attended the pain or the dermatology clinic during the duration of the study.

The main outcome variable in this study was the symptom utility score. For the purposes of statistical analysis, we grouped the primary predictor variables so that each variable had 3 or fewer potential degrees of freedom. Sociodemographic variables were limited to age, sex, race (white, nonwhite), marital status (married, single), education ( ≤high school, college, graduate school), and household income ( <$50  000, $50  000-100  000, >$100  000). Clinical variables included symptom duration ( <1, 1-5, 5-10, or >10 years) and symptom severity. The patients rated their symptom severity over the past 6 weeks “at its worst, ” “at its best, ” “at present, ” and “on average ” using a visual analog scale ranging from 0 to 10; the results were grouped into mild (1-3), moderate (4-6), and severe (7-10). We used the daily average assessment of severity in the statistical analyses in this study.

We used the independent sample t test and the χ² test in univariate analyses to determine the impact of continuous and categorical variables, respectively, on the mean utility score. We developed a generalized linear model to assess the effect of symptom type (pain or pruritus) on the primary outcome variable, ie, the mean utility score, after controlling for symptom severity, symptom duration, age, sex, race, marital status, and educational and income levels. We assessed variance inflation factors for each covariate of interest for multicollinearity.

All tests were 2-sided. P   <  .05 was required for statistical significance. We performed all analyses with SAS software (Version 9.1; SAS Institute, Cary, North Carolina).

Rather than power for an equivalence study, we assumed a minimum effect size that would be meaningful between the 2 sets of utility scores. Power was calculated with the goal of detecting a difference of at least 0.10 in the mean utility score (deemed as a clinically significant difference¹⁰) between the pain and the pruritus groups. We assumed a 2-sided t test, with a common SD of 0.30, an α value of .05, and 80% power to estimate the need for 125 individuals per group. To power our regression analysis, we used the rule of thumb of 10 patients per independent variable. With the 9 anticipated variables, we would need 90 patients.

A total of 138 patients with chronic pain and 73 patients with chronic pruritus were recruited. The demographic characteristics of the participants and the distribution of the severity of symptoms among the 2 groups are shown in Table 1. Continuous variables are reported as mean (SD), and categorical variables are reported as proportions. There were no differences in the demographic characteristics between the 2 groups. Both the pain and the pruritus groups had a mean age of 55 years, were mostly female (62% vs 58%, respectively), and were predominantly white (80% vs 74%). In both the pain and the pruritus groups, the median duration of symptoms was 6 months to 1 year, and almost half of the patients characterized the severity of their symptoms as moderate. A greater percentage (36% vs 28% [P   <  .001]) of the patients in the pain group, however, characterized their symptom as severe. The mean utility score among patients with pruritus was 0.87 (0.27) compared with 0.77 (0.31) for patients with pain (P   <  .01).

We report the parameter estimates of the regression model in Table 2. Single marital status ( −0.12 [P  =  .02]) and symptom severity (0.03 [P   <  .05]) remained significant predictors of the symptom utility score after other subject characteristics were controlled for. Race trended toward significance (P  =  .07). Neither symptom (pain or pruritus) type nor symptom duration was a significant predictor of the symptom utility score.

This study of patients with chronic pruritus and pain has several important findings. First, our data suggest that chronic pruritus carries a considerable burden of disease, as measured by health utility scores. In this study, the mean health utility score of patients with chronic pruritus was 0.87, indicating that the average patient was willing to forfeit 13% of his or her life expectancy to live without pruritus. Previously reported health utility scores for other dermatological conditions in a similar utility range include bullous diseases (0.640) and mycosis fungoides (0.867).⁹

Second, our results from this small study indicate that chronic pruritus may have a QoL impact comparable to that of chronic pain. In our univariate analysis, patients with chronic pruritus had a significantly greater mean utility score (0.87) for their symptom than patients with chronic pain (0.77). However, on multivariate analysis, symptom type was not a significant predictor (P  =  .43). Although it may appear as though patients with chronic pain carry a considerably greater burden of disease than those with chronic pruritus, we believe that this difference exists because a greater proportion of patients characterized their symptom as severe in the pain cohort compared with the pruritus cohort (P   <  .01). Also, on multivariate analysis, after demographic and clinical variables were controlled for, symptom type (pain or pruritus) was not a significant predictor of the primary outcome variable: the mean health utility score. This finding suggests that chronic pruritus has a substantial burden of disease, a burden that may be comparable to that of chronic pain. Indeed, the primary determinants of the mean health utility score, and thus the degree of QoL impact, on multivariate modeling were symptom severity ( −0.03 [P   <  .05]) and marital status ( −0.12 [P  =  .02]). Specifically, greater pruritus severity and single marital status were associated with a lower mean health utility score (ie, worse QoL).

