Immune thrombocytopenia (ITP) World Impact Survey (iWISh): Patient and physician perceptions of diagnosis, signs and symptoms, and treatment

Abstract Immune thrombocytopenia (ITP) is now well‐known to reduce patients' health‐related quality of life. However, data describing which signs and symptoms patients and physicians perceive as having the greatest impact are limited, as is understanding the full effects of ITP treatments. I‐WISh (ITP World Impact Survey) was an exploratory, cross‐sectional survey designed to establish the multifaceted impact of ITP, and its treatments, on patients' lives. It focused on perceptions of 1507 patients and 472 physicians from 13 countries regarding diagnostic pathway, frequency and severity of signs and symptoms, and treatment use. Twenty‐two percent of patients experienced delayed diagnosis (caused by several factors), 73% of whom felt anxious as a result. Patients rated fatigue among the most frequent, severe symptom associated with ITP at diagnosis (58% most frequent; 73% most severe), although physicians assigned it lower priority (30%). Fatigue was one of the few symptoms persisting at survey completion (50% and 65%, respectively) and was the top symptom patients wanted resolved (46%). Participating physicians were experienced at treating ITP, thereby recognizing the need to limit corticosteroid use to newly‐diagnosed or first‐relapse patients and espoused increased use of thrombopoietin receptor agonists and anti‐CD20 after relapse in patients with persistent/chronic disease. Patient and physicians were largely aligned on diagnosis, symptoms, and treatment use. I‐WISh demonstrated that patients and physicians largely align on overall ITP symptom burden, with certain differences, for example, fatigue. Understanding the emotional and clinical toll of ITP on the patient will facilitate shared decision‐management, setting and establishment of treatment goals and disease stage‐appropriate treatment selection.

severity of signs and symptoms, and treatment use. Twenty-two percent of patients experienced delayed diagnosis (caused by several factors), 73% of whom felt anxious as a result. Patients rated fatigue among the most frequent, severe symptom associated with ITP at diagnosis (58% most frequent; 73% most severe), although physicians assigned it lower priority (30%). Fatigue was one of the few symptoms persisting at survey completion (50% and 65%, respectively) and was the top symptom patients wanted resolved (46%). Participating physicians were experienced at treating ITP, thereby recognizing the need to limit corticosteroid use to newlydiagnosed or first-relapse patients and espoused increased use of thrombopoietin receptor agonists and anti-CD20 after relapse in patients with persistent/chronic disease. Patient and physicians were largely aligned on diagnosis, symptoms, and treatment use. I-WISh demonstrated that patients and physicians largely align on overall ITP symptom burden, with certain differences, for example, fatigue. Understanding the emotional and clinical toll of ITP on the patient will facilitate shared decisionmanagement, setting and establishment of treatment goals and disease stageappropriate treatment selection.

| INTRODUCTION
Primary immune thrombocytopenia (ITP) is an autoimmune disorder characterized by reduced platelet counts (<100 × 10 9 /L) and increased bleeding risk in the absence of other causes associated with thrombocytopenia. 1 ITP may affect patients' lives in many ways, including not only hemorrhagic manifestations, fear of bleeding, and secondary complications that can be associated with therapeutic options, but especially anxiety of the unknown (common in chronic diseases) and reduced energy levels. 2 Severe bleeding is rare in ITP, 3 but more frequent in elderly patients. 4,5 Paradoxically, patients with ITP are at risk of thromboembolic events. 4 Important symptoms often reported by patients when asked, but which can be easily overlooked, involve the effects of ITP on health-related quality of life (HRQoL), including unexplained fatigue (up to 45%), 4,6-8 anxiety or depression (29%), and headache (16%). 7 Corticosteroids, intravenous immunoglobulin (IVIg), and anti-D immunoglobulin are standard first-line treatment options for patients with newly-diagnosed primary ITP. 9 None are recommended for use beyond 6-12 weeks, because of lack of curative effects, high costs with transient benefit and drug-related complications. Most patients with primary ITP will relapse after first-line treatment discontinuation. 10 An unknown proportion of adults with newly-diagnosed ITP will improve (with or without treatment) by 1 year from diagnosis, with estimates ranging from 13% of patients to 60% being better within 3 years. 11,12 In the absence of head-to-head trials, rational determination of the most appropriate second-line treatment choice for patients with persistent/chronic ITP is not possible. 2,12,13 Consequently, secondline treatment is individualized according to patient and physician preferences, 13 The international ITP World Impact Survey (I-WISh) study was completed by patients with ITP and physicians experienced in treating them. The goals were to discern how ITP and associated treatments affect patients' lives, and to evaluate how aligned patient and physician perceptions are regarding symptoms, HRQoL and disease management. This primary I-WISh report explores: (a) the diagnostic pathway for patients with ITP, and (b) patient and physician perceptions of symptoms and disease management. The HRQoL results are reported in the companion manuscript. 40 2 | METHODS

