Neuropathic pain: A patient‐centred approach to measuring outcomes

Abstract Background Neuropathic pain (NP) is a complex, chronic pain state initiated by a primary lesion or dysfunction of the nervous system and presents as a variety of symptoms across multiple disease states. Objective To develop a patient‐centred conceptual model of symptoms and impacts in subjects with diabetic peripheral neuropathy (DPN) or post‐herpetic neuralgia (PHN) that can inform the measurement strategy in clinical trials. Method Thirty subjects with DPN or PHN participated in in‐person interviews which were performed until saturation was achieved. Transcripts were analysed in ATLAS.ti. Results Interviews were completed with DPN subjects (United States, n = 10; Japan, n = 10) and PHN subjects (United States, n = 5; Japan, n = 5). Numbness and tingling were frequently reported symptoms in the DPN population while itchiness and hypersensitivity were predominant in PHN. Both populations experienced burning and ache/soreness with similar frequency. DPN subjects experienced pain primarily in their lower extremity (eg feet, ankles), while PHN subjects experienced pain primarily in the chest and back. Impacts reported by DPN subjects included difficulty walking, sleep disturbance and climbing stairs. Impacts in PHN subjects included sleep disturbance, avoidance of physical contact, being angry/frustrated and being sad/depressed. Overall, concepts in Japan were not qualitatively different from the United States. Conceptual models of NP were generated based on the concepts elicited. Conclusions This research highlights core concepts to measure from the patient's perspective. Moreover, it enables the assessment of existing measures, the possible modification of these measures, or if a new NP measure with improved sensitivity and responsiveness is merited.


| INTRODUCTION
The International Association for the Study of Pain defines neuropathic pain as "pain arising as a direct consequence of a lesion or disease affecting the somatosensory system." 1 It is typically accompanied by tissue injury and presents itself with a variety of symptoms across multiple disease states. 2 Diabetic peripheral neuropathy (DPN) and post-herpetic neuralgia (PHN) are 2 common diseases characterized as having a neuropathic pain component. In the literature, the quality of neuropathic pain in DPN and PHN is commonly described as burning pain, numbness, electric shock-like sensation and tingling or skin-crawling sensation. 3 The pain can be spontaneous, constant or triggered in response to a stimuli (eg movement or touch). 4 Despite the high prevalence of DPN and PHN, the mechanism of neuropathic pain is complex and treatment remains a challenge.
Neuropathic pain is largely underreported and undertreated with best prevalence estimates between 7% and 10% 5 in the general population, while conservative estimates place prevalence at 1%-2%. 6 Twenty-six to 47% of people who suffer from diabetes mellitus experience DPN, whereas 25%-50% of people who suffer from herpes zoster viral disorder will develop PHN, depending on timing of antiviral therapy for herpes zoster. 7 Despite its prevalence, many management strategies for neuropathic pain are suboptimal and it remains an area of largely unmet therapeutic need. 8 Treatments include antidepressants (eg duloxetine), anticonvulsants (eg gabapentin, pregabalin), local anaesthetics (eg lidocaine patch, mexiletine, topical capsaicin) and opioid analgesics (eg tramadol) 9 with polypharmacy often used. 10 Patients have reported feeling dissatisfied with the efficacy of management therapies, 10,11 and the financial burden caused by direct medical costs, hospitalizations and outpatient visits, loss of the ability to work, loss of a caregivers' ability to work and possibly greater need for institutionalization or other living assistance. [12][13][14] Furthermore, patients may resort to self-management using non-approved alternative therapies or may learn to accept and live with the pain, adjusting their lifestyle as needed. 15 Neuropathic pain significantly impacts the daily lives of patients with DPN or PHN and is often associated with psychological comorbidities (ie depression and anxiety), sleep disturbance and detriments to self-confidence and self-esteem. 13,14,16,17 Impacts experienced by patients with DPN are reported as decreased physical functioning, mobility, home productivity, lack of energy and vitality, poor sleep quality, limitations in social relationships and enjoyment of life. 16 Similarly, patients with PHN associate their pain with fatigue, insomnia, depression, anxiety, interference with social roles and leisure activity. 7 Across aetiologies, neuropathic pain significantly interferes with physical activity, role and social functions, and negatively affects overall health-related quality of life (HRQoL). 2,14,16,18 There are substantial societal costs associated with neuropathic pain including direct and indirect costs and caregiver costs. 6,12 Patients with DPN incur substantial health-care costs due to outpatient visits and hospitalizations, and humanistic burden increases as pain severity increases, 16,18,19 a scenario also observed in PHN. 11 In Western Europe, direct medical costs have been shown to be twice as high as a control group of chronic pain patients. 13 A patient-centric approach to drug development, where patient needs and preferences are aligned with decision making, is becoming increasingly important to provide cost-effective treatment options. 20 In designing the measurement strategy of clinical studies, patient involvement in the form of qualitative interviews is encouraged by the Food and Drug Administration's (FDA) as evident in their 2009 Patient-Reported Outcome (PRO) Guidance. Previous primary research with patients includes a multicountry focus group study in which the content validity of the Neuropathic Pain Symptom Inventory (NPSI) was explored; 21 however, that research was conducted before the FDA's 2009 PRO Guidance and is limited to assessing the validity of a single, symptom-focused measure. This research expands on that investigation, with the goal of examining the impacts of neuropathic pain in addition to symptoms and seeks to develop an overarching conceptual model that can be compared to existing measures and explore where gaps lie in current measures.

