Are treatment benefits in neuropathic pain reflected in the self assessment of treatment questionnaire?

Neuropathic pain (NP) is one of the most prevalent pain etiologies [1]. It is a disorder of the peripheral nerves—the motor, sensory and autonomic nerves that connect the spinal cord to muscles, skin and internal organs [2, 3], and is a complex type of pain which requires a demonstrable lesion as defined by the updated pain terminology and definitions provided by the International Association for the Study of Pain (IASP); ( http://​www.​iasp-pain.​org/​AM/​Template.​cfm?​Section=​Pain_​Definitions) [4]. Recently, Treede and colleagues [5] redefined NP as “pain arising as a direct consequence of a lesion or disease affecting the somatosensory system.” Damage to the peripheral nervous system can occur as a result of several disorders including diabetes, infections such as herpes zoster (shingles) or HIV, trauma, and autoimmune disorders [1, 6]. While NP can occur in any location, it often manifests in the extremities such as the hands, feet, or legs. The overall prevalence of NP varies from between 0.9 % to 8% in a general population [7, 8]. Among HIV patients between 15% and 35% have NP, whilst the prevalence is 15% in patients with painful diabetic neuropathy or post herpetic neuralgia [9].

The NP symptom experience and the evolution of the condition are highly idiosyncratic with individual experiences varying widely between patients [1, 6]. The course of NP can be episodic and recurrent, slowly progressing over many years, or it can quickly become severe and debilitating. Both direct and indirect costs of NP to society are substantial [10‐14]. The consequences to the individual patient profoundly impair emotional and physical functioning with consequences such as movement limitation, an inability to perform major activities of daily living, needing help for personal care, frequent use of health care services, and work productivity losses [10, 11, 15, 16]. The disease burden is also substantial as shown in a comparison of the world wide burden to patients of suffering from NP assessed in a cross national study [17]. The intensity of the pain as well as the duration of the pain contributes to the distress associated with NP.

Important outcome domains were identified in a survey of people with pain conducted by the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) [18]. These domains formed the basis of the IMMPACT group’s recommendations on a core set of outcome domains to be included in evaluation of treatment benefits [19]. Whilst pain relief measures are used to determine whether the patient has actually benefited from an intervention, and provide valuable information on how effectively pain is being managed, patient satisfaction measures capture the personal evaluation of the intervention provided [20].

Despite newer treatment modalities NP remains difficult to treat. Treatment of NP in the past mainly consisted of regular analgesics, suggesting patients are receiving suboptimal treatment options [21]. However, recently, the European Medicines Agency [22] and the US Food and Drug Administration [23] approved the use of a high-concentration capsaicin topical patch 8% (QUTENZA™). Capsaicin, in high concentrations like the QUTENZA™ patch, is a transient receptor potential vanilloid 1 (TRPV1) agonist that is useful in relieving pain. Whilst the QUTENZA™ EU label covers the treatment of peripheral NP in non-diabetic adults, either alone or in combination with other medicinal products for pain, the US label is licensed for the management of neuropathic pain associated with post herpetic neuralgia [22, 23].

The five-item Self Assessment of Treatment (SAT) questionnaire was developed internally by Neurogesx for use in clinical trials to measure key areas of treatment benefit (i.e. pain, activities, quality of life and treatment satisfaction) recommended by IMMPACT [18]. Since psychometric documentation of the SAT was lacking, a post hoc analysis was conducted using trials data to validate the SAT [24]. The results indicated that the three pain and impact items were reliable and valid, whilst the satisfaction items performed less well [24]. The SAT has been used as a secondary end-point in two trials assessing treatment with Capsaicin in human immunodeficiency virus (HIV)-associated peripheral neuropathy (HIV-PN) with beneficial results for QUTENZA™ cited in the medical review conducted by the FDA [25], as well as by [26] for the treatment of post herpetic neuralgia [22].

The psychometric performance suggested that the SAT items could be further improved with additional patient input to ensure that items are clearly phrased, relevant and comprehensive, easy to understand, and that the response options and recall period are adequate to better assess pain related treatment benefits when used in clinical trials. Therefore, the purpose of this study was to better understand how well the SAT questionnaire reflects what patients perceived as relevant aspects of pain treatment and to document patient understanding of the individual SAT questions. Additionally, recommendations were provided on how to modify the instrument to be better suited to assess the benefits of pain treatment by changing the recall period, based on the information gained in the semi-structured patient interviews.

There are a couple of pain satisfaction surveys. The American Pain Society (APS) Satisfaction Survey is one of these and has shown that satisfaction was influenced largely by effectiveness of medication, independent of pain intensity [31]. This suggests that a satisfaction survey related to the effectiveness of a pain medication is an important measurement tool for use in clinical trials of novel pain treatments which is also consistent with the recommendations of the IMMACT group [18].