| Study design
A comprehensive summary of study design and methods is provided in supplemental materials Appendix S1. So, I-WISh was an exploratory, cross-sectional survey of patients with ITP and physicians who treat ITP from Canada, China, Colombia, Egypt, France, Germany, India, Italy, Japan, Spain, Turkey, UK, and USA. Patients ≥18 years old diagnosed with ITP were recruited. Hospitalized patients were excluded. Patient survey invitations were sent via email to patient association groups (PAG) and physicians, who were responsible for further dissemination to patients either during routine consultations, targeted email, word of mouth, and/or personal contact in PAG meetings. Physicians were asked to invite patients on a consecutive basis but were not required to report the proportion who chose to participate.
Physicians (hematologists/hemato-oncologists) with a minimum, active caseload of three patients with ITP and responsible for treatment decisions when completing the survey, were recruited via local fieldwork agencies (independent contractors) with specialized disease area knowledge.
Parallel patient and physician surveys were developed by a steering group comprising expert ITP clinicians and PAG leads. The patient questionnaire comprised six sections and requested information on demographics and diagnosis (seven questions), ITP symptoms (four questions), HRQoL and emotional impact (12 questions), impact on work, finances, and support (15 questions), treatment received (17 questions), and patient-physician relationship (seven questions).
The physician survey comprised six sections that collected information on demographics (two questions), ITP diagnosis and patient caseload (seven questions), ITP symptoms (five questions), impact of ITP on aspects of patients' physical, emotional, HRQoL, and social health (11 questions), treatment patterns (13 questions), and patientphysician relationship (four questions).
Patients also completed the 10-question ITP Life Quality Index 31 on: working life or school, time taken off work or education, ability to concentrate, social life, sex life, energy levels, ability to undertake daily tasks, ability to provide support, hobbies, and capacity to exercise. Response options were either based on a 4-point scale eg "never", "sometimes", "more than half the time", and "all the time", or on a 7-point Likert scale, where 1 = "never" and 7 = "a great deal".
The 30-minute surveys (supplemental data Appendix S1) were provided to participants either online or by hard copy. Fully deidentified respondent information was collated and aggregated by local fieldwork partners; thus, surveys were unlinked and anonymized.
Survey materials and protocol were reviewed and approved by the Western Institutional Review Board. Patients and physicians were given an overview of the study and ethical approval details; those who wished to participate had to provide consent via a tick/check box before initiating.

| Statistical analyses
Patient and physician surveys were analyzed separately; there was no linkage between patients and physicians. There were no pre-specified hypotheses associated with these exploratory surveys, and as such, data were summarized narratively using descriptive statistics. For numeric variables, the respondent base, mean, and range (minimum and maximum values) were reported. For categorical variables, the total number and percentage of responses are shown. Participants with missing data were removed from all data summaries associated with that variable, but were still eligible for inclusion in other data summaries. Statistical significance analyses were not conducted.

| RESULTS
In total, 472 physicians and 1507 patients from 13 countries completed the survey between December 2017 and August 2018.
Patients were recruited by experienced ITP physicians (43%) or patient association groups (PAG; 57%), with variation by country (Table S1). Reliable estimates of how many people were contacted to participate in the survey could not be obtained.
Sixty-five percent of patients were female, with mean age 47 years and median duration of ITP 5 years (Table 1) (Table 1). Sixty-four percent of patients rated their current health state as high (5-7 on a 7-point Likert scale where 1= very poor health, 7= excellent health), nonetheless 48% reported that they had high or very high symptom burden at diagnosis (defined in supplemental methods Appendix S1).
Patient initial presentation was frequently to a general practitioner (52%), whereas diagnosis was almost always made by a hematologist (86%) ( Figure S1). Two-thirds (66%) of patients were referred to another HCP after initial presentation; the median number of physicians seen before ITP diagnosis was 2.0 (range 1-10). The mean (SD) number of symptoms reported at diagnosis was five (3.1) and this was similar whether the patient had seen one (4.7) or three physicians (5.2).

