Development of a disease-specific quality of life questionnaire for adult patients with hereditary angioedema due to C1 inhibitor deficiency (HAE-QoL): Spanish multi-centre research project

Hereditary angioedema due to C1 inhibitor deficiency (HAE-C1-INH) is a genetic disease that reduces the synthesis or function of C1 inhibitor (C1-INH) [1]. Bradykinin acts as the primary mediator [2]. It is considered a rare disease, and in Spain has a minimal prevalence rate of 1.09 cases/100,000 inhabitants [3]. HAE-C1-INH is characterized by recurrent edema attacks affecting different parts of the body (e.g.: face, extremities, gastrointestinal track, upper airway) [1, 4]. The clinical expression is quite variable; patients may be asymptomatic or suffer from life-threatening angioedema episodes with a wide range of disabling symptoms, severe pain or disfigurement. Attacks are unpredictable. Abdominal attacks may mimic acute abdominal emergencies, prompting unnecessary surgery. Mortality is mainly due to upper airway obstruction caused by laryngeal edema and has been reported to be as high as 30-50% in patients with undiagnosed HAE-C1-INH [4] and 30% in patients diagnosed with HAE-C1-INH whose cases were improperly treated [5].

Conventional treatment for allergic angioedema with adrenaline, antihistamines and glucocorticoids has not been shown efficacious [6].

Therapeutic management of HAE-C1-INH focuses on either preventing or treating attacks. Although new drugs have been developed in recent years, their availability differs from country to country and may even vary among hospitals within a country. Attenuated androgens, antifibrynolytics and plasma derived human C1 inhibitor concentrate (pdhC1INH) are the most commonly used drugs for long term prophylactic treatment. Election treatment for acute episodes are replacement of the lacking protein (pdhC1INH, nanofiltered pdhC1INH or recombinant C1INH concentrate), icatibant acetate or ecallantide and fresh frozen plasma when the aforementioned treatments are not available [7]. Different consensus guidelines for management of HAE-C1-INH have been published [7‐13]. Several drug adverse effects and contraindications have been reported (see Table 1).

The unavailability of some HAE-C1-INH specific therapies, as well as the potential side effects and contraindications of treatments could significantly hinder the management of HAE-C1-INH patients. Moreover, the lack of awareness of HAE-C1-INH among health care professionals may cause patients to fear for their safety and could have a negative effect on health-related Quality of Life (HRQoL).

HRQoL is considered to be a subjective assessment of the impact of disease and treatment across physical, psychological, social and somatic domains of functioning and well-being [23]. Several aspects of HAE-C1-INH (hereditary transmission, improper diagnosis, unpredictability of attacks, disabling symptoms, risk of fatal attacks, unnecessary surgeries, inadequate treatments, frequent need for emergency intervention, lack of information about the disease among health care professionals, adverse effects of treatment or unavailability of specific treatment, etc.) could diminish HRQoL, and the impact of HAE-C1-INH on quality of life should be evaluated. However, there is limited data on this topic in the published literature [24‐28]. To the best of our knowledge, a specific instrument for assessing the impact of HAE-C1-INH on HRQoL has not yet been made available.

This HAE-QoL project aims to develop a sound psychometric tool with which to assess HRQoL in HAE-C1-INH. It has been designed with two phases: the first was for developing the draft version of the questionnaire and the remaining phase will be for assessing its psychometric characteristics. Once the instrument has been validated, we will perform a HRQoL study using the final version of the questionnaire.

The results derived from the first phase of the study, in which domains were chosen and items were identified for the first draft of the HAE-QoL questionnaire, are described in this manuscript. We have also included the results of the expert and patient rating phase in which we received feedback for assessing both content and face validity. Lastly, this paper contains the protocol for the future pilot study and psychometric analysis.

Ethical approval was granted by the Research Ethics Committee of Hospital Universitario La Paz (Madrid). HAE-C1-INH patients 18 years or older and HAE-C1-INH experts from different regions of Spain participated in the study. Inclusion criteria for patients, were that they be 18 years old or older and have a confirmed laboratory diagnosis of HAE-C1-INH (type I or II). Exclusion criteria were cognitive disabilities and lack of fluency in Spanish language.

Participating patients signed an informed consent form. An in-house scoring system for classifying the severity of HAE-C1-INH (asymptomatic, mild, moderate and severe) in patients was established by taking into account the frequency of symptoms, the occurrence of life-threatening angioedema attacks and the need for maintenance therapy (see Table 2). For this step, a convenience sample was formed with patients with different degrees of disease severity. A main priority was to include subjects with the most severe clinical expression since they were the ones who theoretically would suffer a greater burden on their QoL.

The questionnaire was developed according to standard questionnaire development guidelines and methodologies, as well as the protocol for development of a specific HRQoL questionnaire for hemophilia, a disease that has similar characteristics to HAE-C1-INH [29‐36]. A patient-centred perspective based on a qualitative methodology was chosen in order to ensure that the content was appropiate and relevant for the target population and did not omit any issues of importance to the HAE-C1-INH patient.