It may seem peculiar that marital status had a stronger measure of effect than symptom severity. However, epidemiological research indicates that social relationships (with marital status as a proxy measure) may be protective of morbidity by aiding in economic well-being, healthier lifestyles, lower stress, and social support.^11-13 Furthermore, in a large cross-sectional study of individuals with chronic pain and pruritus, Dalgard et al¹⁴ found that those individuals who were most affected by their chronic pain or pruritus tended to be younger and female and reported more depressive symptoms and significantly poorer well-being. Together, these findings underscore the importance of the psychosocial and affective dimensions in affecting patient symptoms. The role of support groups to connect patients who are experiencing similar symptoms, as well as the use of pruritus-specific QoL instruments to track improvements in patient-defined end points, may have a profound effect on reducing the burden of disease.

In this study, the median duration of symptoms for both the pain and the pruritus cohorts was 6 months to 1 year. While symptom severity and marital status were found to contribute significantly to the overall mean utility score, symptom duration (0.06 [P  =  .27]) was not a significant predictor of the mean utility score in the multivariate regression model. The reason that symptom duration did not contribute significantly to the overall mean utility score may be attributable to an adaptation phenomenon whereby patients experiencing chronic illnesses or disabilities are motivated to find ways to accommodate their symptom. Consequently, an actual patient's overall rating of his or her own QoL is often much higher than that of a healthy individual hypothetically picturing life with the disease state of interest.^15,16

Our study has several limitations. As with all survey-based studies, this study was subject to response, recall, and selection biases. Because patients in both the pain and the pruritus cohorts were recruited from Emory University, a tertiary referral center, the study population may have included highly motivated patients with more severe disease than those usually found at a nonacademic center. Similarly, patients who are severely affected may have been more likely to participate. Both of these biases, however, would result in an overestimation of the burden of disease (ie, a lower mean utility score) and for the purposes of comparison would affect both the pain and the pruritus cohorts similarly. Finally, this study did not assess the potential contribution of other comorbidities such as mood states (eg, depression, anxiety) that have been previously reported to contribute to the overall poorer well-being experienced by patients with chronic itch and pain.^3,14

Despite these limitations, the aim of this study was to provide greater insight into the burden of chronic pruritus and to compare the QoL impact of this symptom with that of chronic pain. Although itch is thought to be mediated by a neuronal pathway that is distinct from the pain-processing pathway, the sensations of itch and pain continue to be conceptualized as closely related.^17,18 Our study provides a model to compare common debilitating symptoms that are best self-reported and have very few objective findings. Symptoms other than pain and pruritus include nausea, heartburn, neuralgias, and dysesthesias. The FTF TTO health utility scores show that patients are giving up time, something of understandable value and a unit of measure that can be compared across other health conditions.

Overall, our data support previously published results indicating that patients with chronic pruritus carry a significant burden of disease.^4,14,19 This study provides preliminary data that future studies may build on by incorporating larger and more general populations of patients and by assessing the contribution of mood states (eg, depression, anxiety) to the impact of pruritus on QoL. Such other studies are needed to corroborate our findings. Our results also highlight the value of developing new therapies to improve the QoL of itchy patients and of providing support networks for these patients to discuss their conditions and to commiserate with each other.

Correspondence: Suephy C. Chen, MD, MS, Dermatology Clinical and Outcomes Research Unit, Department of Dermatology, Emory University School of Medicine, Emory Clinic, Bldg A, 1365 Clifton Rd NE, Ste 1100, Atlanta, GA 30322 (schen2@emory.edu).

Accepted for Publication: April 20, 2011.

Published Online: June 16, 2011. doi:10.1001/archdermatol.2011.178

Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: DeLong, McKenzie-Brown, Schaufele, and Chen. Acquisition of data: Kini, DeLong, McKenzie-Brown, and Schaufele. Analysis and interpretation of data: Kini, DeLong, Veledar, and Chen. Drafting of the manuscript: Kini and Chen. Critical revision of the manuscript for important intellectual content: Kini, DeLong, Veledar, McKenzie-Brown, Schaufele, and Chen. Statistical analysis: Kini, Veledar, and Chen. Administrative, technical, and material support: Kini, DeLong, McKenzie-Brown, Schaufele, and Chen. Study supervision: Chen.

Online-Only Material: Listen to an author podcast about this article.

Additional Contributions: Alice Huang, BA, Livia Van, MD, Shinko Lin, MD, Mary Jayne McIlwain, MD, Ricardo Berrios, MD, and Tunisia Finch, MD, graciously assisted with this study.