| Diagnosis
The most common diagnostic tests that physicians recalled using in patients suspected of ITP, and depending on symptom burden (assessed as asymptomatic, low, moderate or high symptom burden at the investigators' discretion), were complete blood count (79%-83%), physical examination (75%-82%) and peripheral blood smear (71%-74%). Physicians reported they would perform bone marrow biopsy/aspirate most frequently in patients they judged to have high symptom burden (73%), and less so in patients with moderate (52%) or low symptom burden (26%). Seventy-one percent of physicians believed that up to one-quarter of patients were misdiagnosed (only 7% believed there were no misdiagnoses); this was consistent across countries except for Germany where 43% of physicians believed 51%-75% of patients were misdiagnosed. Misdiagnoses were most frequently thought to be drug-induced thrombocytopenia (67%), aplastic anemia or myelodysplastic syndromes (49%), liver disease (46%), hypersplenism (38%), or leukemia (29%).

| ITP symptoms and severity
Petechiae (64%, diagnosis; 31%, survey completion) and bruising of unknown origin (65%, diagnosis; 30%, survey completion) were among the most frequent patient-reported signs and symptoms at diagnosis and, though substantially reduced, were prevalent at survey completion ( Figure 1A,B). In contrast, anxiety around maintaining a stable platelet count (34%, diagnosis; 32%, survey completion) and fatigue (discussed below) were also frequent. These four were the top signs and symptoms that patients most wanted resolved. Very few patients reported no symptoms at diagnosis (6%) or survey completion (13%). Physicians reported hearing about fatigue from their patients less frequently (31% overall; 30% diagnosis) than patients reported experiencing fatigue (58% diagnosis; 50% survey completion). Physicians with higher ITP caseload more frequently reported fatigue (37% diagnosis, 40% overall) than physicians with lower ITP caseload (18% and 23%). Fifty-nine percent of physicians believed that having fatigue would greatly impact patient HRQoL ( Figure 1C).

| Treatment and management
Physicians indicated that during the first 6 months following diagnosis, "watch and wait" was the preferred option for 39% of patients, F I G U R E 3 Treatment prescribed by physicians following relapse of ITP. ITP, immune thrombocytopenia; IVIg, intravenous immunoglobulin; TPO-RA, thrombopoietin receptor agonist rather than pharmacological intervention; it was physicians' preferred option for 34% of previously-treated patients with chronic ITP. Factors (>50%) influencing "watch and wait" instead of recommending drug treatment were higher platelet levels (89%; n = 412), patient being asymptomatic (75%; n = 348), and absence of severe bleeding symptoms (59%; n = 273). Over half the patients (54%; n = 806/1501) reported they had been on "watch-and-wait" management at some time, with no clinically meaningful differences between the sexes or age subgroups (data not shown).

| Splenectomy
Physicians were asked what proportion of patients, in their experi-  More than half of patients reported their current health as "high".

