Living with hypoparathyroidism: development of the Hypoparathyroidism Patient Experience Scale-Impact (HPES-Impact)

Hypoparathyroidism (HP) is a rare endocrine disorder characterized by absent or inappropriately low levels of circulating parathyroid hormone (PTH) [1, 2]. Low levels or the absence of PTH circulating in the bloodstream can lead to hypocalcemia (low blood calcium levels), hyperphosphatemia (elevated blood phosphate levels), hypercalciuria (elevated urinary calcium levels), and overly-mineralized bone [1, 3]. Significant physical and cognitive symptoms are associated with the condition, including fatigue, muscle cramping/spasms, paresthesia (sensation of tingling and/or numbness), cognitive dysfunction, and sleep disturbances [1, 4‐8].

Based on data from 2007–2009, the prevalence of HP in the USA has been estimated to be between 70,000—115,000 patients [1, 9‐11]. The most common sources of HP are surgery in the neck region [2, 11], autoimmune disease [2, 12], and idiopathic [2]. HP is typically treated with oral calcium and vitamin D supplements [1‐3, 13]. PTH [1–84] replacement therapy has also been approved by the US Food & Drug Administration (FDA) for adults who do not respond to conventional therapy [1, 14].

Research indicates that patients with HP on conventional and/or PTH replacement therapy may have a reduced quality of life (QOL) [1, 5, 15‐17]. In multiple studies, patients on conventional therapy have had lower SF-36 [5, 6, 18, 19] and WHO-5 Well-Being Index [6] scores compared with the normative reference range and trial controls [6, 18] One study found that patients with post-surgical HP had significantly worse global complaint and subscale scores for anxiety, phobic anxiety, somatization, and their physical equivalents (e.g., heart palpitations) compared with patients who were post-thyroid surgery with intact parathyroid functioning [7]. Another study similarly reported that patients with HP scored worse on the symptom score for anxiety and depression, and the total Hospital Anxiety and Depression score compared with the normative population [18].

Studies of patients on PTH replacement therapy are limited and their findings have varied. In an open-label, uncontrolled PTH replacement study in the USA, significant improvements in the majority of domains of the SF-36 were reported [19]. However, in a randomized, double-blind, placebo-controlled study of PTH replacement therapy in Denmark, no significant difference between placebo and treatment groups was observed in SF-36 and WHO-5 Well-Being Index scores [5]. Although no significant between-group differences were observed in another randomized, double-blind, placebo-controlled study of PTH replacement therapy in the USA, there were significant improvements in multiple domains of the SF-36 for those in the treatment group compared with the placebo group; however, some of these scores remained low in comparison with the normal population [20].

As measured by the SF-36, QOL has further been reported to be reduced for patients with post-surgical HP compared to patients with other sources of HP [18, 20]. Since most patients with post-surgical HP also have post-surgical hypothyroidism, limited data is available on the extent to which the reduced QOL is due to HP or the combination of HP and hypothyroidism [6].

HP may also negatively impact daily life and work productivity. In a web-based survey of 374 adults with HP in the USA, 45% reported significant interference with their life due to HP [8]. Approximately one-fifth reported that symptoms associated with HP directly influenced a change in work status. Respondents further reported an increase in being unemployed, on disability, or retired, and a decrease in working either full- or part-time. Similarly, in a study of patients with HP in Norway, 40% reported receiving either permanent or temporary social security benefits compared with only 14% of the country’s general adult population [18].

Although previous research has provided evidence that QOL may be reduced in patients with HP, the validated questionnaires used in these studies were not disease-specific, and did not assess a number of documented symptoms associated with this condition, such as cognitive deficits, fatigue, or decreased muscle strength [15, 21]. In addition, owing to the lack of disease-specific questionnaires in previous research, impacts associated with these symptoms may also have been inadequately assessed.

The purpose of this study was to develop disease-specific measures of the symptoms and health-related QOL (HRQOL) impacts of HP, and provide evidence for the content validity of the measure based on rigorous qualitative research methodologies for patient-reported outcomes (PRO) development [22‐27]. The development of the first of these measures, the Hypoparathyroidism Patient Experience Scale-Symptom (HPES-Symptom) has been previously reported [28]. This paper describes the development of the Hypoparathyroidism Patient Experience Scale-Impact (HPES-Impact). This paper also describes the development of a preliminary theoretical model to identify relationships among key concepts, impacts, and modifiers related to the new PRO impact measure. Condition-specific instruments reportedly have greater face validity, are more responsive to change over time, and can be more useful to both clinicians and researchers to assess the impact of treatment on patients with HP [22‐25].

