The clinical manifestations of ITP and its management affect patients' everyday activities and well-being. Our research aimed to propose a conceptual model that describes the impact of ITP and its treatments on patients' HRQoL by using information gathered from the published literature and from the patient perspective reported in focus groups. Qualitative methods were used to group the patient reports into one biological variable, two main determinants, and five conceptual domains of HRQoL relevant to patients with ITP.
A conceptual model providing a proposed causal linkage with HRQoL can be useful for several reasons. To begin, it helps to further explore the disease area and proposes a pathway for how benefits and risks of new treatments may impact the lives of ITP patients. Also this pathway will assist researchers in selecting (or developing) appropriate outcome measures to evaluate a treatment's efficacy.
Although the etiology of ITP is poorly understood, patients with low platelet counts are at greater risk of complications. Also, platelet counts, with or without the presence of symptoms, can dictate whether to treat ITP. It is to be expected, then, that in the focus groups, patients reported keen awareness and close monitoring of the ups and downs of their platelet counts. The emphasis placed on the fluctuations in this clinical marker appears to cause nearly as much anxiety for the patient as the actual disease symptoms, particularly for patients that have undergone splenectomy with the hope of full permanent remission. Further, the focus group patients reported some distress about the confusion surrounding the disease, specifically with relation to childbearing. Future efforts to educate those affected by the disease could reduce the discrepancy between the perceptions and the medical facts.
Symptoms of the disease were found to be important factors characterizing patient well-being. According to patients, the bruising and bleeding resulting from ITP significantly worsen HRQoL. In addition, since fatigue substantially hindered patients' ability to perform their daily activities, the management of fatigue could potentially improve overall HRQoL. However, despite our finding that over 90% of the patients in the CA and OK focus groups mentioned fatigue, currently fatigue is not sufficiently often considered by clinicians who manage persons with ITP. In fact, fatigue is seldom recorded as an adverse event in clinical trials of ITP patients.
In addition to potentially severe clinical outcomes, the effects of the various ITP treatments impact multiple facets of the patients' lives. However, the adverse effects of corticosteroids, such as weight gain and mood swings, were most emphasized during the focus group discussions. Patients reported side effects of other treatments, such as hair loss and susceptibility to infection, less frequently. Since treatment with corticosteroids is usually first-line therapy for patients with ITP, it is likely that all patients in the focus groups had received treatment with corticosteroids at some time.
Patients reported some concern over increased susceptibility to infection as a result of splenectomy. These reported fears may seem unfounded when reviewing some reports of high success rates (i.e., increased platelet counts) one year following splenectomy [21, 26]. However, some evidence suggests that longer-term outcomes may not be as favorable bringing to question the risk-benefit ratio of splenectomy in the face of all the anxiety. Kojouri, et al. found that 66% of the patients had a complete response to splenectomy, defined as achievement and maintenance of a normal platelet count, (> 150 × 109/L or as defined in the original report and > 100 × 109/L for all measurements 30 days or longer after splenectomy, and with no additional treatment for ITP, except for the tapering of perioperative glucocorticoids or other treatments) with a median follow-up of 29 months. They also found surgical complication rates of 12.9% and 9.6% for laparotomy and laparoscopic splenectomy, respectively. However, they acknowledge that the follow-up duration may not have been long enough to provide a valid assessment of long-term risks associated with splenectomy, especially since relapse rates increase with duration of follow up . Portielje, et al. found that, of those patients who had experienced complete response (platelet count >100 × 109/L) within 2 years of diagnosis, 45% had experienced at least one ITP-related hospital admission in the 10-year follow-up period . McMillan and Durette studied 105 chronic ITP adults who were refractory to splenectomy. During the median follow-up period of 110 months, 6 patients (5.6%) died of treatment-related complications (including sepsis associated with long periods of immunosuppression, postoperative pancreatitis, and transfusion-related hepatitis C) .
It is worth noting that just as the model provides a pathway for describing how negative outcomes or adverse events of treatment affect the patients' HRQoL; the model also holds in the case of positive effects of treatment. Although it was not emphasized explicitly in the focus groups, any effects of a particular treatment that patients perceive as positive (e.g., a stable platelet count, increase in energy and vitality, decrease in anxiety) would likely improve the domains of HRQoL.