| Ethics
Study documents were submitted to Copernicus Group Independent Review Board for ethics review and approval. In Japan, local ethics board approval was obtained. Research practices were guided by the Good Clinical Practice and regulatory requirements as applicable.

| Recruitment
Subjects were recruited from the United States and Japan, with a target population of 15 subjects with DPN and 15 subjects with PHN (5 subjects within each group were targeted in Japan). Clinicians were asked to attempt to recruit subjects with differing ages, genders and ethnicities (ethnicity only applicable to the United States).
Clinicians used existing medical records and the pre-specified inclusion/exclusion criteria to identify potentially eligible subjects.
Informed consent was obtained from the subject prior to participation, and the case report form was completed to document the subject's eligibility. Subjects also completed a basic demographic information form. All documents were reviewed to confirm subject eligibility and correct completion prior to conducting the interviews.

| Inclusion/exclusion criteria
Subjects recruited for the study were only enrolled if they satisfied the following inclusion criteria: 18 years of age or older; a confirmed diagnosis with DPN or PHN; an average daily pain score ≥4 on an 11-point 0-10 numerical rating scale at time of screening; in good general and psychological health and capable of completing a 60-minute face-to-face interview; able to speak, read, write and comprehend US English (US subjects) or Japanese (Japanese subjects  Probed concepts were not part of evaluating saturation because saturation is assessed by documenting concept emergence. After the first 25% of interviews were completed, interviewers discussed the findings at that stage and made adjustments to the interview process as needed. Subjects who participated in interviews were compensated for their time and travel costs.

| Transcription/translation and analysis
The audio recordings of the interviews were transcribed verbatim and analysed for content using the qualitative data analysis software, ATLAS.ti. Transcripts of interviews with US subjects were analysed by native English speakers, and transcripts with Japanese subjects were analysed by native Japanese speakers. Concepts identified during analysis of Japanese interviews were translated to English for comparison with US subjects.
The coding process was guided by established qualitative research methods, including grounded theory and constant comparative method. In grounded theory, inductive, yet systematic analytic strategies are applied to qualitative data to conceptually analyse individual experiences. 22,23 Coding schemes were developed to identify thematic trends in subject descriptions of symptoms and impacts related to their neuropathic pain experience. There were 3 coders, and harmonization meetings were held periodically (ie after first transcript, after 10 transcripts) to compare, reconcile and update the thematic coding scheme. The wording of the codes was harmonized to be conceptually equivalent and accurate (eg "feeling hot to the touch," "hot pain" and "burning pain" consolidated to "burning"). The concepts and domains identified, along with examples of subject quotes, were compared across aetiologies and subject populations and informed the development of a conceptual model in DPN and PHN.