The HPES-Impact was developed in accordance with the FDA guidance [29] on best research practices for PRO measure development [22‐25]. The methods used to carry out this research, including eligibility criteria, have been described in detail in the HPES-Symptom article [28]. To briefly summarize, qualitative, individual concept elicitation (CE) interviews conducted by telephone with clinical experts and adults with HP were used to identify key concepts, the impacts of HP, that were both relevant and important to people with this condition to inform the development of the PRO measure and develop the preliminary theoretical model [24]. The clinical experts were well-established endocrinologists in the USA and Denmark with publications on HP who were currently treating patients with HP and had been recruited through a key opinion leader. Patient participants were adults in the USA with a diagnosis of HP who had been stable on standard-of-care (oral calcium and vitamin D supplements) and/or short-acting PTH replacement therapy for at least 3 months.

They were recruited via a national hypoparathyroidism organization and required to verify their eligibility by answering telephone screening questions and providing documentation of their diagnosis.

CE interviews with clinical experts and patient participants were completed by trained qualitative research interviewers familiar with HP who used a semi-structured interview guide, based on a literature review and discussions with clinical experts, that explored the signs and symptoms of HP and their impacts on QOL. Interviews lasted approximately one hour. Data were qualitatively analyzed through an adapted grounded theory approach, entailing developing and refining a theory based on concepts derived during the research process [26]. To carry out this analysis, participant and clinical expert transcripts were analyzed for content using Dedoose (www.​dedoose.​com), a qualitative analysis software program. First, the research team developed a preliminary code list based on the discussion guides’ sensitizing (initial, general) concepts [30] and subsequently revised the list based on concepts derived from the interviews during the coding process. For the portion of the study related to the development of the HPES-Impact measure, the overarching theme was the impacts associated with HP, and the key concepts within these were the specific impacts reported by participants and/or clinical experts. Within each of these study populations, each transcript was initially read, then coded, and reviewed multiple times to ensure accuracy and consistency. Transcripts were coded in the order in which interviews had been conducted, and emerging concepts were incorporated into the coding scheme as they were identified. Prior transcripts were then re-examined for the presence of newly emerged concepts. The coding scheme was refined throughout the analysis, including merging, modifying, or deleting codes as the researcher team gained further insights into particular concepts and the extent to which study participants considered them to be relevant and important. Lastly, following this iterative process and a series of internal deliberations to reach consensus among the research team members, the coded concepts were organized into larger categories of major themes (the impact domains).

Thematic saturation, defined as that point in time when no new major themes or subthemes emerged from the interviews and sufficient data had been collected to comprehensively identify, develop and understand the depth, range and relationships of the concepts and categories of interest [22, 27], was evaluated and confirmed through the development of a saturation grid populated with the key concepts discussed in each patient interview. A total of 101 impacts associated with the disease burden of HP emerged from the participant interviews. After the 14^th participant interview, 75% of impacts had been discussed, and by the 24^th interview, 95% of impacts were covered. Assessment of additional saturation criteria, including the frequency, severity and bothersomeness described by participants who reported experiencing each impact provided confirmatory evidence that saturation had been reached.

To address the question raised in the literature about the extent to which impaired QOL may be due to HP as a standalone condition versus HP combined with hypothyroidism, comparative frequencies for the overall endorsement rates of impacts were examined based on the proportions of participants with vs. without hypothyroidism. Differences of at least 20% in the proportions of participant endorsement rates between these subgroups were considered to be potentially meaningful. However, meaningful difference was not explored in the interviews; therefore, the 20% threshold cannot be considered a clinically meaningful difference.

A three-day, in-person item generation meeting was held by the project research team (n = 4). In accordance with FDA guidelines and good practices for PRO measure development, the team sought to establish criteria to identify impacts that were important and relevant to patients, including patients with varying levels of disease severity and demographic characteristics [22‐25, 29], as well as patients with or without hypothyroidism as a comorbidity. All issues that did not fulfill these criteria were categorized as minor, distal, or a modifier.

Based on decisions regarding major and minor impacts, a theoretical model was developed for the relationships among the key concepts of interest, and to identify potential modifiers [22]. Thus informed by the qualitative analysis, the major impacts identified, and the preliminary theoretical model, the team then generated the draft items and created an item definition table, which specified the intent of each item and acceptable synonyms, using the language of the participants as closely as possible.

Following item generation, cognitive debriefing (CD) interviews were conducted to reach consensus on the appropriate item format, structure, and wording, to ensure that items, response options, and instructions were clear and relevant, the recall period was appropriate, and the content was comprehensive [22, 25]. The CD interviews were conducted by telephone in an independent sample of patients with HP who met the same eligibility criteria as the CE interview sample using an item-by-item structured interview guide with probes for clarification. Participants were instructed to complete the measure within 24–48 h of their scheduled interview. This timeframe was chosen since it would be recent enough to avoid recall issues, while also ensuring that participants would have ample time to complete the measure without feeling pressured to finish it immediately prior to the interview. After the CD interviews were completed, a draft version of the measure ready for evaluation of measurement properties was produced, and the preliminary theoretical model was updated to incorporate the interview findings.

The study was approved by an independent Institutional Review Board (IRB), Copernicus Group IRB, located in Cary, North Carolina, USA. Informed consent was obtained from all patient participants.