Even though the APS Satisfaction Survey focused on pain management in general practice, it was not directly related to satisfaction with medication. The same applies to another pain survey, the Pain Treatment Satisfaction Scale (PTSS) [32] which measured satisfaction in patients receiving treatment for acute or chronic pain; it was not designed for use in clinical trials, as it evaluates satisfaction with medical care received as well as pain medication. Although a variety of instruments capturing patient satisfaction exists, none of these measures were identified as appropriate for assessing patient satisfaction with medication used to evaluate the effect of treatment of chronic pain in the context of a clinical trial, which led to the development of the SAT.

In all, forty-four patients across three NP populations were interviewed which can be regarded as a sufficiently large sample size. The consistency of the findings support content validity and concept saturation as defined by Leidy and Vernon [33]and Magasi et al. [34]. Specifically, they define saturation as the point at which no substantially new information/concepts continue to emerge beyond what has been previously mentioned when interviewing the last few patients. Additional support can be found in the FDA guidance document [35], “The number of patients is not as critical as interview quality and patient diversity included in the sample in relation to intended clinical trial population characteristics.” In general, the 3 NP populations were similar in the way they described pain and the impact of pain symptoms. Clinicians reported a high variability in terms of pain reporting across patients but not across different NP types. Although a small number of clinicians were interviewed for this study, they provided similar input on the variability of pain experiences. The pain descriptors used by patients also varied, but were consistent irrespective of NP type and very also similar to the descriptors included in the McGill short form pain questionnaire where pain is referred to as debilitating, burning, sharp, jabbing, deep, and aching, and can persist for months or years [36, 37]. The similarity in pain description across all three NP types indicates that the content of the SAT is applicable for use across the different NP populations.

Similarities supporting the use of the SAT across all three NP populations prove that the concept of satisfaction with treatment assessment is fundamental in the evaluation of pain medication with both the clinical experts and the patients rating the reduction in pain as the most important treatment attribute followed by the ability to perform usual activities (by the PDPN population) and duration of pain relief (by the PHN and HIV-AN populations).

As pain reduction is unanimously identified as the most important aspect of treatment, it is not surprising that changes in pain have been shown to be closely linked to improvement in quality of life [13]. Once again, in the interviews, no difference was seen between the groups in terms of the impact of NP and the disruption of ability to perform daily activities and its impact on quality of life. With similarities in terms of pain definitions and similarities in the way that it affects individuals across all NP types as well as the fact that pain reduction is a critical component of treatment, it appears that the SAT would be appropriate for use in all NP types.

Pain is one area of research where the subjective experience and reporting by the patient is well-established and the concept of self assessment is particularly relevant. The importance of patient report/self assessment can be seen in work conducted by Taylor-Stokes and colleagues [38] where there was significant association between patient- and physician- rated severity but physician and patient ratings were discordant with physicians underestimating severity in 46.7% of patients who reported severe PDPN. Therefore, a patient-reported tool is critical in accessing the correct data. With patients identifying a reduction in pain as the most important treatment attribute, question 1 of the SAT is designed to assess this upfront. The patient interviews provided significant insight into the impact that pain has on patients’ lives and as a result questions 2 and 3 have been modified to reflect the multiple areas on which pain has an impact. Daily activities, physical activities, social activities and relationships were identified by participants with varying types and degrees of NP as facets of life that were affected by pain. These aspects are also captured by Ko and colleagues [39] who report that pain from PDPN interfered with general activity, mood, walking, normal work, relationships, sleep, and enjoyment of life. The addition of sleep and the emotional impact of pain identified by the participants are mirrored in the domain suggestions of the by the IMMPACT group [7, 19, 40]. The suggested modifications of the SAT enables patients to provide more details, as uncovered in the patient interviews, about the way NP affects their lives and the resulting impact of treatment in areas of life that matter to NP patients. Additionally, the consistency in pain and impact reports supports the content validity of the revised SAT, a critical component in any self report instrument [34, 41, 42].

With the introduction of baseline and follow up questionnaires and the inclusion of onset and duration of action of NP treatment, two important treatment attributes mentioned by patients in the interviews can be more efficiently captured. The modifications to the SAT provide an instrument that is more appropriate for use in clinical trials providing a baseline assessment and follow up assessments.

The previous recall bias is also amended and is more in line with FDA recommendations than the previous version as a result of the shorter recall period. The questions on the revised SAT are phrased in a consistent fashion to facilitate the respondents’ answering. Both the questions and the response options follow a similar pattern as well as the inclusion of examples derived from the interviews.

The content validity of the revised SAT has been supported by extensive qualitative research among diverse NP populations and clinicians and NP treatment benefits are now reflected in a more accurate and informed manner in the measure. Further alterations may be necessary once data on the measure performance has been collected but currently the most fundamental concept that of content validity has been assured. Without validity, a measure with good psychometric properties but not based on something real is not valid and has no use.

Astellas Pharma Global Development provided funding for this study.