| Saturation
Saturation is considered to be achieved at the point when additional interviews are unlikely to yield new information (ie new concepts of importance and relevance to subjects). 22 To evaluate conceptual saturation, concepts spontaneously emerging from the interviews were documented per subject, constantly comparing the total number of concepts that have already emerged from the previous subject(s) to the subsequent subject. Concepts were also compared in sets, in the order that data were collected. An example would be a group of 10 subjects, where the first set of interviews (eg n = 3) are compared with the next set (eg n = 4). Both of these sets of interviews (eg n = 7) are then compared with the final set of interviews (eg n = 3). The goal of this process is to compare the amount of novel information that is observed in the first interview set compared with the second interview set and so forth. A low number of subjects reporting a certain concept are neither an indication of whether saturation is achieved or not achieved. Instead, the timing of the concept occurrence during the interview process determines whether there is adequate evidence of saturation.
Saturation in the US population was evaluated and then compared to the Japanese population to determine if saturation was achieved in the total sample. In addition to confirming the adequacy of the sample size, this process highlights the emergence of new concepts to develop a comprehensive list of concepts, as well as the emergence of subconcepts that will help to saturate broader concepts.
Ten US subjects with PHN participated in the concept elicitation interviews. However, during the interview, 1 subject revealed that neuropathic pain only occurred during herpetic episodes, which resulted in removal of this subject during data analysis. US subjects with PHN were above the age of 41, with 3 subjects between 51 and 60 years old (n = 3, 33.3%) and 3 above the age of 71 (n = 3, 33.3%). The gender ratio was near equal, with 5 male subjects (n = 5, 55.6%) and approximately half of subjects were Caucasian (n = 5, 55.6%).

| Japanese cohort
All 5 Japanese DPN subjects had a diagnosis of Type 2 diabetes and were older than 41, with the average age around 63. Most subjects were male (n = 4, 80.0%).
Overall, 5 Japanese PHN subjects participated in the interviews.
All but one of these subjects was above the age of 71 (n = 4, 80.0%) with the remaining subject in the age range of 61-70 (n = 1, 20.0%) and 2 subjects were male (n = 2, 40.0%). Hispanic; Mexican-American; b With child; c With partner, children and parents; d Some patients marked more than one response, thus percentages in each category may total more than 100.0% for both Japanese and US populations.

| US cohort
Numbness was reported by every US DPN subject (100.0%). Some subjects had difficulties characterizing their numbness beyond "numbness," while other subjects described the sensation as "a loss of sense of feeling" or "doesn't have any sensation." Subjects tended to describe numbness in regard to how frequently it occurred and most reported experiencing their numbness "constantly." Tingling (90.0%) was also frequently reported and described using expressions such as "walking on rocks" and feeling like the foot "fell asleep." All 9 subjects described their tingling by severity indicating that inactivity (typically in the evenings) led to feeling the tingling sensation more acutely.
Subjects also frequently reported burning (60.0%), sharp pain (50.0%) and ache/soreness (50.0%). While the frequency of the burning sensation varied among subjects, most described the sensation as "intense" and "very painful." Descriptions ranged from "burning sensation," to "hot feet," to "on fire," and wanting to "put feet in ice water." Most subjects described burning and tingling as concurrent or similar sensations. Descriptors of sharp pain varied among subjects from "crushing pain" to "splitting" and "shooting" to being "stuck with a pin" or "sharp needle." However, all subjects described the sensation as sudden or spontaneous, frequent and short lasting. Ache/soreness was related to their level of physical activity or exertion, and relatively frequent and long in duration. However, common descriptors included a less severe sensation, such as "solid ache" or "dull ache." Among US PHN subjects, burning (66.7%), itchiness (66.7%), ache/soreness (55.6%) and hypersensitivity (55.6%) were the most frequently reported sensations. Most subjects described burning as a "feeling of heat" on the skin, as if someone had "thrown hot oil on you" and as frequent in occurrence but short in duration. Burning was also generally described as "intense" or "excruciating." However, 2 subjects described their burning as a deeper, "more muscular" sensation, which was mild in severity but constantly present. In regard to itchiness, 3 subjects focused on the depth of the sensation describing it as "skindeep" or "internal." In general, itchiness was not described as very severe. Frequency descriptions varied significantly between subjects and ranged from once a week to constantly. Ache/soreness was described by PHN subjects as if one had "hit your arm" or your body had been "beat on one side" with severity ranging from causing one to "scream" or "cry" to "a little sore." In terms of frequency, subjects experienced aching/soreness as varying on a daily basis or as a constant sensation.
Subjects described hypersensitivity in regard to their skin being "very sensitive to touch" and frequently reported having to avoid physical contact. Generally, subjects discussed hypersensitivity in regard to their outbreak period and not as a constant sensation.
Among the Japanese PHN subjects, no more than 2 subjects reported experiencing the same symptom. However, hypersensitivity, shooting, throbbing and tingling were somewhat frequently reported (40.0%).   Table 5 summarizes the most frequent, spontaneously reported impacts across both aetiologies and study populations associated with neuropathic pain as described by interviewed subjects. pain; although for some subjects, it was because of their fear of falling, which was caused by numbness in their feet.