ITP researchers have acknowledged the value that PRO measures would bring to understanding the signs and symptoms of ITP and its treatments effects on HRQoL. After conducting a literature review aimed at assessing variability and terminology used in the diagnosis and treatment of ITP, Ruggeri, et al. suggested that further studies consider the effect of treatment options on quality of life evaluations as well as treatment effects on platelet counts . Bussel, et al. also recognized that improved treatment outcomes with new treatments can have an effect on patient HRQoL . Our literature searches yielded only four instances in which a questionnaire was used to assess quality of life [17, 32, 56, 57]. The PRO endpoints pursued in the studies were consistent with the concerns voiced by the patients in the focus groups, e.g. fear of bleeding and extreme fatigue.
A multi-dimensional measure of HRQoL in ITP can be particularly useful when comparing HRQoL outcomes of treatments that may have differential impacts on HRQoL or when some aspects of HRQoL may be improved at the expense of others. Our proposed conceptual model delineates concepts that may be considered in a PRO questionnaire to assess HRQoL changes in persons with ITP. Most of these concepts are included in the ITP-PAQ; any differences between the model and the ITP-PAQ reflect that the questionnaire was developed for use in clinical trials and could not contain all issues that may be relevant to patients. The model will also assist in the selection of items, scales, and/or questionnaires that may be appropriate to utilize when assessing HRQoL in ITP patients.
This research has some limitations. For one, the conceptual model focuses specifically on the relationship between symptoms, side-effects, and HRQoL and did not address other concepts such as treatment satisfaction, treatment decision-making, and medication compliance. Research looking into the roles of these other outcomes on patient HRQoL in ITP may be important.
The decision to use focus groups to gather patient contributions also created limitations. For instance, only English-speaking patients in the US were sampled; therefore the conceptual model may need further testing to be generalizable in other populations. The time commitment required for focus group participation and the use of an honorarium as an incentive for participating may have introduced a self-selection bias into the sample population. Also, the entry criteria for participation in the focus groups were very broad. No strata were used to guarantee a distribution in severity of disease, success of treatment, duration of disease, or age. All patients were recruited by physicians from tertiary care centers that tend to treat more severe or refractory patients. Therefore, this sample may be biased by over representing patients that are, in general, more ill or more aware of their illness. It is worth noting that although group interaction can stimulate participant ideas that might not have been available on an individual basis, it may also lead to overemphasis of the opinions of the groups' participants. For this reason, it is critical to the success of a focus group to have a highly experienced moderator guide the discussion.
Finally, because we could not match demographic or clinical history to individual focus group participants, it was not possible to distinguish whether the concepts affected some patients more than others. In particular, gender and age may play an important role in determining the HRQoL of adult patients living with chronic ITP. Further studies on the varying severity and progress of ITP may lead to slightly different conclusions regarding the pathway from platelet counts through signs and symptoms and HRQoL domains to overall HRQoL. However, the intent of this research was to develop a broad conceptual model.
Future research will build on the findings of this study. The sample size needed for qualitative research is determined by the concept of theoretical saturation. In order to achieve saturation, data must be collected until no new information is obtained. For this model our empirical data source, the existing transcripts of contributions from 23 patients, was not designed to reach saturation of themes or concepts. Researchers may wish to collect additional patient data either in focus groups or individual interviews to reach saturation on all areas of interest.
While the literature searches yielded several questionnaires that have been used to assess HRQoL in persons with ITP, only the ITP-PAQ, developed based on the feedback from these very focus groups of ITP patients, was designed and validated in patients with ITP. The ITP-PAQ covers many of the themes identified in our conceptual model for ITP, however other concepts important to patients, e.g. ability to perform daily activities, feelings of social isolation, inability to travel, financial stress, and reduction in libido, are not included in the ITP-PAQ. Patient-reported data captured with questionnaires such as the ITP-PAQ may be used to refine the conceptual model presented here. We expect further targeted research, in the style of Moore, et al. , to be used to test the relationships proposed in the ITP conceptual model, e.g., observing the changes in HRQoL of a subset of patients with a specific treatment or patients with a specific symptom.