The data from this research indicates that the impact burden among patients with HP is substantial, even when on conventional and PTH replacement treatment. This burden is evident in the wide range of impacts as well as their reported frequency, severity, and bothersomeness.

Notably, over 40% (n = 18, 43%) of participant-reported interference with work productivity that they attributed to the physical and/or cognitive symptoms experienced with HP. This represented the vast majority of participants who were currently employed (n = 21, 86%). An additional one-third of the sample (n = 14, 33%) reported no longer being able to work due to their HP symptoms. These findings underscore the debilitating nature of this condition and suggest that HP may have economic implications which should be quantitatively examined in future research.

Higher proportions of participants with hypothyroidism reported experiencing several daily life and psychological impacts compared with participants without this comorbidity. This may be due to an increased burden resulting from experiencing these chronic conditions simultaneously. It should be further noted that not all impacts were disproportionately experienced by those with both HP and hypothyroidism. There may be several explanations for this, including the influence of other comorbidities on patients without hypothyroidism, the severity of HP among some of those without hypothyroidism, the source of one’s HP, length of time with HP, and/or differences in treatment regimens among patients for both their HP and hypothyroidism. Regarding this latter point, since the focus of the study was on HP, data was not collected on the specific treatments participants were taking to manage their hypothyroidism. Further research with a larger study sample and a quantitative study design would be appropriate to further investigate the potential differences in QOL and their causal factors between these subgroups.

It is anticipated that a brief, disease-specific validated measure of the impacts of HP on patients may benefit clinical care by concretely assessing their specific effects on QOL, and by providing standardized language to describe the nature and severity of these impacts in individuals. This is especially important given the wide range and severity of both the symptoms and corresponding impacts experienced among patients with HP. The HPES-Impact measure provides a convenient way for clinicians to determine where their individual patients fall within this range, facilitating provider-patient discussions and treatment decisions, with a tool that captures the patient perspective. The measure may also be used in future research to evaluate disease burden or the impacts of treatment efficacy on HRQOL.

This study importantly provides evidence for content validity of the impacts captured in the HPES-Impact. Previous disease-specific measures that have been developed for this condition have focused primarily on symptoms and not captured the broad spectrum of both symptom and disease impacts [32]. We believe the HPES-Impact measure is the first disease-specific measure to capture the broad spectrum of disease impacts experienced by adult patients with HP.

As with all research, there are study limitations to consider when interpreting findings. These have been noted in a prior publication but should be reiterated here [28]. While efforts were made to recruit a diverse sample of participants, the majority were female and had post-surgical HP, which is similar to the sample compositions in other studies of patients with this condition [5, 8, 33]. The majority of participants also had hypothyroidism, which, as previously noted, is a common comorbidity with this condition [6]. Although none of the participants in the CE sample had HP with an autoimmune etiology, two participants with autoimmune HP were included in the CD sample, and both confirmed that the measure items were relevant and that the recall period was appropriate based on their experience with the condition. In addition, most participants were white/Caucasian, and it was a US-based population. However, the clinical experts from Denmark reported observing similar impacts among their patients, and studies describing patients with HP in Canada, India and multiple European countries provide evidence that the measure may be appropriate for assessments in other populations [5, 7, 18, 34, 35]. Cultural and linguistic validation of the measure would be required for use in non-US populations. This study was also limited to an adult population. Future research should examine the impacts of HP for children, as impacts may differ. Given the impacts reported by adults on work, for example, it is important to assess if school attendance and work are affected as well.

Given the sample sizes within each CE subgroup and the differences in the numbers of participants within the hypothyroidism vs. non-hypothyroidism subgroups (n = 31 vs. n = 11), study findings based on subgroups should be interpreted with caution. There were also variations within each subgroup that may impact the burden of illness, such as other comorbidities, duration of time with HP, and a high proportion of participants without hypothyroidism who were taking PTH replacement therapy (8/11, 73%). A quantitative survey study would be appropriate to further explore variations among subgroups based on treatment status, comorbidities, and other potential modifiers such as gender, age, disease etiology, and length of time with the condition.

Lastly, while several study participants reported that they had experienced a reduction in the frequency and severity of impacts due to their treatment, it was not possible to systematically assess these treatment benefits due to the potential for recall bias. It is anticipated that the HPES-Impact will serve as a useful tool for measuring responsiveness to treatment in patients who can be assessed before and after initially starting treatment or changing treatment regimens pending research to evaluate its measurement properties.

A better understanding of the impact burden experienced by patients with HP is important to inform clinician treatment decisions and address unmet treatment needs. This study provides evidence of content validity for the validation-ready HPES-Impact, which includes 26 items to be completed by adults ages 18 years and older with HP, across four domains of QOL: Physical Functioning, Daily Life, Psychological Well-Being, and Social Life and Relationships. Additional research is needed to assess the measurement properties of the HPES-Impact in patients with HP to determine whether it is a reliable measure. To the best of our knowledge, the HPES-Impact measure is the first disease-specific measure to capture the broad spectrum of disease impacts experienced by adult patients with HP.