The primary theoretical driver was a concurrent quantitative study12 (submitted). The present study was underpinned by a post-positive philosophical position paying attention to: (i) the quality of the data, whereby questionnaires are one dimensional and do not give voice to the context and reasons behind answers; (ii) the use of an integrated approach whereby subjective experiences account for some of the pain outcome being studied; and (iii) the context of the targeted phenomenon, in this case, the development of CS pain.13 This study is reported according to the Standards for Reporting of Qualitative Research (SRQR).14

This is a qualitative, exploratory study, using a concurrent nested design within a larger mixed methods protocol.

Participants were recruited from the quantitative sample by purposive sampling.15 Recruitment for the original quantitative sample was performed by clinical physical therapists experienced in pain neurophysiology and management, from physical therapy-led outpatient and pain clinics in England, Ireland and New Zealand. The inclusion criteria for the quantitative sample were for adults (age 18–64) with chronic (>3 months), non-specific (no specific pathology identified) low back pain (NSCLBP), with a predominant central sensitisation (CS) pain presentation according to the current published clinical guidelines for the clinical recognition of predominant central sensitisation in low back pain populations.2,7,9 The quantitative study recruited n=165 participants with NSCLBP and predominant CS (n=126 female) with a mean age of 45 (SD+/−12), 32% of whom were not on pain-related medication. A pilot study of the concurrent quantitative component had identified a prevalence of trait sensory hyper- and hypo-sensitivity profiles16,17 and anxiety-related personality types18 in individuals with NSCLBP and CS pain.19 Using this information, potential participants were stratified into n=6 groups based on their sensory profiles and personality type combinations. These groups were based on three trait anxiety-related personality type groups (defensive high anxious, high anxious, and repressors20) and two sensory profile (sensory hyper- or hypo-sensitive16,17) combinations. n=12 individuals were purposively selected (n=2 from each group). Individuals were provided with information about the study prior to giving written consent. Those who consented were given the option of being interviewed individually in their own home, at a physical therapy clinic, or the University campus.

Questions for the interview schedule were derived from the literature around sensory sensitivity and trait anxiety and from information which emerged from the pilot study.19 The interview schedule is summarised in Table 1. Interviews were digitally recorded and transcribed verbatim. Transcripts were sent to each participant for verification. Participant anonymity was assured through allocation of a number (P1-9). Data were protected by password encryption and managed using NVivo 10 software.21

Ethical approval was granted by Manchester Metropolitan University, Manchester, United Kingdom (ref: 1205) and permission was granted by the Northern Y Ethics committee, New Zealand.

Data were analysed using a ‘theoretical’ thematic analysis approach. Analysis was driven by the analytic focus of the primary theoretical driver in the concurrent quantitative study. ‘Theoretical’ thematic analysis provides a detailed analysis of certain aspects of interest predetermined by the researcher.22 The aspects of interest were around pre-morbid trait characteristics related to sensory processing and personality.

Transcripts and field notes were used for analysis. Field notes assisted understanding contexts in which interview data were given. Transcription data were analysed using thematic analysis according to guidelines by Braun and Clarke.22 Data were coded from participant data extracts and collated into categories. Categories were grouped into themes and sub-themes. Two researchers [JC and GY] analysed the data independently and were in substantial agreement over the final themes following the discussion of any semantic differences.

Participant demographics will be presented followed by data under the theme and sub-theme headings.

n=12 participants were initially identified from NZ and the UK. n=4 participants were lost to the study: n=2 participants could not be contacted (n=1 in UK, n=1 in NZ), n=1 UK participant declined due to unavailability at the time of the interviews, and n=1 because she felt she had recovered from her NSCLBP. n=1 further participant was recruited from NZ. In total, nine in-depth, face-to-face, semi-structured interviews were undertaken, n=8 from NZ and n=1 from UK. Seven of the participants elected to be interviewed in their own homes and two male participants elected to be interviewed at the physical therapy clinic of the primary investigator (JC) who conducted all the interviews. Each interview took place over a period of 60–100min. Any follow-up clarification of unclear information was performed by the primary investigator over the telephone.