| Pharmacological treatments
This may reflect a perceived normalization of their disease burden over time, for example, getting acclimatized to certain conditions, although they had not changed, and/or developing "work arounds". Petechiae and bruising were two of the most frequent patientreported symptoms at diagnosis but were reduced by half at survey completion, a median of 5 years later. However, two other frequent patient-reported symptoms at diagnosis, fatigue and anxiety about stable platelet counts, did not substantially decrease over time. It is not clear whether this was caused by lack of treatment, failure to respond to it, or patient anxiety independent of the state of the ITP.
The fall in bleeding symptoms in 50% of patients (and even greater reductions in HMB) suggests that persistence of fatigue and platelet count anxiety were likely independent of the platelet count, although it might not have been the same patients reporting fatigue and anxiety at diagnosis and survey completion. The anxiety experienced around maintaining stable blood counts requires further research to establish if it is intrinsic or arises from external sources such as a physician, family members, or the internet. There was likely an inadvertent bias in that patients choosing to participate may have done so because they continued to experience these issues. The proportion of patients experiencing each ITP symptom and its reported severity did not differ greatly among the subgroups analyzed, suggesting that the burden of ITP affects all patients to a similar extent whether men or women, younger or older adults. This was unexpected given the likelihood that fatigue would be more prevalent in older age; this may also speak to the need to reset expectations of "normal" among patients with ITP.
For patients within 6 months of diagnosis and with mild symptoms, physicians indicated a preference for "watch-and-wait". Pharmacological treatments most commonly prescribed either at diagnosis or first relapse were corticosteroids and IVIg; both were used for first relapse but use decreased after second relapse. For patients with persistent/chronic disease or after subsequent relapse, TPO-RAs and anti-CD20 were most commonly prescribed, with splenectomy relatively infrequently selected; however, it had been performed in one-fifth of participating patients, with wide national variability. In those nations perceived to have more resources, for example, the United States, splenectomy rates were higher than in India and China, which was surprising. Again, this may have reflected the bias of patients who entered the study.

| Fatigue is often persistent and severe
The findings of I-WISh expand on a smaller previous survey of Japanese physicians and patients with ITP 7 that also reported high fatigue occurrence. [6][7][8] In I-WISh, fatigue was one of the top three symptoms (along with bleeding and unstable platelet counts) that patients most wanted resolved. Physicians considered fatigue as important, but they reported it substantially less often than did patients; physicians noted its role in impairment of quality of life and prioritized increasing energy levels when fatigue was present. Physicians believed fatigue would decrease as platelet count increased, and thus that fatigue was intrinsically related to disease activity. However, fatigue persisted, despite amelioration of other components of ITP between diagnosis and survey completion, suggesting that increasing platelet count and reducing bleeding is not sufficient to abrogate fatigue. Indeed, a recent study of fatigue in children and adolescents with ITP showed that fatigue scores did not correlate with bleeding symptoms, platelet count, or platelet response to treatment. 16 Underestimation by physicians of fatigue frequency in patients with ITP, although reported more often by physicians with longer clinical experience, has many potential causes. On the one hand, it could reflect an "insufficient" assessment in patients as the visit time may be focused on issues such as bleeding or other medical concerns.
Alternatively, patients may not always vocalize their experiences of fatigue, which could also be compounded by the large proportion of patients describing good health. These results highlight the importance of the physician fully inquiring into the degree of fatigue, including raised anxiety levels and other causes, and also patients being more forthcoming about their daily life experiences and not accepting sub-optimal "normal".
The best approach to treating fatigue remains unresolved but, if present, attempts at amelioration should certainly be made. Among many options, raising platelet count (ideally without using steroids), antidepressants and/or anti-inflammatories could be considered.
Clinical trials in this area are needed.
Despite I-WISh being the largest survey of its kind, including physicians and not just patients, the limitations include the inability to estimate response rate, and potential biases introduced inadvertently because some patients were more motivated to participate than others, for example, those with high levels of anxiety or more

| CONCLUSIONS
ITP is a serious condition which, even if infrequently fatal (thanks at least in part to an array of effective treatments), has a substantial adverse impact on the quality of many patients' daily lives. Fatigue was a frequently reported, persistent and severe symptom associated with ITP, which at best did not appear to consistently respond to platelet-increasing. Although participating physicians were experienced in ITP management and believed that fatigue greatly affects patients, they reported fatigue less frequently than did patients, likely reflecting multiple etiologies. Patients indicated the need for considerable support from multiple sources, especially at diagnosis, not only from physicians, family and friends, but also other patients highlighting the important role for PAGs. These findings should be integrated into the clinical management of each patient with ITP in an individual fashion according to each patient's needs. Further research, including I-WISh 2.0, is needed to assess and hopefully verify these exploratory findings through an appropriately powered, hypothesis-driven methodology linking the clinical to the quality of life and fatigue-related issues.

ACKNOWLEDGMENTS
We extend enormous thanks to all the patients and physicians who took the time to complete this survey. We also gratefully thank Cath-