Semi-structured individual interviews of HAE-C1-INH patients and experts were conducted using a written self-administered questionnaire. Subjects were asked 8 questions about their personal views on the disease (Table 3). Answers were transcribed and a qualitative content analysis of the responses was carried out according to published methodology [33, 34], in order to build a theoretical model. The procedure followed an inductive approach whereby transcripts were encoded in order to develop conceptual categories (domains), in which to classify issues that appeared in the data. Identified responses were organized in terms of frequency and issue, and common issues were grouped into domains. Lastly, items were generated using the reports on the verbatims from patients and experts. Every procedure was double checked by at least two members of the research team in order to ensure accuracy.

Our HAE-C1-INH patient group (29 females and 16 males) had a mean age of 39 years (18–74 years) and different degrees of HAE-C1-INH severity. These patients and the 8 HAE-C1-INH experts who participated in the study came from 8 regions of Spain (Madrid, Andalusia, Catalonia, Galicia, Asturias, the Valencian Community, Castille and Leon, and the Basque Country).

The most representative verbatim reports were used to formulate questions for developing an optimal item pool (examples are shown in Table 4). Content analysis led to the creation of 10 conceptual categories or domains: health conditions, physical functioning, social role, emotional role, physical role, symptoms, general health, aesthetics, mental health and treatment. A set of 64 items was developed and they were grouped into these domains. A theoretical model of the questionnaire was built based on these domains and served as the basis for the draft version of the HRQoL questionnaire (HAE-QoL v 1.0). A 5 or 6-point Likert scale (or summative scale) will be used.

The definitions of the preliminary domains of the questionnaire and the number of items per domain in the draft version are shown in Table 5.

In order to assess the content and face validity of the questionnaire, the draft version was initially evaluated by a group of 8 HAE-C1-INH experts and subsequently by a group of 16 HAE-C1-INH patients, all from different regions of Spain (expert and patient rating phase). Here, the sample was also a convenience one, in which subjects were selected in order to include a representative pool of patients that was heterogeneous with regard to sex, age, level of studies, geographical origin and severity of disease. A standardized evaluation form was used by experts to assess clarity of the items, relevance for HAE-C1-INH and assignment of dimensions. The condition established for deciding to keep, reformulate or remove items was an agreement rate of 80%. Qualitative comments suggesting that items be edited or added were also taken into account. Following the expert evaluation, 21 items were removed, 1 was added, 30 were reworded, and 6 items were reassigned to another dimension. Researchers agreed to remove the clinical dimension of the questionnaire and add a supplementary specific HAE-C1-INH clinical questionnaire specially developed for this study in order to obtain objective clinical data. Another major reason for removal was to avoid the use of similar items or to eliminate ambiguous or poorly worded items. No items were deemed irrelevant. Of the 30 items that were modified, 12 were for an agreement rate of 25%, 2 for a 37.5%, 1 for a 50% and in 14 items as result of qualitative comments. The resulting 44 item version was evaluated by the patients for clarity and relevance. Their evaluation resulted in the deletion of 1 item (due to qualitative comments), the modification of 8 items (1 item with an agreement rate of 31.25%, 3 with a 25% and 4 items due to qualitative comments) and the addition of 1 item. None of the items were removed due to a lack of relevance in either evaluation. Demographic characteristics and severity scores of the patients who participated in the semi-structured interview phase and rating phase are shown in Table 6.

The preliminary version of the questionnaire (HAE-QOL v 1.1.) contained 44 items compiled into 9 domains. Examples of the expert and patient rating phase are shown in Tables 7 and 8.

Disease-specific measures to assess HRQoL in HAE-C1-INH patients have not been available until now. This paper describes the initial steps toward developing a disease-specific QoL questionnaire for adults with HAE-C1-INH as part of a national multi-centre research project. Such an instrument can provide a more comprehensive and accurate evaluation of the HRQoL status of HAE-C1-INH patients and insights on how the illness affects their daily life. Moreover, it could be an instrumental tool for building health policies or implementing measures to improve the health and well-being of patients with HAE-C1-INH.

The methodological approach of this study is premised on the idea that a patient´s perception of illness and related impairments is a key issue to consider when assessing HRQoL [37]. Opinions of HAE-C1-INH experts were also used as the basis for item selection. Items were modified or removed based on the responses and feedback provided by patients and experts in order to ensure content validity. Our goal was to create an instrument that patients considered thorough, respondent-friendly and useful for expanding our knowledge of the impact of HAE-C1-INH on HRQoL.

Over the last few years, HRQoL has emerged as an important outcome measure of the degree of patient suffering, as well as a tool for comparing the efficacy of different treatments and assessing the results of health policy planning. However, this outcome measure is of little value if the instruments employed lack validity or reliability.