Participants were aged 28–64 years, n=6 female and n=3 male. Based on scores from the Adolescent, Adult Sensory Profile,16n=3 participants scored in the extreme ranges for the Low Registration sensory profile. All scored in the extreme ranges for the Sensory Sensitive Sensory and Sensation Avoidance profiles. Based on measures of trait anxiety and defensiveness,18 personality types included n=4 Defensive High Anxious, n=3 Repressor and n=2 High Anxious individuals. n=5 participants had widespread pain (WP) as well as non-specific chronic low back pain (NSCLBP), the others had regional NSCLBP (Table 2). Data saturation was reached from the 9 interviews.

Four themes emerged: Developmental Learning Experiences, Personal Characteristics, Sensitivity and Trauma (Table 3). Findings in relation to each theme were divided into sub-themes and sub-headings. Due to the complexity and number of elements included, sub-themes overlap. The data extracts related to each theme are listed in Table 4.

Theme 1, Developmental Learning Experiences, related to childhood and adolescent experiences of school and the environment in which they grew up. Participants described diagnosed difficulties in learning such as dyslexia, or related disorders such as dyspraxia and ADHD. Other participants described differences in learning preferences but who had never received any formal diagnoses. Difficulties holding attention in class and a tendency towards boredom was reported by nearly half of the participants. A few participants claimed to be poorly coordinated in sports and play activities as a child; this was found in participants who had learning difficulties and a Low Registration sensory profile, characterised by reduced sensory feedback and slower movement.17

Participants with learning difficulties also displayed learning strengths in other aspects of their development. Contrary to learning difficulties, two of the participants demonstrated all round learning strengths and intelligence, such as reading, writing and maths. Conversely to having all-round sports strengths, two participants (both with extreme scores in the Low Registration sensory profile) described themselves as having strength in only running sport but being uncoordinated in other motor skills.

Theme 2, Personal Characteristics, determines how a person copes with challenges and unexpected events as well as aspects of resilience and confidence. Personal characteristics also determine behavioural responses and their tendency to interpret situations in a positive light or, alternatively, as situations which are threatening. The theme of personal characteristics encompasses two sub-themes, Confidence and Sense of Control.

Confidence covers aspects of attitudes and behaviours related to self-confidence, such as self-efficacy, self-esteem and self-identity. Having a strong sense of self-identity appeared to have a positive and a negative impact on individuals. Self-identities appeared to be enduring and learned from childhood. A negative impact of holding to these strong identities meant that a few of the participants developed a tendency to push themselves past their limits of wellbeing. A mismatch between expectation and the real lived experiences often led to low self-esteem and a sense of being weaker than peers.

A Sense of Control encompasses the categories of diligence and responses to adversity, both of which relate to behaviours of staying in control; and being controlled by others. Responses to adversity to regain a sense of control varied from active through to passive responses across the group and this appeared to relate closely to their individual trait sensory profiles and personality types.

Theme 3, Sensory Sensitivity, relates to the reception and perception of sensory stimuli, with higher sensitivity implying a lower neurological threshold for stimulation. Sensory sensitivity was apparent in the lives of all participants and presented in various forms of emotional and physical sensitivities. A large component of emotional sensitivity was anxiety, in which individuals were prone to responding to potential or actual threats with anxiety. Other emotional sensitivities emerged which were sensitivity to the opinion of others and sensitivity to injustice or unmet expectations. Physical sensitivity included sensitivity to touch, sound, light, smell and body feedback, as well as sensitivity to certain foods and food textures. Participants varied in their responses to sensory discomfort depending on their individual sensory profile.

All participants reported various physical sensitivities. Most participants reported high sensitivity, not handling as much as they considered they should in terms of certain physical stimuli, possibly suggestive of sensory hyper-sensitivity. A small number also reported they were unable to feel or sense certain sensory stimuli as much as others or as much as they thought they should, possibly suggesting sensory hypo-sensitivity. There was clearly overlap between sensory discomfort and emotional response, such as feelings of anxiety, linking together physical and emotional sensitivity.

Theme 4, Trauma, relates to both emotional and physical trauma. Emotional trauma was experienced in the forms of abuse, being controlled by others and in personal loss and overlaps with physical trauma. Physical trauma was experienced by many participants in the form of physical abuse and accident related trauma. Emotional trauma was prevalent in many of the participants, some as a result of abuse from others and some as a result of circumstances outside of their control, such as loss, or being controlled by others.