| US cohort
Among impacts on daily living activities, subjects frequently reported impact on ability to wear certain clothing (50%), specifically footwear, and difficulty completing household activities (50%), which subjects often attributed to physical mobility limitations (eg difficulty standing or kneeling).
Subjects describing feelings of anger/irritability noted that these feelings manifested as short temperedness and impatience around friends or family because of their pain. Most subjects described their fear or worry in terms of falling.
Within the social domain, the only frequently reported impact was relying on others (50%), which was closely linked with subjects' diffi- The only frequently reported impact in the physical domain was avoiding physical contact to prevent triggering/exacerbating pain (77.8%) (ie touch from others, clothing or themselves). Lastly, subjects frequently reported that their pain experience interfered with their sex life (55.6%) attributing this impact more often to a loss of desire when in pain as opposed to a physical limitation. No frequently reported impacts were part of the cognitive functioning or work/school domains.
Within the activities of daily living domain, the only frequently reported impact was hygiene/personal care (60.0%). There were no frequently reported impacts in the social domain or in the work/school domain.
Within the sleep domain, difficulty falling asleep (60.0%) was the only frequently reported impact. Lastly, an impact on subjects' ability to do leisure activities (60.0%) was also frequently reported. There were no impacts on cognitive or work/school domains that were frequently/consistently reported.
Frequently reported impacts in the emotional domain were feeling depressed (60.0%) and a loss of motivation/interest (60.0%).
Within the activities of daily living domain, frequently reported impacts included avoiding physical activities (80.0%), household activities (80.0%), clothing limitations (60.0%), participating in exercise/ sports (60.0%) and going outside (60.0%). Of the impacts on physical mobility, difficulty getting to an upright position from a lying position (60.0%) was frequently reported by subjects.
Frequently reported social impacts were subjects' pain interfering with social activities in general (60.0%) and having to rely on others (60.0%). In regard to impacts on sleep, sleep disturbance (80.0%) was reported by most subjects.
In regard to cognitive functioning, subjects frequently reported being distracted by their pain (60.0%). Lastly, an impact on subjects' ability to partake in leisure activities (80.0%) was also frequently reported.

| The conceptual model
The

| DISCUSSION
Few neuropathic pain studies have taken a patient-centred approach in estimating symptom-and neuropathic pain-specific impact burden. Although there are common symptoms across DPN and PHN, selected symptoms should be considered specific to each aetiology.
This research found that burning sensation and ache/soreness were common to both aetiologies, whereas numbness, tingling and sharp pain were more commonly reported in DPN, and itchiness and hypersensitivity were more commonly reported in PHN. Therefore, outcome measures should be specific to the underlying disease-causing neuropathy or should focus on common core neuropathic pain symptoms with aetiology-specific questions as a supplement.
Current thinking in the field of outcomes assessment suggests that symptoms are best collected on a daily basis in a diary. 25