Although generic HRQoL questionnaires are readily available, they often lack the specificity necessary to adequately measure how certain aspects of a disease are related to QoL. In such cases, a disease-specific questionnaire would be preferable.

Data linking HAE-C1-INH to a burden on quality of life has been available for some time [38], however this is the first time that data has been collected using semi-structured questions. Huang et al. [25] showed that 46% of HAE-C1-INH patients were dissatisfied with the management of the disease, 85% had a constant fear of airway closure and 65% had a fear of pain. Other studies revealed that disruptive attacks caused patients to miss opportunities for career and personal development and they may have failed to reach their full potential because of high anxiety levels and depression [26‐28].

In our study, the issues cited most often by both experts and patients include potentially life-threatening attacks; the adverse side effects of medication (in several cases associated with chronic treatment); the unavailability of acute specific treatment at several health care centres; hereditary transmission; the lack of a known trigger to be avoided in many cases; and the fact that it is a rare disease about which health care professionals know very little. It is especially worth noting that the relevant aspects of HAE-C1-INH were not the same for experts and patients. Aesthetics was an aspect mentioned more often by patients than by experts. On the other hand, experts were more likely to mention the adverse side effects of treatment. This finding supports the general opinion that the clinician´s view of disease severity does not necessarily match with the patient´s perception of disability. Thus, measuring HRQoL impairment could become an important part of HAE-C1-INH management. Moreover, the recent HAE-C1-INH International Working Group consensus has recommended yearly assessments be carried out using a HRQoL questionnaire [12].

There is limited data available on HRQoL in patients with HAE-C1-INH. Lumry et al. [28] administered the SF-12 to patients and found that HAE-C1-INH created a considerable humanistic burden across physical and mental health domains. HAE-C1-INH patients also had higher mean scores than population norms in the Hamilton Depression Inventory-Short form (HDI-SF). Over 42% scored >8.5, which is indicative of depressive symptomatology. A determinant factor for depression was the severity of the disease. Patients who were under long term prophylactic treatment with attenuated androgens were shown to report even lower levels of HRQoL when compared with the overall HAE-C1-INH population.

Dermatology Life Quality Index (DLQI), SF-36, and an adaptation of the Pain Disability Index have been used in HAE-C1-INH patients [26, 27], but none of these questionnaires, including the SF-12, have been specifically developed or validated for HAE-C1-INH.

The HAE-QoL questionnaire is a disease-specific HRQoL measure that could be used in future research on clinical and economic matters and the identification of the determinants of HRQoL in HAE-C1-INH. This information could thereby aid health care providers and professionals in their quest to design and implement measures for improving the HRQoL of HAE-C1-INH patients.

The next step in the HAE-QoL project will be to set up a field study using the draft questionnaire (HAE-QoL v. 1.1). We will assess data quality, scaling assumption, reliability and evidences for validity.

Lastly, the final phase of this protocol will be to perform a HRQoL study with the final version of the validated questionnaire. Assessments of HRQoL can be repeated at different times in order to discover how changes in the management of the disease and changes in drug availability influence the HRQoL of HAE-C1-INH patients. We believe this is the first multi-centre research project designed to develop a specific instrument to assess HRQoL in adults with HAE-C1-INH.

NP, TC and ER have made substantial contribution to initial idea and design, acquisition, analysis and interpretation of data, and have been involved in drafting and revising the manuscript. In addition, they have reviewed the manuscript for important intellectual content and have given final approval of the version to be published. CG, CL, RC, JC, AC, VC, SC, TG, MG, DH, CM, MR and MAT have made substantial contributions to the acquisition, interpretation, and analysis of data. They have provided critical reviews of the manuscript and have given final approval of the version to be published. All authors read and approved the final manuscript.

Dr. Nieves Prior, Dr. Carmen Gómez-Traseira¹, Dr. Concepción López-Serrano¹, Dr. Rosario Cabañas¹, Dr. Stefan Cimbollek⁵, Dr. Teresa González-Quevedo⁵, Dr. Mar Guilarte⁴and Dr. Carmen Marcos are members of the Spanish Study Group on Bradykinin-Induced Angioedema (SGBA) (Grupo Español de Estudio del Angioedema mediado por Bradicinina: GEAB).

Dr. Teresa Caballero is an investigator from the IdiPAZ program for promoting research activities (2009), member of the Biomedical Research Network on Rare Diseases U754 (CIBERER) and coordinator of of Spanish Study Group on Bradykinin-Induced Angioedema (SGBA) (Grupo Español de Estudio del Angioedema mediado por Bradicinina: GEAB).

This work has been partially supported by a grant from SEAIC (Spanish Society of Allergy and Clinical Immunology, Type B Grant, 2003) and FIS (Fondo de Investigaciones Sanitarias, Grant n° 060843). Dr. Teresa Caballero is an investigator from the IdiPAZ program for promoting research activities (2009).